{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.3391812865497075,
  "eval_steps": 500,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.691974544525147,
      "epoch": 0.005847953216374269,
      "grad_norm": 13.5,
      "learning_rate": 2e-06,
      "loss": 10.8755,
      "mean_token_accuracy": 0.0,
      "num_tokens": 13999.0,
      "step": 5
    },
    {
      "entropy": 10.691928291320801,
      "epoch": 0.011695906432748537,
      "grad_norm": 13.1875,
      "learning_rate": 4.5e-06,
      "loss": 10.7881,
      "mean_token_accuracy": 8.032128680497408e-05,
      "num_tokens": 25888.0,
      "step": 10
    },
    {
      "entropy": 10.690763854980469,
      "epoch": 0.017543859649122806,
      "grad_norm": 10.5625,
      "learning_rate": 7e-06,
      "loss": 10.5299,
      "mean_token_accuracy": 0.029781045974232256,
      "num_tokens": 37699.0,
      "step": 15
    },
    {
      "entropy": 10.669985389709472,
      "epoch": 0.023391812865497075,
      "grad_norm": 5.9375,
      "learning_rate": 9.5e-06,
      "loss": 10.2235,
      "mean_token_accuracy": 0.05048878975212574,
      "num_tokens": 49648.0,
      "step": 20
    },
    {
      "entropy": 10.5446457862854,
      "epoch": 0.029239766081871343,
      "grad_norm": 4.0,
      "learning_rate": 1.2e-05,
      "loss": 9.9371,
      "mean_token_accuracy": 0.049999012425541875,
      "num_tokens": 62255.0,
      "step": 25
    },
    {
      "entropy": 10.506792068481445,
      "epoch": 0.03508771929824561,
      "grad_norm": 3.4375,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 9.8635,
      "mean_token_accuracy": 0.048928847163915636,
      "num_tokens": 76003.0,
      "step": 30
    },
    {
      "entropy": 10.482833862304688,
      "epoch": 0.04093567251461988,
      "grad_norm": 2.484375,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 9.7065,
      "mean_token_accuracy": 0.06288735270500183,
      "num_tokens": 87585.0,
      "step": 35
    },
    {
      "entropy": 10.464082145690918,
      "epoch": 0.04678362573099415,
      "grad_norm": 2.984375,
      "learning_rate": 1.95e-05,
      "loss": 9.6752,
      "mean_token_accuracy": 0.061649443581700325,
      "num_tokens": 100404.0,
      "step": 40
    },
    {
      "entropy": 10.373864269256591,
      "epoch": 0.05263157894736842,
      "grad_norm": 2.25,
      "learning_rate": 2.2e-05,
      "loss": 9.6279,
      "mean_token_accuracy": 0.0631986454129219,
      "num_tokens": 113130.0,
      "step": 45
    },
    {
      "entropy": 10.265812587738036,
      "epoch": 0.05847953216374269,
      "grad_norm": 3.265625,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.4722,
      "mean_token_accuracy": 0.0745375208556652,
      "num_tokens": 125430.0,
      "step": 50
    },
    {
      "entropy": 10.30671148300171,
      "epoch": 0.06432748538011696,
      "grad_norm": 2.703125,
      "learning_rate": 2.7e-05,
      "loss": 9.5423,
      "mean_token_accuracy": 0.06626624949276447,
      "num_tokens": 137550.0,
      "step": 55
    },
    {
      "entropy": 10.23580093383789,
      "epoch": 0.07017543859649122,
      "grad_norm": 2.609375,
      "learning_rate": 2.95e-05,
      "loss": 9.4244,
      "mean_token_accuracy": 0.08311915919184684,
      "num_tokens": 149530.0,
      "step": 60
    },
    {
      "entropy": 10.123470211029053,
      "epoch": 0.07602339181286549,
      "grad_norm": 2.828125,
      "learning_rate": 3.2e-05,
      "loss": 9.4496,
      "mean_token_accuracy": 0.07575741671025753,
      "num_tokens": 161671.0,
      "step": 65
    },
    {
      "entropy": 10.18219289779663,
      "epoch": 0.08187134502923976,
      "grad_norm": 2.078125,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.3854,
      "mean_token_accuracy": 0.07742468379437924,
      "num_tokens": 174160.0,
      "step": 70
    },
    {
      "entropy": 10.220508003234864,
      "epoch": 0.08771929824561403,
      "grad_norm": 2.515625,
      "learning_rate": 3.7e-05,
      "loss": 9.3404,
      "mean_token_accuracy": 0.07739944010972977,
      "num_tokens": 186969.0,
      "step": 75
    },
    {
      "entropy": 10.152502727508544,
      "epoch": 0.0935672514619883,
      "grad_norm": 2.078125,
      "learning_rate": 3.95e-05,
      "loss": 9.2504,
      "mean_token_accuracy": 0.08422459289431572,
      "num_tokens": 199470.0,
      "step": 80
    },
    {
      "entropy": 10.011490631103516,
      "epoch": 0.09941520467836257,
      "grad_norm": 2.484375,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.1723,
      "mean_token_accuracy": 0.09044785127043724,
      "num_tokens": 211306.0,
      "step": 85
    },
    {
      "entropy": 10.162246799468994,
      "epoch": 0.10526315789473684,
      "grad_norm": 2.21875,
      "learning_rate": 4.45e-05,
      "loss": 9.1514,
      "mean_token_accuracy": 0.08413446620106697,
      "num_tokens": 223323.0,
      "step": 90
    },
    {
      "entropy": 9.993253993988038,
      "epoch": 0.1111111111111111,
      "grad_norm": 2.046875,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 9.0369,
      "mean_token_accuracy": 0.09222856536507607,
      "num_tokens": 235896.0,
      "step": 95
    },
    {
      "entropy": 10.006341552734375,
      "epoch": 0.11695906432748537,
      "grad_norm": 2.375,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 9.0269,
      "mean_token_accuracy": 0.08629260584712029,
      "num_tokens": 247879.0,
      "step": 100
    },
    {
      "entropy": 9.960891246795654,
      "epoch": 0.12280701754385964,
      "grad_norm": 2.265625,
      "learning_rate": 5.2e-05,
      "loss": 8.8842,
      "mean_token_accuracy": 0.08838528990745545,
      "num_tokens": 259714.0,
      "step": 105
    },
    {
      "entropy": 9.847960281372071,
      "epoch": 0.1286549707602339,
      "grad_norm": 1.84375,
      "learning_rate": 5.45e-05,
      "loss": 8.8575,
      "mean_token_accuracy": 0.09311703070998192,
      "num_tokens": 272429.0,
      "step": 110
    },
    {
      "entropy": 9.824352073669434,
      "epoch": 0.13450292397660818,
      "grad_norm": 2.0625,
      "learning_rate": 5.7e-05,
      "loss": 8.834,
      "mean_token_accuracy": 0.08550347462296486,
      "num_tokens": 285924.0,
      "step": 115
    },
    {
      "entropy": 9.759407234191894,
      "epoch": 0.14035087719298245,
      "grad_norm": 1.3828125,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.6848,
      "mean_token_accuracy": 0.09789064452052117,
      "num_tokens": 299453.0,
      "step": 120
    },
    {
      "entropy": 9.61653537750244,
      "epoch": 0.14619883040935672,
      "grad_norm": 2.046875,
      "learning_rate": 6.2e-05,
      "loss": 8.6997,
      "mean_token_accuracy": 0.09431982859969139,
      "num_tokens": 312295.0,
      "step": 125
    },
    {
      "entropy": 9.595601081848145,
      "epoch": 0.15204678362573099,
      "grad_norm": 1.8671875,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.5501,
      "mean_token_accuracy": 0.09463789388537407,
      "num_tokens": 324730.0,
      "step": 130
    },
    {
      "entropy": 9.421235179901123,
      "epoch": 0.15789473684210525,
      "grad_norm": 1.4296875,
      "learning_rate": 6.7e-05,
      "loss": 8.4792,
      "mean_token_accuracy": 0.09828749150037766,
      "num_tokens": 336481.0,
      "step": 135
    },
    {
      "entropy": 9.258519268035888,
      "epoch": 0.16374269005847952,
      "grad_norm": 1.4296875,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.4926,
      "mean_token_accuracy": 0.0937332771718502,
      "num_tokens": 348987.0,
      "step": 140
    },
    {
      "entropy": 9.255396842956543,
      "epoch": 0.1695906432748538,
      "grad_norm": 1.15625,
      "learning_rate": 7.2e-05,
      "loss": 8.4024,
      "mean_token_accuracy": 0.09496198520064354,
      "num_tokens": 360390.0,
      "step": 145
    },
    {
      "entropy": 9.080035209655762,
      "epoch": 0.17543859649122806,
      "grad_norm": 1.40625,
      "learning_rate": 7.45e-05,
      "loss": 8.3663,
      "mean_token_accuracy": 0.0986220009624958,
      "num_tokens": 372456.0,
      "step": 150
    },
    {
      "entropy": 9.131995296478271,
      "epoch": 0.18128654970760233,
      "grad_norm": 1.3359375,
      "learning_rate": 7.7e-05,
      "loss": 8.3326,
      "mean_token_accuracy": 0.09041213691234588,
      "num_tokens": 384185.0,
      "step": 155
    },
    {
      "entropy": 8.906043529510498,
      "epoch": 0.1871345029239766,
      "grad_norm": 1.203125,
      "learning_rate": 7.950000000000001e-05,
      "loss": 8.2872,
      "mean_token_accuracy": 0.0966079480946064,
      "num_tokens": 395399.0,
      "step": 160
    },
    {
      "entropy": 8.800198173522949,
      "epoch": 0.19298245614035087,
      "grad_norm": 1.9140625,
      "learning_rate": 8.2e-05,
      "loss": 8.2863,
      "mean_token_accuracy": 0.09909554198384285,
      "num_tokens": 408676.0,
      "step": 165
    },
    {
      "entropy": 8.711225223541259,
      "epoch": 0.19883040935672514,
      "grad_norm": 1.0859375,
      "learning_rate": 8.450000000000001e-05,
      "loss": 8.2698,
      "mean_token_accuracy": 0.09550312235951423,
      "num_tokens": 421861.0,
      "step": 170
    },
    {
      "entropy": 8.798204135894775,
      "epoch": 0.2046783625730994,
      "grad_norm": 1.484375,
      "learning_rate": 8.7e-05,
      "loss": 8.3183,
      "mean_token_accuracy": 0.09797023683786392,
      "num_tokens": 434852.0,
      "step": 175
    },
    {
      "entropy": 8.719051837921143,
      "epoch": 0.21052631578947367,
      "grad_norm": 1.4296875,
      "learning_rate": 8.95e-05,
      "loss": 8.2497,
      "mean_token_accuracy": 0.09236074090003968,
      "num_tokens": 447313.0,
      "step": 180
    },
    {
      "entropy": 8.575885391235351,
      "epoch": 0.21637426900584794,
      "grad_norm": 1.125,
      "learning_rate": 9.2e-05,
      "loss": 8.1896,
      "mean_token_accuracy": 0.10111410841345787,
      "num_tokens": 460271.0,
      "step": 185
    },
    {
      "entropy": 8.562645149230956,
      "epoch": 0.2222222222222222,
      "grad_norm": 1.3203125,
      "learning_rate": 9.45e-05,
      "loss": 8.1582,
      "mean_token_accuracy": 0.1076015666127205,
      "num_tokens": 473006.0,
      "step": 190
    },
    {
      "entropy": 8.600567626953126,
      "epoch": 0.22807017543859648,
      "grad_norm": 1.1796875,
      "learning_rate": 9.7e-05,
      "loss": 8.1814,
      "mean_token_accuracy": 0.09954929873347282,
      "num_tokens": 486100.0,
      "step": 195
    },
    {
      "entropy": 8.427424430847168,
      "epoch": 0.23391812865497075,
      "grad_norm": 1.5390625,
      "learning_rate": 9.95e-05,
      "loss": 8.0686,
      "mean_token_accuracy": 0.108884796500206,
      "num_tokens": 499460.0,
      "step": 200
    },
    {
      "entropy": 8.508369827270508,
      "epoch": 0.23976608187134502,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000102,
      "loss": 8.0849,
      "mean_token_accuracy": 0.10188826620578766,
      "num_tokens": 511795.0,
      "step": 205
    },
    {
      "entropy": 8.331426906585694,
      "epoch": 0.24561403508771928,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00010449999999999999,
      "loss": 8.1497,
      "mean_token_accuracy": 0.10387411639094353,
      "num_tokens": 524897.0,
      "step": 210
    },
    {
      "entropy": 8.393550872802734,
      "epoch": 0.25146198830409355,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000107,
      "loss": 8.0901,
      "mean_token_accuracy": 0.10539199337363243,
      "num_tokens": 538325.0,
      "step": 215
    },
    {
      "entropy": 8.538016510009765,
      "epoch": 0.2573099415204678,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001095,
      "loss": 8.1375,
      "mean_token_accuracy": 0.09665202274918556,
      "num_tokens": 550979.0,
      "step": 220
    },
    {
      "entropy": 8.389378452301026,
      "epoch": 0.2631578947368421,
      "grad_norm": 1.15625,
      "learning_rate": 0.000112,
      "loss": 8.1055,
      "mean_token_accuracy": 0.10522033721208572,
      "num_tokens": 563001.0,
      "step": 225
    },
    {
      "entropy": 8.339937973022462,
      "epoch": 0.26900584795321636,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001145,
      "loss": 8.0235,
      "mean_token_accuracy": 0.1068884126842022,
      "num_tokens": 573600.0,
      "step": 230
    },
    {
      "entropy": 8.336192607879639,
      "epoch": 0.27485380116959063,
      "grad_norm": 1.296875,
      "learning_rate": 0.00011700000000000001,
      "loss": 8.0318,
      "mean_token_accuracy": 0.10861181318759919,
      "num_tokens": 585287.0,
      "step": 235
    },
    {
      "entropy": 8.343888473510741,
      "epoch": 0.2807017543859649,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011949999999999999,
      "loss": 8.0583,
      "mean_token_accuracy": 0.10137449800968171,
      "num_tokens": 597386.0,
      "step": 240
    },
    {
      "entropy": 8.288441467285157,
      "epoch": 0.28654970760233917,
      "grad_norm": 1.015625,
      "learning_rate": 0.000122,
      "loss": 8.0481,
      "mean_token_accuracy": 0.10464474633336067,
      "num_tokens": 609256.0,
      "step": 245
    },
    {
      "entropy": 8.411810398101807,
      "epoch": 0.29239766081871343,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001245,
      "loss": 8.0474,
      "mean_token_accuracy": 0.10834256708621978,
      "num_tokens": 621506.0,
      "step": 250
    },
    {
      "entropy": 8.100932836532593,
      "epoch": 0.2982456140350877,
      "grad_norm": 1.328125,
      "learning_rate": 0.000127,
      "loss": 7.9646,
      "mean_token_accuracy": 0.1123759962618351,
      "num_tokens": 633512.0,
      "step": 255
    },
    {
      "entropy": 8.300220108032226,
      "epoch": 0.30409356725146197,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001295,
      "loss": 8.0895,
      "mean_token_accuracy": 0.10339951142668724,
      "num_tokens": 647098.0,
      "step": 260
    },
    {
      "entropy": 8.343167591094971,
      "epoch": 0.30994152046783624,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000132,
      "loss": 8.0972,
      "mean_token_accuracy": 0.09793601483106613,
      "num_tokens": 659559.0,
      "step": 265
    },
    {
      "entropy": 8.258354759216308,
      "epoch": 0.3157894736842105,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00013450000000000002,
      "loss": 8.0199,
      "mean_token_accuracy": 0.10353473275899887,
      "num_tokens": 672621.0,
      "step": 270
    },
    {
      "entropy": 8.298446559906006,
      "epoch": 0.3216374269005848,
      "grad_norm": 1.234375,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.9906,
      "mean_token_accuracy": 0.10663355737924576,
      "num_tokens": 685652.0,
      "step": 275
    },
    {
      "entropy": 8.19245777130127,
      "epoch": 0.32748538011695905,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001395,
      "loss": 8.04,
      "mean_token_accuracy": 0.11108417361974716,
      "num_tokens": 696910.0,
      "step": 280
    },
    {
      "entropy": 8.219241619110107,
      "epoch": 0.3333333333333333,
      "grad_norm": 1.78125,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.9193,
      "mean_token_accuracy": 0.11071920469403267,
      "num_tokens": 709054.0,
      "step": 285
    },
    {
      "entropy": 8.26666088104248,
      "epoch": 0.3391812865497076,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0001445,
      "loss": 7.9479,
      "mean_token_accuracy": 0.11016275808215141,
      "num_tokens": 720902.0,
      "step": 290
    },
    {
      "entropy": 8.227052402496337,
      "epoch": 0.34502923976608185,
      "grad_norm": 1.21875,
      "learning_rate": 0.000147,
      "loss": 8.0106,
      "mean_token_accuracy": 0.10988251566886902,
      "num_tokens": 733319.0,
      "step": 295
    },
    {
      "entropy": 8.17465419769287,
      "epoch": 0.3508771929824561,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001495,
      "loss": 7.8917,
      "mean_token_accuracy": 0.11573786139488221,
      "num_tokens": 744817.0,
      "step": 300
    },
    {
      "entropy": 8.232967758178711,
      "epoch": 0.3567251461988304,
      "grad_norm": 1.078125,
      "learning_rate": 0.000152,
      "loss": 7.9535,
      "mean_token_accuracy": 0.10650328993797302,
      "num_tokens": 756582.0,
      "step": 305
    },
    {
      "entropy": 8.144847440719605,
      "epoch": 0.36257309941520466,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.8627,
      "mean_token_accuracy": 0.12130056619644165,
      "num_tokens": 769273.0,
      "step": 310
    },
    {
      "entropy": 8.138186359405518,
      "epoch": 0.3684210526315789,
      "grad_norm": 1.296875,
      "learning_rate": 0.000157,
      "loss": 7.9351,
      "mean_token_accuracy": 0.11499258056282997,
      "num_tokens": 782001.0,
      "step": 315
    },
    {
      "entropy": 8.08138952255249,
      "epoch": 0.3742690058479532,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001595,
      "loss": 7.926,
      "mean_token_accuracy": 0.11798002570867538,
      "num_tokens": 794575.0,
      "step": 320
    },
    {
      "entropy": 8.198786306381226,
      "epoch": 0.38011695906432746,
      "grad_norm": 1.171875,
      "learning_rate": 0.000162,
      "loss": 7.9014,
      "mean_token_accuracy": 0.11587951257824898,
      "num_tokens": 805583.0,
      "step": 325
    },
    {
      "entropy": 8.112969303131104,
      "epoch": 0.38596491228070173,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.9467,
      "mean_token_accuracy": 0.1160537801682949,
      "num_tokens": 817617.0,
      "step": 330
    },
    {
      "entropy": 8.17340841293335,
      "epoch": 0.391812865497076,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.8426,
      "mean_token_accuracy": 0.11603786572813987,
      "num_tokens": 830043.0,
      "step": 335
    },
    {
      "entropy": 8.051109790802002,
      "epoch": 0.39766081871345027,
      "grad_norm": 1.203125,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.8356,
      "mean_token_accuracy": 0.12449322417378425,
      "num_tokens": 843200.0,
      "step": 340
    },
    {
      "entropy": 8.152585649490357,
      "epoch": 0.40350877192982454,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.8698,
      "mean_token_accuracy": 0.12035520151257514,
      "num_tokens": 856031.0,
      "step": 345
    },
    {
      "entropy": 8.055157375335693,
      "epoch": 0.4093567251461988,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.8494,
      "mean_token_accuracy": 0.1174808993935585,
      "num_tokens": 869119.0,
      "step": 350
    },
    {
      "entropy": 7.974756145477295,
      "epoch": 0.4152046783625731,
      "grad_norm": 1.0625,
      "learning_rate": 0.000177,
      "loss": 7.8566,
      "mean_token_accuracy": 0.12319247648119927,
      "num_tokens": 880802.0,
      "step": 355
    },
    {
      "entropy": 8.131376028060913,
      "epoch": 0.42105263157894735,
      "grad_norm": 1.125,
      "learning_rate": 0.0001795,
      "loss": 7.8644,
      "mean_token_accuracy": 0.11949887052178383,
      "num_tokens": 894237.0,
      "step": 360
    },
    {
      "entropy": 8.112016677856445,
      "epoch": 0.4269005847953216,
      "grad_norm": 1.109375,
      "learning_rate": 0.000182,
      "loss": 7.8648,
      "mean_token_accuracy": 0.12209260165691375,
      "num_tokens": 906695.0,
      "step": 365
    },
    {
      "entropy": 8.052251434326172,
      "epoch": 0.4327485380116959,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001845,
      "loss": 7.8327,
      "mean_token_accuracy": 0.1234503522515297,
      "num_tokens": 918955.0,
      "step": 370
    },
    {
      "entropy": 7.96677598953247,
      "epoch": 0.43859649122807015,
      "grad_norm": 1.015625,
      "learning_rate": 0.000187,
      "loss": 7.8616,
      "mean_token_accuracy": 0.12289973944425583,
      "num_tokens": 931812.0,
      "step": 375
    },
    {
      "entropy": 8.08064661026001,
      "epoch": 0.4444444444444444,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001895,
      "loss": 7.8693,
      "mean_token_accuracy": 0.11234893426299095,
      "num_tokens": 945261.0,
      "step": 380
    },
    {
      "entropy": 8.010916709899902,
      "epoch": 0.4502923976608187,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000192,
      "loss": 7.781,
      "mean_token_accuracy": 0.12615325823426246,
      "num_tokens": 958376.0,
      "step": 385
    },
    {
      "entropy": 7.934221315383911,
      "epoch": 0.45614035087719296,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001945,
      "loss": 7.7977,
      "mean_token_accuracy": 0.12733107730746268,
      "num_tokens": 971709.0,
      "step": 390
    },
    {
      "entropy": 8.095316505432129,
      "epoch": 0.4619883040935672,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.772,
      "mean_token_accuracy": 0.12458922639489174,
      "num_tokens": 983847.0,
      "step": 395
    },
    {
      "entropy": 8.061276531219482,
      "epoch": 0.4678362573099415,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.8479,
      "mean_token_accuracy": 0.11282772347331047,
      "num_tokens": 997034.0,
      "step": 400
    },
    {
      "entropy": 7.900276470184326,
      "epoch": 0.47368421052631576,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000202,
      "loss": 7.8811,
      "mean_token_accuracy": 0.11630931496620178,
      "num_tokens": 1010213.0,
      "step": 405
    },
    {
      "entropy": 8.164872455596925,
      "epoch": 0.47953216374269003,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020449999999999998,
      "loss": 7.7587,
      "mean_token_accuracy": 0.12265534400939941,
      "num_tokens": 1021618.0,
      "step": 410
    },
    {
      "entropy": 7.90818042755127,
      "epoch": 0.4853801169590643,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000207,
      "loss": 7.7779,
      "mean_token_accuracy": 0.119289730489254,
      "num_tokens": 1033809.0,
      "step": 415
    },
    {
      "entropy": 8.035586547851562,
      "epoch": 0.49122807017543857,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002095,
      "loss": 7.7602,
      "mean_token_accuracy": 0.122083730250597,
      "num_tokens": 1045291.0,
      "step": 420
    },
    {
      "entropy": 7.94893217086792,
      "epoch": 0.49707602339181284,
      "grad_norm": 1.28125,
      "learning_rate": 0.000212,
      "loss": 7.7345,
      "mean_token_accuracy": 0.12310368940234184,
      "num_tokens": 1057516.0,
      "step": 425
    },
    {
      "entropy": 7.834074544906616,
      "epoch": 0.5029239766081871,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002145,
      "loss": 7.7166,
      "mean_token_accuracy": 0.13022998943924904,
      "num_tokens": 1070675.0,
      "step": 430
    },
    {
      "entropy": 7.962926816940308,
      "epoch": 0.5087719298245614,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00021700000000000002,
      "loss": 7.7871,
      "mean_token_accuracy": 0.12778857946395875,
      "num_tokens": 1082965.0,
      "step": 435
    },
    {
      "entropy": 7.948134422302246,
      "epoch": 0.5146198830409356,
      "grad_norm": 1.0,
      "learning_rate": 0.0002195,
      "loss": 7.767,
      "mean_token_accuracy": 0.11752260848879814,
      "num_tokens": 1095922.0,
      "step": 440
    },
    {
      "entropy": 7.899146366119385,
      "epoch": 0.52046783625731,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000222,
      "loss": 7.7519,
      "mean_token_accuracy": 0.1246542438864708,
      "num_tokens": 1109002.0,
      "step": 445
    },
    {
      "entropy": 8.014074802398682,
      "epoch": 0.5263157894736842,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002245,
      "loss": 7.7662,
      "mean_token_accuracy": 0.11872777864336967,
      "num_tokens": 1121199.0,
      "step": 450
    },
    {
      "entropy": 7.865769338607788,
      "epoch": 0.5321637426900585,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00022700000000000002,
      "loss": 7.7036,
      "mean_token_accuracy": 0.12603166699409485,
      "num_tokens": 1134721.0,
      "step": 455
    },
    {
      "entropy": 7.8770805358886715,
      "epoch": 0.5380116959064327,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022950000000000002,
      "loss": 7.788,
      "mean_token_accuracy": 0.12093541622161866,
      "num_tokens": 1147954.0,
      "step": 460
    },
    {
      "entropy": 7.915914726257324,
      "epoch": 0.543859649122807,
      "grad_norm": 1.125,
      "learning_rate": 0.00023200000000000003,
      "loss": 7.7121,
      "mean_token_accuracy": 0.12637170255184174,
      "num_tokens": 1161208.0,
      "step": 465
    },
    {
      "entropy": 7.918968534469604,
      "epoch": 0.5497076023391813,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023449999999999998,
      "loss": 7.8051,
      "mean_token_accuracy": 0.11931493803858757,
      "num_tokens": 1173789.0,
      "step": 470
    },
    {
      "entropy": 7.964196968078613,
      "epoch": 0.5555555555555556,
      "grad_norm": 1.0,
      "learning_rate": 0.000237,
      "loss": 7.7054,
      "mean_token_accuracy": 0.12410269007086754,
      "num_tokens": 1185765.0,
      "step": 475
    },
    {
      "entropy": 7.763233613967896,
      "epoch": 0.5614035087719298,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002395,
      "loss": 7.7242,
      "mean_token_accuracy": 0.12448163107037544,
      "num_tokens": 1197954.0,
      "step": 480
    },
    {
      "entropy": 7.995425701141357,
      "epoch": 0.5672514619883041,
      "grad_norm": 1.046875,
      "learning_rate": 0.000242,
      "loss": 7.6458,
      "mean_token_accuracy": 0.12746693715453147,
      "num_tokens": 1211285.0,
      "step": 485
    },
    {
      "entropy": 7.84964427947998,
      "epoch": 0.5730994152046783,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002445,
      "loss": 7.7247,
      "mean_token_accuracy": 0.12084417417645454,
      "num_tokens": 1223197.0,
      "step": 490
    },
    {
      "entropy": 7.825186395645142,
      "epoch": 0.5789473684210527,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000247,
      "loss": 7.6581,
      "mean_token_accuracy": 0.12425737157464027,
      "num_tokens": 1234817.0,
      "step": 495
    },
    {
      "entropy": 7.8655126094818115,
      "epoch": 0.5847953216374269,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002495,
      "loss": 7.7119,
      "mean_token_accuracy": 0.13079003244638443,
      "num_tokens": 1249305.0,
      "step": 500
    },
    {
      "epoch": 0.5847953216374269,
      "eval_entropy": 7.929424790857816,
      "eval_loss": 7.730849742889404,
      "eval_mean_token_accuracy": 0.11598945001640239,
      "eval_num_tokens": 1249305.0,
      "eval_runtime": 2.3042,
      "eval_samples_per_second": 1174.802,
      "eval_steps_per_second": 147.121,
      "step": 500
    },
    {
      "entropy": 7.852921438217163,
      "epoch": 0.5906432748538012,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000252,
      "loss": 7.6822,
      "mean_token_accuracy": 0.12737054675817489,
      "num_tokens": 1262182.0,
      "step": 505
    },
    {
      "entropy": 7.7925210952758786,
      "epoch": 0.5964912280701754,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002545,
      "loss": 7.7134,
      "mean_token_accuracy": 0.13160952627658845,
      "num_tokens": 1274616.0,
      "step": 510
    },
    {
      "entropy": 7.7616315364837645,
      "epoch": 0.6023391812865497,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000257,
      "loss": 7.7244,
      "mean_token_accuracy": 0.11796518042683601,
      "num_tokens": 1287306.0,
      "step": 515
    },
    {
      "entropy": 7.773230743408203,
      "epoch": 0.6081871345029239,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002595,
      "loss": 7.5783,
      "mean_token_accuracy": 0.13205838426947594,
      "num_tokens": 1299472.0,
      "step": 520
    },
    {
      "entropy": 7.825972223281861,
      "epoch": 0.6140350877192983,
      "grad_norm": 1.7578125,
      "learning_rate": 0.000262,
      "loss": 7.5578,
      "mean_token_accuracy": 0.13092489391565323,
      "num_tokens": 1312636.0,
      "step": 525
    },
    {
      "entropy": 7.714103269577026,
      "epoch": 0.6198830409356725,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026450000000000003,
      "loss": 7.5719,
      "mean_token_accuracy": 0.1383853517472744,
      "num_tokens": 1324058.0,
      "step": 530
    },
    {
      "entropy": 7.7577794075012205,
      "epoch": 0.6257309941520468,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00026700000000000004,
      "loss": 7.6831,
      "mean_token_accuracy": 0.12188005968928337,
      "num_tokens": 1337605.0,
      "step": 535
    },
    {
      "entropy": 7.853874206542969,
      "epoch": 0.631578947368421,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026950000000000005,
      "loss": 7.6867,
      "mean_token_accuracy": 0.12546794787049292,
      "num_tokens": 1350658.0,
      "step": 540
    },
    {
      "entropy": 7.805501413345337,
      "epoch": 0.6374269005847953,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00027200000000000005,
      "loss": 7.6069,
      "mean_token_accuracy": 0.12714534625411034,
      "num_tokens": 1362091.0,
      "step": 545
    },
    {
      "entropy": 7.723537731170654,
      "epoch": 0.6432748538011696,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002745,
      "loss": 7.6142,
      "mean_token_accuracy": 0.1300032578408718,
      "num_tokens": 1374055.0,
      "step": 550
    },
    {
      "entropy": 7.724965763092041,
      "epoch": 0.6491228070175439,
      "grad_norm": 1.25,
      "learning_rate": 0.000277,
      "loss": 7.5059,
      "mean_token_accuracy": 0.12956581264734268,
      "num_tokens": 1385269.0,
      "step": 555
    },
    {
      "entropy": 7.69310941696167,
      "epoch": 0.6549707602339181,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002795,
      "loss": 7.6122,
      "mean_token_accuracy": 0.13100477755069734,
      "num_tokens": 1396559.0,
      "step": 560
    },
    {
      "entropy": 7.651605224609375,
      "epoch": 0.6608187134502924,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00028199999999999997,
      "loss": 7.6413,
      "mean_token_accuracy": 0.13066264986991882,
      "num_tokens": 1408838.0,
      "step": 565
    },
    {
      "entropy": 7.753659629821778,
      "epoch": 0.6666666666666666,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002845,
      "loss": 7.5345,
      "mean_token_accuracy": 0.1306803986430168,
      "num_tokens": 1421023.0,
      "step": 570
    },
    {
      "entropy": 7.645546102523804,
      "epoch": 0.672514619883041,
      "grad_norm": 1.140625,
      "learning_rate": 0.000287,
      "loss": 7.5958,
      "mean_token_accuracy": 0.13120171204209327,
      "num_tokens": 1433912.0,
      "step": 575
    },
    {
      "entropy": 7.768090343475341,
      "epoch": 0.6783625730994152,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002895,
      "loss": 7.6014,
      "mean_token_accuracy": 0.12672622352838517,
      "num_tokens": 1445262.0,
      "step": 580
    },
    {
      "entropy": 7.660789060592651,
      "epoch": 0.6842105263157895,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000292,
      "loss": 7.5561,
      "mean_token_accuracy": 0.1315220259130001,
      "num_tokens": 1457927.0,
      "step": 585
    },
    {
      "entropy": 7.648988533020019,
      "epoch": 0.6900584795321637,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002945,
      "loss": 7.6251,
      "mean_token_accuracy": 0.12571160346269608,
      "num_tokens": 1470116.0,
      "step": 590
    },
    {
      "entropy": 7.753493022918701,
      "epoch": 0.695906432748538,
      "grad_norm": 1.375,
      "learning_rate": 0.000297,
      "loss": 7.5099,
      "mean_token_accuracy": 0.1337502546608448,
      "num_tokens": 1482067.0,
      "step": 595
    },
    {
      "entropy": 7.677929210662842,
      "epoch": 0.7017543859649122,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002995,
      "loss": 7.598,
      "mean_token_accuracy": 0.12390129417181014,
      "num_tokens": 1493902.0,
      "step": 600
    },
    {
      "entropy": 7.769023275375366,
      "epoch": 0.7076023391812866,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000302,
      "loss": 7.5527,
      "mean_token_accuracy": 0.12497177794575691,
      "num_tokens": 1505689.0,
      "step": 605
    },
    {
      "entropy": 7.566888380050659,
      "epoch": 0.7134502923976608,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003045,
      "loss": 7.5007,
      "mean_token_accuracy": 0.14484675377607345,
      "num_tokens": 1518002.0,
      "step": 610
    },
    {
      "entropy": 7.677926254272461,
      "epoch": 0.7192982456140351,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000307,
      "loss": 7.5296,
      "mean_token_accuracy": 0.12993534058332443,
      "num_tokens": 1529307.0,
      "step": 615
    },
    {
      "entropy": 7.742590284347534,
      "epoch": 0.7251461988304093,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003095,
      "loss": 7.6741,
      "mean_token_accuracy": 0.11497443467378617,
      "num_tokens": 1542351.0,
      "step": 620
    },
    {
      "entropy": 7.50280613899231,
      "epoch": 0.7309941520467836,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000312,
      "loss": 7.5082,
      "mean_token_accuracy": 0.13412842527031898,
      "num_tokens": 1556263.0,
      "step": 625
    },
    {
      "entropy": 7.664288806915283,
      "epoch": 0.7368421052631579,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003145,
      "loss": 7.5242,
      "mean_token_accuracy": 0.1329628087580204,
      "num_tokens": 1568906.0,
      "step": 630
    },
    {
      "entropy": 7.606091737747192,
      "epoch": 0.7426900584795322,
      "grad_norm": 1.046875,
      "learning_rate": 0.000317,
      "loss": 7.5109,
      "mean_token_accuracy": 0.13439518883824347,
      "num_tokens": 1580631.0,
      "step": 635
    },
    {
      "entropy": 7.604536008834839,
      "epoch": 0.7485380116959064,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003195,
      "loss": 7.5301,
      "mean_token_accuracy": 0.1288958840072155,
      "num_tokens": 1592369.0,
      "step": 640
    },
    {
      "entropy": 7.587004566192627,
      "epoch": 0.7543859649122807,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000322,
      "loss": 7.5573,
      "mean_token_accuracy": 0.1358466237783432,
      "num_tokens": 1604888.0,
      "step": 645
    },
    {
      "entropy": 7.692623138427734,
      "epoch": 0.7602339181286549,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032450000000000003,
      "loss": 7.4648,
      "mean_token_accuracy": 0.13601355329155923,
      "num_tokens": 1617128.0,
      "step": 650
    },
    {
      "entropy": 7.592813205718994,
      "epoch": 0.7660818713450293,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032700000000000003,
      "loss": 7.5261,
      "mean_token_accuracy": 0.12374127358198166,
      "num_tokens": 1629350.0,
      "step": 655
    },
    {
      "entropy": 7.661840629577637,
      "epoch": 0.7719298245614035,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032950000000000004,
      "loss": 7.5617,
      "mean_token_accuracy": 0.1267032027244568,
      "num_tokens": 1643082.0,
      "step": 660
    },
    {
      "entropy": 7.53505277633667,
      "epoch": 0.7777777777777778,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00033200000000000005,
      "loss": 7.5229,
      "mean_token_accuracy": 0.13297795206308366,
      "num_tokens": 1654785.0,
      "step": 665
    },
    {
      "entropy": 7.66111946105957,
      "epoch": 0.783625730994152,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033450000000000005,
      "loss": 7.565,
      "mean_token_accuracy": 0.12604237273335456,
      "num_tokens": 1667704.0,
      "step": 670
    },
    {
      "entropy": 7.700377321243286,
      "epoch": 0.7894736842105263,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000337,
      "loss": 7.5012,
      "mean_token_accuracy": 0.12841907143592834,
      "num_tokens": 1680459.0,
      "step": 675
    },
    {
      "entropy": 7.515329790115357,
      "epoch": 0.7953216374269005,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003395,
      "loss": 7.4115,
      "mean_token_accuracy": 0.1341998517513275,
      "num_tokens": 1692912.0,
      "step": 680
    },
    {
      "entropy": 7.572557640075684,
      "epoch": 0.8011695906432749,
      "grad_norm": 2.421875,
      "learning_rate": 0.000342,
      "loss": 7.4348,
      "mean_token_accuracy": 0.13281576186418534,
      "num_tokens": 1705329.0,
      "step": 685
    },
    {
      "entropy": 7.603346347808838,
      "epoch": 0.8070175438596491,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034449999999999997,
      "loss": 7.5435,
      "mean_token_accuracy": 0.12343821972608567,
      "num_tokens": 1717283.0,
      "step": 690
    },
    {
      "entropy": 7.513457489013672,
      "epoch": 0.8128654970760234,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000347,
      "loss": 7.483,
      "mean_token_accuracy": 0.1284686468541622,
      "num_tokens": 1730308.0,
      "step": 695
    },
    {
      "entropy": 7.676495838165283,
      "epoch": 0.8187134502923976,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003495,
      "loss": 7.5498,
      "mean_token_accuracy": 0.1264888271689415,
      "num_tokens": 1742716.0,
      "step": 700
    },
    {
      "entropy": 7.500397348403931,
      "epoch": 0.8245614035087719,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000352,
      "loss": 7.4877,
      "mean_token_accuracy": 0.13529630452394487,
      "num_tokens": 1754183.0,
      "step": 705
    },
    {
      "entropy": 7.494405746459961,
      "epoch": 0.8304093567251462,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003545,
      "loss": 7.4888,
      "mean_token_accuracy": 0.1257370926439762,
      "num_tokens": 1766601.0,
      "step": 710
    },
    {
      "entropy": 7.553955984115601,
      "epoch": 0.8362573099415205,
      "grad_norm": 1.7109375,
      "learning_rate": 0.000357,
      "loss": 7.3649,
      "mean_token_accuracy": 0.14501583501696585,
      "num_tokens": 1778756.0,
      "step": 715
    },
    {
      "entropy": 7.401547574996949,
      "epoch": 0.8421052631578947,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003595,
      "loss": 7.4454,
      "mean_token_accuracy": 0.13954100236296654,
      "num_tokens": 1791485.0,
      "step": 720
    },
    {
      "entropy": 7.540654993057251,
      "epoch": 0.847953216374269,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000362,
      "loss": 7.4696,
      "mean_token_accuracy": 0.12822862416505815,
      "num_tokens": 1803712.0,
      "step": 725
    },
    {
      "entropy": 7.457326412200928,
      "epoch": 0.8538011695906432,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003645,
      "loss": 7.4343,
      "mean_token_accuracy": 0.12846877947449684,
      "num_tokens": 1818387.0,
      "step": 730
    },
    {
      "entropy": 7.5405213832855225,
      "epoch": 0.8596491228070176,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000367,
      "loss": 7.4252,
      "mean_token_accuracy": 0.1321238525211811,
      "num_tokens": 1830113.0,
      "step": 735
    },
    {
      "entropy": 7.537390804290771,
      "epoch": 0.8654970760233918,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003695,
      "loss": 7.5229,
      "mean_token_accuracy": 0.13313855454325677,
      "num_tokens": 1841789.0,
      "step": 740
    },
    {
      "entropy": 7.509795761108398,
      "epoch": 0.8713450292397661,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000372,
      "loss": 7.389,
      "mean_token_accuracy": 0.13464695289731027,
      "num_tokens": 1853669.0,
      "step": 745
    },
    {
      "entropy": 7.390686273574829,
      "epoch": 0.8771929824561403,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003745,
      "loss": 7.456,
      "mean_token_accuracy": 0.1308678649365902,
      "num_tokens": 1865322.0,
      "step": 750
    },
    {
      "entropy": 7.4917685985565186,
      "epoch": 0.8830409356725146,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000377,
      "loss": 7.3569,
      "mean_token_accuracy": 0.13735375180840492,
      "num_tokens": 1879039.0,
      "step": 755
    },
    {
      "entropy": 7.56513204574585,
      "epoch": 0.8888888888888888,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003795,
      "loss": 7.499,
      "mean_token_accuracy": 0.12612639144062995,
      "num_tokens": 1893078.0,
      "step": 760
    },
    {
      "entropy": 7.541068363189697,
      "epoch": 0.8947368421052632,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000382,
      "loss": 7.5139,
      "mean_token_accuracy": 0.12583210840821266,
      "num_tokens": 1907399.0,
      "step": 765
    },
    {
      "entropy": 7.450762844085693,
      "epoch": 0.9005847953216374,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003845,
      "loss": 7.4699,
      "mean_token_accuracy": 0.1259562186896801,
      "num_tokens": 1919290.0,
      "step": 770
    },
    {
      "entropy": 7.545792961120606,
      "epoch": 0.9064327485380117,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038700000000000003,
      "loss": 7.4182,
      "mean_token_accuracy": 0.13090129122138022,
      "num_tokens": 1931658.0,
      "step": 775
    },
    {
      "entropy": 7.500565242767334,
      "epoch": 0.9122807017543859,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00038950000000000003,
      "loss": 7.4377,
      "mean_token_accuracy": 0.12877111062407492,
      "num_tokens": 1943881.0,
      "step": 780
    },
    {
      "entropy": 7.516105222702026,
      "epoch": 0.9181286549707602,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039200000000000004,
      "loss": 7.3502,
      "mean_token_accuracy": 0.13375887349247934,
      "num_tokens": 1956288.0,
      "step": 785
    },
    {
      "entropy": 7.317897605895996,
      "epoch": 0.9239766081871345,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039450000000000005,
      "loss": 7.2505,
      "mean_token_accuracy": 0.1360914021730423,
      "num_tokens": 1967265.0,
      "step": 790
    },
    {
      "entropy": 7.435814285278321,
      "epoch": 0.9298245614035088,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039700000000000005,
      "loss": 7.4049,
      "mean_token_accuracy": 0.13047077506780624,
      "num_tokens": 1980039.0,
      "step": 795
    },
    {
      "entropy": 7.459825086593628,
      "epoch": 0.935672514619883,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003995,
      "loss": 7.361,
      "mean_token_accuracy": 0.13110343888401985,
      "num_tokens": 1992735.0,
      "step": 800
    },
    {
      "entropy": 7.465135145187378,
      "epoch": 0.9415204678362573,
      "grad_norm": 1.015625,
      "learning_rate": 0.000402,
      "loss": 7.3883,
      "mean_token_accuracy": 0.13711876720190047,
      "num_tokens": 2004545.0,
      "step": 805
    },
    {
      "entropy": 7.346128177642822,
      "epoch": 0.9473684210526315,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004045,
      "loss": 7.3759,
      "mean_token_accuracy": 0.13430287092924117,
      "num_tokens": 2017100.0,
      "step": 810
    },
    {
      "entropy": 7.417292737960816,
      "epoch": 0.9532163742690059,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040699999999999997,
      "loss": 7.3286,
      "mean_token_accuracy": 0.14560126513242722,
      "num_tokens": 2029868.0,
      "step": 815
    },
    {
      "entropy": 7.466277170181274,
      "epoch": 0.9590643274853801,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004095,
      "loss": 7.3827,
      "mean_token_accuracy": 0.12977704107761384,
      "num_tokens": 2041334.0,
      "step": 820
    },
    {
      "entropy": 7.440905380249023,
      "epoch": 0.9649122807017544,
      "grad_norm": 1.078125,
      "learning_rate": 0.000412,
      "loss": 7.448,
      "mean_token_accuracy": 0.1350324772298336,
      "num_tokens": 2054797.0,
      "step": 825
    },
    {
      "entropy": 7.432639074325562,
      "epoch": 0.9707602339181286,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004145,
      "loss": 7.3314,
      "mean_token_accuracy": 0.13245931789278984,
      "num_tokens": 2068337.0,
      "step": 830
    },
    {
      "entropy": 7.450676774978637,
      "epoch": 0.9766081871345029,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000417,
      "loss": 7.363,
      "mean_token_accuracy": 0.1392313800752163,
      "num_tokens": 2079984.0,
      "step": 835
    },
    {
      "entropy": 7.329059696197509,
      "epoch": 0.9824561403508771,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004195,
      "loss": 7.3717,
      "mean_token_accuracy": 0.1394517168402672,
      "num_tokens": 2093069.0,
      "step": 840
    },
    {
      "entropy": 7.384470462799072,
      "epoch": 0.9883040935672515,
      "grad_norm": 1.0,
      "learning_rate": 0.000422,
      "loss": 7.3291,
      "mean_token_accuracy": 0.13665454760193824,
      "num_tokens": 2104753.0,
      "step": 845
    },
    {
      "entropy": 7.348030519485474,
      "epoch": 0.9941520467836257,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004245,
      "loss": 7.3418,
      "mean_token_accuracy": 0.1328301377594471,
      "num_tokens": 2116576.0,
      "step": 850
    },
    {
      "entropy": 7.415113210678101,
      "epoch": 1.0,
      "grad_norm": 1.140625,
      "learning_rate": 0.000427,
      "loss": 7.3202,
      "mean_token_accuracy": 0.1307046242058277,
      "num_tokens": 2129484.0,
      "step": 855
    },
    {
      "entropy": 7.309794139862061,
      "epoch": 1.0058479532163742,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004295,
      "loss": 7.1455,
      "mean_token_accuracy": 0.14334668442606927,
      "num_tokens": 2142000.0,
      "step": 860
    },
    {
      "entropy": 7.259340381622314,
      "epoch": 1.0116959064327484,
      "grad_norm": 1.078125,
      "learning_rate": 0.000432,
      "loss": 7.1,
      "mean_token_accuracy": 0.1472266711294651,
      "num_tokens": 2152965.0,
      "step": 865
    },
    {
      "entropy": 7.345311832427979,
      "epoch": 1.0175438596491229,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004345,
      "loss": 7.1904,
      "mean_token_accuracy": 0.13219141513109206,
      "num_tokens": 2165078.0,
      "step": 870
    },
    {
      "entropy": 7.180283641815185,
      "epoch": 1.023391812865497,
      "grad_norm": 1.171875,
      "learning_rate": 0.000437,
      "loss": 7.1359,
      "mean_token_accuracy": 0.14372573569417,
      "num_tokens": 2176706.0,
      "step": 875
    },
    {
      "entropy": 7.374202919006348,
      "epoch": 1.0292397660818713,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004395,
      "loss": 7.1526,
      "mean_token_accuracy": 0.13850091844797136,
      "num_tokens": 2190157.0,
      "step": 880
    },
    {
      "entropy": 7.3007509231567385,
      "epoch": 1.0350877192982457,
      "grad_norm": 1.015625,
      "learning_rate": 0.000442,
      "loss": 7.1248,
      "mean_token_accuracy": 0.14188719019293786,
      "num_tokens": 2201821.0,
      "step": 885
    },
    {
      "entropy": 7.164031553268432,
      "epoch": 1.04093567251462,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004445,
      "loss": 7.0664,
      "mean_token_accuracy": 0.1440157599747181,
      "num_tokens": 2214504.0,
      "step": 890
    },
    {
      "entropy": 7.243292570114136,
      "epoch": 1.0467836257309941,
      "grad_norm": 1.078125,
      "learning_rate": 0.000447,
      "loss": 7.1924,
      "mean_token_accuracy": 0.13835916370153428,
      "num_tokens": 2226318.0,
      "step": 895
    },
    {
      "entropy": 7.315338039398194,
      "epoch": 1.0526315789473684,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044950000000000003,
      "loss": 7.2242,
      "mean_token_accuracy": 0.13328551575541497,
      "num_tokens": 2237892.0,
      "step": 900
    },
    {
      "entropy": 7.251867389678955,
      "epoch": 1.0584795321637426,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045200000000000004,
      "loss": 7.1719,
      "mean_token_accuracy": 0.1382812075316906,
      "num_tokens": 2250656.0,
      "step": 905
    },
    {
      "entropy": 7.2664087295532225,
      "epoch": 1.064327485380117,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00045450000000000004,
      "loss": 7.1879,
      "mean_token_accuracy": 0.13563519567251206,
      "num_tokens": 2263685.0,
      "step": 910
    },
    {
      "entropy": 7.246600580215454,
      "epoch": 1.0701754385964912,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00045700000000000005,
      "loss": 7.1603,
      "mean_token_accuracy": 0.13542255833745004,
      "num_tokens": 2275413.0,
      "step": 915
    },
    {
      "entropy": 7.257514142990113,
      "epoch": 1.0760233918128654,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00045950000000000006,
      "loss": 7.1339,
      "mean_token_accuracy": 0.13871516659855843,
      "num_tokens": 2288165.0,
      "step": 920
    },
    {
      "entropy": 7.2856926918029785,
      "epoch": 1.0818713450292399,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000462,
      "loss": 7.1488,
      "mean_token_accuracy": 0.1338664099574089,
      "num_tokens": 2300574.0,
      "step": 925
    },
    {
      "entropy": 7.196798849105835,
      "epoch": 1.087719298245614,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004645,
      "loss": 7.148,
      "mean_token_accuracy": 0.1378339521586895,
      "num_tokens": 2311631.0,
      "step": 930
    },
    {
      "entropy": 7.23977427482605,
      "epoch": 1.0935672514619883,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000467,
      "loss": 7.169,
      "mean_token_accuracy": 0.1378619410097599,
      "num_tokens": 2324312.0,
      "step": 935
    },
    {
      "entropy": 7.282365465164185,
      "epoch": 1.0994152046783625,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004695,
      "loss": 7.2368,
      "mean_token_accuracy": 0.1315642088651657,
      "num_tokens": 2337747.0,
      "step": 940
    },
    {
      "entropy": 7.191779136657715,
      "epoch": 1.1052631578947367,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000472,
      "loss": 7.1814,
      "mean_token_accuracy": 0.13839799612760545,
      "num_tokens": 2349830.0,
      "step": 945
    },
    {
      "entropy": 7.2542595863342285,
      "epoch": 1.1111111111111112,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004745,
      "loss": 7.1209,
      "mean_token_accuracy": 0.14533643126487733,
      "num_tokens": 2361825.0,
      "step": 950
    },
    {
      "entropy": 7.1857686042785645,
      "epoch": 1.1169590643274854,
      "grad_norm": 0.921875,
      "learning_rate": 0.000477,
      "loss": 7.0716,
      "mean_token_accuracy": 0.14158036708831787,
      "num_tokens": 2373963.0,
      "step": 955
    },
    {
      "entropy": 7.213093662261963,
      "epoch": 1.1228070175438596,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004795,
      "loss": 7.2192,
      "mean_token_accuracy": 0.13446253091096877,
      "num_tokens": 2386348.0,
      "step": 960
    },
    {
      "entropy": 7.232397890090942,
      "epoch": 1.128654970760234,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000482,
      "loss": 7.1337,
      "mean_token_accuracy": 0.13307930007576943,
      "num_tokens": 2398338.0,
      "step": 965
    },
    {
      "entropy": 7.1732768535614015,
      "epoch": 1.1345029239766082,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004845,
      "loss": 7.0945,
      "mean_token_accuracy": 0.1407437168061733,
      "num_tokens": 2411855.0,
      "step": 970
    },
    {
      "entropy": 7.160850143432617,
      "epoch": 1.1403508771929824,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000487,
      "loss": 7.1619,
      "mean_token_accuracy": 0.14437102526426315,
      "num_tokens": 2424271.0,
      "step": 975
    },
    {
      "entropy": 7.1891566753387455,
      "epoch": 1.1461988304093567,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004895,
      "loss": 7.1708,
      "mean_token_accuracy": 0.1375895693898201,
      "num_tokens": 2436668.0,
      "step": 980
    },
    {
      "entropy": 7.174351406097412,
      "epoch": 1.1520467836257309,
      "grad_norm": 1.0625,
      "learning_rate": 0.000492,
      "loss": 7.1176,
      "mean_token_accuracy": 0.13666361346840858,
      "num_tokens": 2449451.0,
      "step": 985
    },
    {
      "entropy": 7.188272762298584,
      "epoch": 1.1578947368421053,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004945,
      "loss": 7.2191,
      "mean_token_accuracy": 0.13468535766005515,
      "num_tokens": 2463107.0,
      "step": 990
    },
    {
      "entropy": 7.275109195709229,
      "epoch": 1.1637426900584795,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000497,
      "loss": 7.1589,
      "mean_token_accuracy": 0.13581722751259803,
      "num_tokens": 2475443.0,
      "step": 995
    },
    {
      "entropy": 7.135958623886109,
      "epoch": 1.1695906432748537,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004995,
      "loss": 7.0839,
      "mean_token_accuracy": 0.1393029898405075,
      "num_tokens": 2486884.0,
      "step": 1000
    },
    {
      "epoch": 1.1695906432748537,
      "eval_entropy": 7.071001939013996,
      "eval_loss": 7.3009033203125,
      "eval_mean_token_accuracy": 0.12748390563920273,
      "eval_num_tokens": 2486884.0,
      "eval_runtime": 2.2417,
      "eval_samples_per_second": 1207.555,
      "eval_steps_per_second": 151.223,
      "step": 1000
    },
    {
      "entropy": 7.080128574371338,
      "epoch": 1.1754385964912282,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999996875147963,
      "loss": 7.0353,
      "mean_token_accuracy": 0.14896628335118295,
      "num_tokens": 2499310.0,
      "step": 1005
    },
    {
      "entropy": 7.226775979995727,
      "epoch": 1.1812865497076024,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999984180451436,
      "loss": 7.1649,
      "mean_token_accuracy": 0.1328574500977993,
      "num_tokens": 2512230.0,
      "step": 1010
    },
    {
      "entropy": 7.102473497390747,
      "epoch": 1.1871345029239766,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999961720662221,
      "loss": 7.2272,
      "mean_token_accuracy": 0.13584866672754287,
      "num_tokens": 2525255.0,
      "step": 1015
    },
    {
      "entropy": 7.121642780303955,
      "epoch": 1.1929824561403508,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999929495877796,
      "loss": 7.118,
      "mean_token_accuracy": 0.13906984254717827,
      "num_tokens": 2537672.0,
      "step": 1020
    },
    {
      "entropy": 7.139839220046997,
      "epoch": 1.198830409356725,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999887506238018,
      "loss": 7.1251,
      "mean_token_accuracy": 0.13929466605186464,
      "num_tokens": 2550371.0,
      "step": 1025
    },
    {
      "entropy": 7.115478038787842,
      "epoch": 1.2046783625730995,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999835751925124,
      "loss": 7.1417,
      "mean_token_accuracy": 0.13262443989515305,
      "num_tokens": 2563708.0,
      "step": 1030
    },
    {
      "entropy": 7.243052911758423,
      "epoch": 1.2105263157894737,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999774233163734,
      "loss": 7.1046,
      "mean_token_accuracy": 0.13376289308071138,
      "num_tokens": 2576529.0,
      "step": 1035
    },
    {
      "entropy": 7.117304182052612,
      "epoch": 1.2163742690058479,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999702950220841,
      "loss": 7.1329,
      "mean_token_accuracy": 0.13227491527795793,
      "num_tokens": 2588970.0,
      "step": 1040
    },
    {
      "entropy": 7.104359102249146,
      "epoch": 1.2222222222222223,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999621903405819,
      "loss": 7.0833,
      "mean_token_accuracy": 0.136895339936018,
      "num_tokens": 2601134.0,
      "step": 1045
    },
    {
      "entropy": 7.140307426452637,
      "epoch": 1.2280701754385965,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499953109307042,
      "loss": 7.1338,
      "mean_token_accuracy": 0.13351348638534546,
      "num_tokens": 2612670.0,
      "step": 1050
    },
    {
      "entropy": 7.080586051940918,
      "epoch": 1.2339181286549707,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999430519608762,
      "loss": 7.1057,
      "mean_token_accuracy": 0.14540162831544876,
      "num_tokens": 2625157.0,
      "step": 1055
    },
    {
      "entropy": 7.201329231262207,
      "epoch": 1.239766081871345,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999320183457346,
      "loss": 7.1085,
      "mean_token_accuracy": 0.13505723252892493,
      "num_tokens": 2638318.0,
      "step": 1060
    },
    {
      "entropy": 7.053661489486695,
      "epoch": 1.2456140350877192,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999200085095035,
      "loss": 7.074,
      "mean_token_accuracy": 0.13898903355002404,
      "num_tokens": 2651024.0,
      "step": 1065
    },
    {
      "entropy": 6.962784957885742,
      "epoch": 1.2514619883040936,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999070225043068,
      "loss": 7.0091,
      "mean_token_accuracy": 0.14594187438488007,
      "num_tokens": 2665568.0,
      "step": 1070
    },
    {
      "entropy": 7.270829439163208,
      "epoch": 1.2573099415204678,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998930603865042,
      "loss": 7.14,
      "mean_token_accuracy": 0.13617520555853843,
      "num_tokens": 2678748.0,
      "step": 1075
    },
    {
      "entropy": 7.042156457901001,
      "epoch": 1.263157894736842,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998781222166929,
      "loss": 7.0922,
      "mean_token_accuracy": 0.13991621881723404,
      "num_tokens": 2690682.0,
      "step": 1080
    },
    {
      "entropy": 7.046918058395386,
      "epoch": 1.2690058479532165,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998622080597052,
      "loss": 7.1164,
      "mean_token_accuracy": 0.1319088101387024,
      "num_tokens": 2703077.0,
      "step": 1085
    },
    {
      "entropy": 7.098138236999512,
      "epoch": 1.2748538011695907,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998453179846098,
      "loss": 7.0887,
      "mean_token_accuracy": 0.13736322820186614,
      "num_tokens": 2714609.0,
      "step": 1090
    },
    {
      "entropy": 7.1358263969421385,
      "epoch": 1.280701754385965,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998274520647109,
      "loss": 7.0558,
      "mean_token_accuracy": 0.14192912206053734,
      "num_tokens": 2726907.0,
      "step": 1095
    },
    {
      "entropy": 7.00035605430603,
      "epoch": 1.286549707602339,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998086103775478,
      "loss": 7.0525,
      "mean_token_accuracy": 0.1388276733458042,
      "num_tokens": 2739248.0,
      "step": 1100
    },
    {
      "entropy": 7.098528242111206,
      "epoch": 1.2923976608187133,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004997887930048947,
      "loss": 7.0155,
      "mean_token_accuracy": 0.14455852061510086,
      "num_tokens": 2753842.0,
      "step": 1105
    },
    {
      "entropy": 6.9865405559539795,
      "epoch": 1.2982456140350878,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004997680000327607,
      "loss": 7.0665,
      "mean_token_accuracy": 0.13506068885326386,
      "num_tokens": 2767249.0,
      "step": 1110
    },
    {
      "entropy": 7.14464693069458,
      "epoch": 1.304093567251462,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997462315513884,
      "loss": 7.0602,
      "mean_token_accuracy": 0.1384397678077221,
      "num_tokens": 2779657.0,
      "step": 1115
    },
    {
      "entropy": 7.082661390304565,
      "epoch": 1.3099415204678362,
      "grad_norm": 1.6328125,
      "learning_rate": 0.000499723487655255,
      "loss": 6.9763,
      "mean_token_accuracy": 0.14285208955407142,
      "num_tokens": 2792760.0,
      "step": 1120
    },
    {
      "entropy": 6.954230690002442,
      "epoch": 1.3157894736842106,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004996997684430703,
      "loss": 7.0164,
      "mean_token_accuracy": 0.1394075021147728,
      "num_tokens": 2805277.0,
      "step": 1125
    },
    {
      "entropy": 7.065880632400512,
      "epoch": 1.3216374269005848,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004996750740177775,
      "loss": 7.1166,
      "mean_token_accuracy": 0.13852448090910913,
      "num_tokens": 2818341.0,
      "step": 1130
    },
    {
      "entropy": 7.145817613601684,
      "epoch": 1.327485380116959,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004996494044865521,
      "loss": 7.0604,
      "mean_token_accuracy": 0.13684254363179207,
      "num_tokens": 2830199.0,
      "step": 1135
    },
    {
      "entropy": 6.950229167938232,
      "epoch": 1.3333333333333333,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996227599608018,
      "loss": 6.9667,
      "mean_token_accuracy": 0.14581984430551528,
      "num_tokens": 2842141.0,
      "step": 1140
    },
    {
      "entropy": 7.08269419670105,
      "epoch": 1.3391812865497075,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995951405561658,
      "loss": 7.0522,
      "mean_token_accuracy": 0.13312725201249123,
      "num_tokens": 2854929.0,
      "step": 1145
    },
    {
      "entropy": 6.987738466262817,
      "epoch": 1.345029239766082,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004995665463925142,
      "loss": 7.0516,
      "mean_token_accuracy": 0.14139380380511285,
      "num_tokens": 2867034.0,
      "step": 1150
    },
    {
      "entropy": 7.075216960906983,
      "epoch": 1.3508771929824561,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004995369775939478,
      "loss": 6.9576,
      "mean_token_accuracy": 0.14617282450199126,
      "num_tokens": 2881254.0,
      "step": 1155
    },
    {
      "entropy": 6.949263334274292,
      "epoch": 1.3567251461988303,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004995064342887972,
      "loss": 7.0713,
      "mean_token_accuracy": 0.14057645201683044,
      "num_tokens": 2894592.0,
      "step": 1160
    },
    {
      "entropy": 7.0024169921875,
      "epoch": 1.3625730994152048,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994749166096226,
      "loss": 6.9747,
      "mean_token_accuracy": 0.14208125099539756,
      "num_tokens": 2907196.0,
      "step": 1165
    },
    {
      "entropy": 7.013747358322144,
      "epoch": 1.368421052631579,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004994424246932129,
      "loss": 7.0602,
      "mean_token_accuracy": 0.13577224165201188,
      "num_tokens": 2920398.0,
      "step": 1170
    },
    {
      "entropy": 6.958837985992432,
      "epoch": 1.3742690058479532,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994089586805856,
      "loss": 6.9887,
      "mean_token_accuracy": 0.1430194616317749,
      "num_tokens": 2932052.0,
      "step": 1175
    },
    {
      "entropy": 7.202078723907471,
      "epoch": 1.3801169590643274,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004993745187169852,
      "loss": 7.054,
      "mean_token_accuracy": 0.1366309255361557,
      "num_tokens": 2943620.0,
      "step": 1180
    },
    {
      "entropy": 6.892934131622314,
      "epoch": 1.3859649122807016,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499339104951884,
      "loss": 7.0283,
      "mean_token_accuracy": 0.14119249433279038,
      "num_tokens": 2956159.0,
      "step": 1185
    },
    {
      "entropy": 7.062256336212158,
      "epoch": 1.391812865497076,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00049930271753898,
      "loss": 7.0003,
      "mean_token_accuracy": 0.13841674402356147,
      "num_tokens": 2967332.0,
      "step": 1190
    },
    {
      "entropy": 6.983926677703858,
      "epoch": 1.3976608187134503,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992653566361974,
      "loss": 7.0041,
      "mean_token_accuracy": 0.13496125042438506,
      "num_tokens": 2980168.0,
      "step": 1195
    },
    {
      "entropy": 7.037136030197144,
      "epoch": 1.4035087719298245,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004992270224056849,
      "loss": 7.0935,
      "mean_token_accuracy": 0.1383035495877266,
      "num_tokens": 2993509.0,
      "step": 1200
    },
    {
      "entropy": 7.023516464233398,
      "epoch": 1.409356725146199,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991877150138162,
      "loss": 7.0183,
      "mean_token_accuracy": 0.1392554260790348,
      "num_tokens": 3005827.0,
      "step": 1205
    },
    {
      "entropy": 6.972181892395019,
      "epoch": 1.4152046783625731,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004991474346311879,
      "loss": 6.9922,
      "mean_token_accuracy": 0.14359172955155372,
      "num_tokens": 3018968.0,
      "step": 1210
    },
    {
      "entropy": 7.086374950408936,
      "epoch": 1.4210526315789473,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004991061814326198,
      "loss": 7.0242,
      "mean_token_accuracy": 0.13482476994395257,
      "num_tokens": 3031892.0,
      "step": 1215
    },
    {
      "entropy": 6.958857440948487,
      "epoch": 1.4269005847953216,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004990639555971539,
      "loss": 6.9632,
      "mean_token_accuracy": 0.14786728024482726,
      "num_tokens": 3045994.0,
      "step": 1220
    },
    {
      "entropy": 6.935246562957763,
      "epoch": 1.4327485380116958,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004990207573080533,
      "loss": 6.9662,
      "mean_token_accuracy": 0.14441566094756125,
      "num_tokens": 3058304.0,
      "step": 1225
    },
    {
      "entropy": 6.971540212631226,
      "epoch": 1.4385964912280702,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004989765867528014,
      "loss": 6.9762,
      "mean_token_accuracy": 0.1433918684720993,
      "num_tokens": 3071102.0,
      "step": 1230
    },
    {
      "entropy": 6.901489162445069,
      "epoch": 1.4444444444444444,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004989314441231019,
      "loss": 7.0027,
      "mean_token_accuracy": 0.13700876906514167,
      "num_tokens": 3084653.0,
      "step": 1235
    },
    {
      "entropy": 7.029861402511597,
      "epoch": 1.4502923976608186,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000498885329614877,
      "loss": 7.0022,
      "mean_token_accuracy": 0.14563447535037993,
      "num_tokens": 3097425.0,
      "step": 1240
    },
    {
      "entropy": 6.994398736953736,
      "epoch": 1.456140350877193,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004988382434282668,
      "loss": 7.0236,
      "mean_token_accuracy": 0.13943543583154677,
      "num_tokens": 3111898.0,
      "step": 1245
    },
    {
      "entropy": 7.006692504882812,
      "epoch": 1.4619883040935673,
      "grad_norm": 1.015625,
      "learning_rate": 0.000498790185767629,
      "loss": 6.9668,
      "mean_token_accuracy": 0.14214081391692163,
      "num_tokens": 3124271.0,
      "step": 1250
    },
    {
      "entropy": 6.997828435897827,
      "epoch": 1.4678362573099415,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000498741156841537,
      "loss": 7.0595,
      "mean_token_accuracy": 0.14109789729118347,
      "num_tokens": 3136344.0,
      "step": 1255
    },
    {
      "entropy": 6.945074844360351,
      "epoch": 1.4736842105263157,
      "grad_norm": 1.109375,
      "learning_rate": 0.00049869115686278,
      "loss": 6.9611,
      "mean_token_accuracy": 0.14131287783384322,
      "num_tokens": 3148299.0,
      "step": 1260
    },
    {
      "entropy": 7.050410270690918,
      "epoch": 1.47953216374269,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004986401860483615,
      "loss": 7.0176,
      "mean_token_accuracy": 0.13987123295664788,
      "num_tokens": 3161537.0,
      "step": 1265
    },
    {
      "entropy": 6.907403945922852,
      "epoch": 1.4853801169590644,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004985882446194984,
      "loss": 6.9717,
      "mean_token_accuracy": 0.13868435025215148,
      "num_tokens": 3173676.0,
      "step": 1270
    },
    {
      "entropy": 7.050081634521485,
      "epoch": 1.4912280701754386,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004985353328016204,
      "loss": 6.9217,
      "mean_token_accuracy": 0.15403224006295205,
      "num_tokens": 3186542.0,
      "step": 1275
    },
    {
      "entropy": 6.8447894096374515,
      "epoch": 1.4970760233918128,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004984814508243685,
      "loss": 6.9633,
      "mean_token_accuracy": 0.1440097950398922,
      "num_tokens": 3197867.0,
      "step": 1280
    },
    {
      "entropy": 7.016148805618286,
      "epoch": 1.5029239766081872,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004984265989215942,
      "loss": 6.919,
      "mean_token_accuracy": 0.14048407971858978,
      "num_tokens": 3209671.0,
      "step": 1285
    },
    {
      "entropy": 6.8506049633026125,
      "epoch": 1.5087719298245614,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004983707773313589,
      "loss": 6.9187,
      "mean_token_accuracy": 0.14848409295082093,
      "num_tokens": 3220707.0,
      "step": 1290
    },
    {
      "entropy": 6.872060537338257,
      "epoch": 1.5146198830409356,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004983139862959322,
      "loss": 6.8841,
      "mean_token_accuracy": 0.1535121612250805,
      "num_tokens": 3235122.0,
      "step": 1295
    },
    {
      "entropy": 6.986389875411987,
      "epoch": 1.52046783625731,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004982562260617912,
      "loss": 6.8776,
      "mean_token_accuracy": 0.13864699825644494,
      "num_tokens": 3246833.0,
      "step": 1300
    },
    {
      "entropy": 6.929480314254761,
      "epoch": 1.526315789473684,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004981974968796197,
      "loss": 6.9988,
      "mean_token_accuracy": 0.13592579811811448,
      "num_tokens": 3260493.0,
      "step": 1305
    },
    {
      "entropy": 6.903130149841308,
      "epoch": 1.5321637426900585,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004981377990043064,
      "loss": 6.916,
      "mean_token_accuracy": 0.14037951976060867,
      "num_tokens": 3272445.0,
      "step": 1310
    },
    {
      "entropy": 6.899147319793701,
      "epoch": 1.5380116959064327,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004980771326949442,
      "loss": 6.9349,
      "mean_token_accuracy": 0.1383832648396492,
      "num_tokens": 3284360.0,
      "step": 1315
    },
    {
      "entropy": 6.97801103591919,
      "epoch": 1.543859649122807,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004980154982148296,
      "loss": 6.934,
      "mean_token_accuracy": 0.14036118239164352,
      "num_tokens": 3296841.0,
      "step": 1320
    },
    {
      "entropy": 6.864973831176758,
      "epoch": 1.5497076023391814,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004979528958314604,
      "loss": 6.9121,
      "mean_token_accuracy": 0.13837594985961915,
      "num_tokens": 3309393.0,
      "step": 1325
    },
    {
      "entropy": 6.999245882034302,
      "epoch": 1.5555555555555556,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004978893258165354,
      "loss": 6.9954,
      "mean_token_accuracy": 0.14061468243598937,
      "num_tokens": 3322094.0,
      "step": 1330
    },
    {
      "entropy": 6.905670976638794,
      "epoch": 1.5614035087719298,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004978247884459532,
      "loss": 6.9197,
      "mean_token_accuracy": 0.14624207094311714,
      "num_tokens": 3334468.0,
      "step": 1335
    },
    {
      "entropy": 6.932331085205078,
      "epoch": 1.5672514619883042,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004977592839998105,
      "loss": 6.9064,
      "mean_token_accuracy": 0.1421583592891693,
      "num_tokens": 3346367.0,
      "step": 1340
    },
    {
      "entropy": 6.8350423812866214,
      "epoch": 1.5730994152046782,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004976928127624013,
      "loss": 6.8967,
      "mean_token_accuracy": 0.1457481265068054,
      "num_tokens": 3357463.0,
      "step": 1345
    },
    {
      "entropy": 6.8896355628967285,
      "epoch": 1.5789473684210527,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004976253750222157,
      "loss": 6.8837,
      "mean_token_accuracy": 0.1439684234559536,
      "num_tokens": 3369465.0,
      "step": 1350
    },
    {
      "entropy": 6.872499275207519,
      "epoch": 1.5847953216374269,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004975569710719382,
      "loss": 7.0025,
      "mean_token_accuracy": 0.13503989055752755,
      "num_tokens": 3382859.0,
      "step": 1355
    },
    {
      "entropy": 6.969785070419311,
      "epoch": 1.590643274853801,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004974876012084468,
      "loss": 6.9562,
      "mean_token_accuracy": 0.13781241104006767,
      "num_tokens": 3395431.0,
      "step": 1360
    },
    {
      "entropy": 6.84591851234436,
      "epoch": 1.5964912280701755,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004974172657328116,
      "loss": 6.8481,
      "mean_token_accuracy": 0.1415877401828766,
      "num_tokens": 3408388.0,
      "step": 1365
    },
    {
      "entropy": 6.912861585617065,
      "epoch": 1.6023391812865497,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004973459649502939,
      "loss": 6.864,
      "mean_token_accuracy": 0.14842946082353592,
      "num_tokens": 3421275.0,
      "step": 1370
    },
    {
      "entropy": 6.861267900466919,
      "epoch": 1.608187134502924,
      "grad_norm": 1.0,
      "learning_rate": 0.0004972736991703439,
      "loss": 6.9159,
      "mean_token_accuracy": 0.143263141810894,
      "num_tokens": 3433537.0,
      "step": 1375
    },
    {
      "entropy": 6.905822086334228,
      "epoch": 1.6140350877192984,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004972004687066005,
      "loss": 6.9023,
      "mean_token_accuracy": 0.14614611566066743,
      "num_tokens": 3445167.0,
      "step": 1380
    },
    {
      "entropy": 6.768242454528808,
      "epoch": 1.6198830409356724,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000497126273876889,
      "loss": 6.91,
      "mean_token_accuracy": 0.1506970815360546,
      "num_tokens": 3458529.0,
      "step": 1385
    },
    {
      "entropy": 6.879518270492554,
      "epoch": 1.6257309941520468,
      "grad_norm": 1.0,
      "learning_rate": 0.0004970511150032205,
      "loss": 6.9002,
      "mean_token_accuracy": 0.14469428062438966,
      "num_tokens": 3470315.0,
      "step": 1390
    },
    {
      "entropy": 6.876277732849121,
      "epoch": 1.631578947368421,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004969749924117896,
      "loss": 6.7924,
      "mean_token_accuracy": 0.14415119141340255,
      "num_tokens": 3481899.0,
      "step": 1395
    },
    {
      "entropy": 6.8515345573425295,
      "epoch": 1.6374269005847952,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004968979064329741,
      "loss": 6.9329,
      "mean_token_accuracy": 0.1479893445968628,
      "num_tokens": 3494076.0,
      "step": 1400
    },
    {
      "entropy": 7.007955503463745,
      "epoch": 1.6432748538011697,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004968198574013323,
      "loss": 6.934,
      "mean_token_accuracy": 0.13703711330890656,
      "num_tokens": 3507670.0,
      "step": 1405
    },
    {
      "entropy": 6.893377113342285,
      "epoch": 1.6491228070175439,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000496740845655603,
      "loss": 6.8404,
      "mean_token_accuracy": 0.14385494217276573,
      "num_tokens": 3519704.0,
      "step": 1410
    },
    {
      "entropy": 6.873383617401123,
      "epoch": 1.654970760233918,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004966608715387029,
      "loss": 6.8984,
      "mean_token_accuracy": 0.14520218521356582,
      "num_tokens": 3531665.0,
      "step": 1415
    },
    {
      "entropy": 6.817680501937867,
      "epoch": 1.6608187134502925,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004965799353977248,
      "loss": 6.9069,
      "mean_token_accuracy": 0.14890925586223602,
      "num_tokens": 3543876.0,
      "step": 1420
    },
    {
      "entropy": 6.816280269622803,
      "epoch": 1.6666666666666665,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004964980375839379,
      "loss": 6.8107,
      "mean_token_accuracy": 0.15025360584259034,
      "num_tokens": 3556517.0,
      "step": 1425
    },
    {
      "entropy": 6.824383068084717,
      "epoch": 1.672514619883041,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004964151784527844,
      "loss": 6.8378,
      "mean_token_accuracy": 0.15114935636520385,
      "num_tokens": 3569378.0,
      "step": 1430
    },
    {
      "entropy": 6.922686195373535,
      "epoch": 1.6783625730994152,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000496331358363879,
      "loss": 6.833,
      "mean_token_accuracy": 0.14508106857538222,
      "num_tokens": 3581684.0,
      "step": 1435
    },
    {
      "entropy": 6.764388227462769,
      "epoch": 1.6842105263157894,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004962465776810066,
      "loss": 6.7812,
      "mean_token_accuracy": 0.15873703435063363,
      "num_tokens": 3594961.0,
      "step": 1440
    },
    {
      "entropy": 6.858131456375122,
      "epoch": 1.6900584795321638,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004961608367721217,
      "loss": 6.8265,
      "mean_token_accuracy": 0.14417641311883928,
      "num_tokens": 3607197.0,
      "step": 1445
    },
    {
      "entropy": 6.812173652648926,
      "epoch": 1.695906432748538,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004960741360093459,
      "loss": 6.8627,
      "mean_token_accuracy": 0.1441764861345291,
      "num_tokens": 3618728.0,
      "step": 1450
    },
    {
      "entropy": 6.915625095367432,
      "epoch": 1.7017543859649122,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004959864757689669,
      "loss": 6.9563,
      "mean_token_accuracy": 0.13298699259757996,
      "num_tokens": 3630872.0,
      "step": 1455
    },
    {
      "entropy": 6.860816144943238,
      "epoch": 1.7076023391812867,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004958978564314361,
      "loss": 6.8413,
      "mean_token_accuracy": 0.14621406346559523,
      "num_tokens": 3642657.0,
      "step": 1460
    },
    {
      "entropy": 6.870521593093872,
      "epoch": 1.7134502923976607,
      "grad_norm": 0.96875,
      "learning_rate": 0.000495808278381368,
      "loss": 6.864,
      "mean_token_accuracy": 0.144855272769928,
      "num_tokens": 3654505.0,
      "step": 1465
    },
    {
      "entropy": 6.7053485870361325,
      "epoch": 1.719298245614035,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004957177420075377,
      "loss": 6.7807,
      "mean_token_accuracy": 0.15822129100561141,
      "num_tokens": 3668554.0,
      "step": 1470
    },
    {
      "entropy": 6.885648488998413,
      "epoch": 1.7251461988304093,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004956262477028797,
      "loss": 6.936,
      "mean_token_accuracy": 0.1396143242716789,
      "num_tokens": 3681448.0,
      "step": 1475
    },
    {
      "entropy": 6.84114933013916,
      "epoch": 1.7309941520467835,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004955337958644857,
      "loss": 6.8637,
      "mean_token_accuracy": 0.13967747539281844,
      "num_tokens": 3693677.0,
      "step": 1480
    },
    {
      "entropy": 6.7674092769622805,
      "epoch": 1.736842105263158,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004954403868936032,
      "loss": 6.8596,
      "mean_token_accuracy": 0.14602801874279975,
      "num_tokens": 3706204.0,
      "step": 1485
    },
    {
      "entropy": 6.906683778762817,
      "epoch": 1.7426900584795322,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000495346021195634,
      "loss": 6.8156,
      "mean_token_accuracy": 0.1473531424999237,
      "num_tokens": 3718255.0,
      "step": 1490
    },
    {
      "entropy": 6.838267850875854,
      "epoch": 1.7485380116959064,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004952506991801319,
      "loss": 6.8289,
      "mean_token_accuracy": 0.14677032232284545,
      "num_tokens": 3730623.0,
      "step": 1495
    },
    {
      "entropy": 6.830192804336548,
      "epoch": 1.7543859649122808,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004951544212608015,
      "loss": 6.8732,
      "mean_token_accuracy": 0.14355233311653137,
      "num_tokens": 3743636.0,
      "step": 1500
    },
    {
      "epoch": 1.7543859649122808,
      "eval_entropy": 6.78361423880653,
      "eval_loss": 6.987274169921875,
      "eval_mean_token_accuracy": 0.1345269678072636,
      "eval_num_tokens": 3743636.0,
      "eval_runtime": 2.238,
      "eval_samples_per_second": 1209.556,
      "eval_steps_per_second": 151.474,
      "step": 1500
    },
    {
      "entropy": 6.85185809135437,
      "epoch": 1.7602339181286548,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004950571878554956,
      "loss": 6.9139,
      "mean_token_accuracy": 0.13569407239556314,
      "num_tokens": 3755608.0,
      "step": 1505
    },
    {
      "entropy": 6.905317783355713,
      "epoch": 1.7660818713450293,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004949589993862143,
      "loss": 6.8189,
      "mean_token_accuracy": 0.14117416441440583,
      "num_tokens": 3768208.0,
      "step": 1510
    },
    {
      "entropy": 6.804564189910889,
      "epoch": 1.7719298245614035,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004948598562791025,
      "loss": 6.8354,
      "mean_token_accuracy": 0.14516011774539947,
      "num_tokens": 3780550.0,
      "step": 1515
    },
    {
      "entropy": 6.734309101104737,
      "epoch": 1.7777777777777777,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004947597589644484,
      "loss": 6.817,
      "mean_token_accuracy": 0.14395568445324897,
      "num_tokens": 3792417.0,
      "step": 1520
    },
    {
      "entropy": 6.883447170257568,
      "epoch": 1.7836257309941521,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004946587078766818,
      "loss": 6.7954,
      "mean_token_accuracy": 0.14844498932361602,
      "num_tokens": 3803887.0,
      "step": 1525
    },
    {
      "entropy": 6.835381889343262,
      "epoch": 1.7894736842105263,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004945567034543712,
      "loss": 6.9313,
      "mean_token_accuracy": 0.13484521508216857,
      "num_tokens": 3816534.0,
      "step": 1530
    },
    {
      "entropy": 6.8271290302276615,
      "epoch": 1.7953216374269005,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004944537461402234,
      "loss": 6.8922,
      "mean_token_accuracy": 0.14591248333454132,
      "num_tokens": 3827790.0,
      "step": 1535
    },
    {
      "entropy": 6.8720885753631595,
      "epoch": 1.801169590643275,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004943498363810807,
      "loss": 6.8323,
      "mean_token_accuracy": 0.1483864262700081,
      "num_tokens": 3839800.0,
      "step": 1540
    },
    {
      "entropy": 6.697543144226074,
      "epoch": 1.807017543859649,
      "grad_norm": 1.125,
      "learning_rate": 0.0004942449746279184,
      "loss": 6.7608,
      "mean_token_accuracy": 0.15420471429824828,
      "num_tokens": 3852621.0,
      "step": 1545
    },
    {
      "entropy": 6.82108359336853,
      "epoch": 1.8128654970760234,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004941391613358444,
      "loss": 6.8153,
      "mean_token_accuracy": 0.14005157276988028,
      "num_tokens": 3864184.0,
      "step": 1550
    },
    {
      "entropy": 6.835338401794433,
      "epoch": 1.8187134502923976,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000494032396964096,
      "loss": 6.7941,
      "mean_token_accuracy": 0.14192988201975823,
      "num_tokens": 3876750.0,
      "step": 1555
    },
    {
      "entropy": 6.804272985458374,
      "epoch": 1.8245614035087718,
      "grad_norm": 0.953125,
      "learning_rate": 0.000493924681976038,
      "loss": 6.8399,
      "mean_token_accuracy": 0.1471085824072361,
      "num_tokens": 3888788.0,
      "step": 1560
    },
    {
      "entropy": 6.753233766555786,
      "epoch": 1.8304093567251463,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004938160168391614,
      "loss": 6.8905,
      "mean_token_accuracy": 0.13889118731021882,
      "num_tokens": 3901816.0,
      "step": 1565
    },
    {
      "entropy": 6.806808280944824,
      "epoch": 1.8362573099415205,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004937064020250804,
      "loss": 6.7818,
      "mean_token_accuracy": 0.15054665878415108,
      "num_tokens": 3913127.0,
      "step": 1570
    },
    {
      "entropy": 6.792178058624268,
      "epoch": 1.8421052631578947,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004935958380095316,
      "loss": 6.7941,
      "mean_token_accuracy": 0.14823683351278305,
      "num_tokens": 3924878.0,
      "step": 1575
    },
    {
      "entropy": 6.834120273590088,
      "epoch": 1.8479532163742691,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004934843252723702,
      "loss": 6.8727,
      "mean_token_accuracy": 0.1455922856926918,
      "num_tokens": 3938334.0,
      "step": 1580
    },
    {
      "entropy": 6.78785252571106,
      "epoch": 1.8538011695906431,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00049337186429757,
      "loss": 6.8358,
      "mean_token_accuracy": 0.14139243289828302,
      "num_tokens": 3949733.0,
      "step": 1585
    },
    {
      "entropy": 6.91608624458313,
      "epoch": 1.8596491228070176,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004932584555732192,
      "loss": 6.8612,
      "mean_token_accuracy": 0.14494034871459008,
      "num_tokens": 3963109.0,
      "step": 1590
    },
    {
      "entropy": 6.812306976318359,
      "epoch": 1.8654970760233918,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004931440995915203,
      "loss": 6.8472,
      "mean_token_accuracy": 0.1428239457309246,
      "num_tokens": 3975477.0,
      "step": 1595
    },
    {
      "entropy": 6.742663955688476,
      "epoch": 1.871345029239766,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000493028796848786,
      "loss": 6.8052,
      "mean_token_accuracy": 0.15210191905498505,
      "num_tokens": 3987757.0,
      "step": 1600
    },
    {
      "entropy": 6.828237295150757,
      "epoch": 1.8771929824561404,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004929125478454389,
      "loss": 6.6838,
      "mean_token_accuracy": 0.14859880805015563,
      "num_tokens": 3998807.0,
      "step": 1605
    },
    {
      "entropy": 6.669670343399048,
      "epoch": 1.8830409356725146,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004927953530860077,
      "loss": 6.7556,
      "mean_token_accuracy": 0.14744624197483064,
      "num_tokens": 4010320.0,
      "step": 1610
    },
    {
      "entropy": 6.863097715377807,
      "epoch": 1.8888888888888888,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004926772130791261,
      "loss": 6.8995,
      "mean_token_accuracy": 0.13792884200811387,
      "num_tokens": 4023432.0,
      "step": 1615
    },
    {
      "entropy": 6.776876592636109,
      "epoch": 1.8947368421052633,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004925581283375303,
      "loss": 6.8713,
      "mean_token_accuracy": 0.13947884514927864,
      "num_tokens": 4036282.0,
      "step": 1620
    },
    {
      "entropy": 6.789672327041626,
      "epoch": 1.9005847953216373,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004924380993780566,
      "loss": 6.7491,
      "mean_token_accuracy": 0.15059926509857177,
      "num_tokens": 4048164.0,
      "step": 1625
    },
    {
      "entropy": 6.733688116073608,
      "epoch": 1.9064327485380117,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004923171267216392,
      "loss": 6.7895,
      "mean_token_accuracy": 0.14438801854848862,
      "num_tokens": 4059287.0,
      "step": 1630
    },
    {
      "entropy": 6.773806571960449,
      "epoch": 1.912280701754386,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004921952108933083,
      "loss": 6.7883,
      "mean_token_accuracy": 0.14180546924471854,
      "num_tokens": 4071976.0,
      "step": 1635
    },
    {
      "entropy": 6.726539087295532,
      "epoch": 1.9181286549707601,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004920723524221871,
      "loss": 6.8096,
      "mean_token_accuracy": 0.15039053112268447,
      "num_tokens": 4085116.0,
      "step": 1640
    },
    {
      "entropy": 6.895672273635864,
      "epoch": 1.9239766081871346,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004919485518414902,
      "loss": 6.8046,
      "mean_token_accuracy": 0.140822733938694,
      "num_tokens": 4096778.0,
      "step": 1645
    },
    {
      "entropy": 6.680618095397949,
      "epoch": 1.9298245614035088,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004918238096885213,
      "loss": 6.7161,
      "mean_token_accuracy": 0.14424919709563255,
      "num_tokens": 4109490.0,
      "step": 1650
    },
    {
      "entropy": 6.785288667678833,
      "epoch": 1.935672514619883,
      "grad_norm": 0.84375,
      "learning_rate": 0.00049169812650467,
      "loss": 6.7552,
      "mean_token_accuracy": 0.14924068748950958,
      "num_tokens": 4122832.0,
      "step": 1655
    },
    {
      "entropy": 6.841014194488525,
      "epoch": 1.9415204678362574,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004915715028354102,
      "loss": 6.783,
      "mean_token_accuracy": 0.14009461998939515,
      "num_tokens": 4134585.0,
      "step": 1660
    },
    {
      "entropy": 6.72946720123291,
      "epoch": 1.9473684210526314,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004914439392302981,
      "loss": 6.7209,
      "mean_token_accuracy": 0.14602409675717354,
      "num_tokens": 4145909.0,
      "step": 1665
    },
    {
      "entropy": 6.716705989837647,
      "epoch": 1.9532163742690059,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004913154362429686,
      "loss": 6.6342,
      "mean_token_accuracy": 0.15917537361383438,
      "num_tokens": 4157360.0,
      "step": 1670
    },
    {
      "entropy": 6.692678165435791,
      "epoch": 1.95906432748538,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004911859944311341,
      "loss": 6.8247,
      "mean_token_accuracy": 0.1407847747206688,
      "num_tokens": 4169789.0,
      "step": 1675
    },
    {
      "entropy": 6.856312084197998,
      "epoch": 1.9649122807017543,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004910556143565812,
      "loss": 6.8312,
      "mean_token_accuracy": 0.14292604625225067,
      "num_tokens": 4183131.0,
      "step": 1680
    },
    {
      "entropy": 6.738270616531372,
      "epoch": 1.9707602339181287,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004909242965851688,
      "loss": 6.7986,
      "mean_token_accuracy": 0.14529500007629395,
      "num_tokens": 4194865.0,
      "step": 1685
    },
    {
      "entropy": 6.79417314529419,
      "epoch": 1.976608187134503,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004907920416868253,
      "loss": 6.819,
      "mean_token_accuracy": 0.1439730167388916,
      "num_tokens": 4207406.0,
      "step": 1690
    },
    {
      "entropy": 6.730139255523682,
      "epoch": 1.9824561403508771,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004906588502355467,
      "loss": 6.7533,
      "mean_token_accuracy": 0.14599005728960038,
      "num_tokens": 4221623.0,
      "step": 1695
    },
    {
      "entropy": 6.724703598022461,
      "epoch": 1.9883040935672516,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004905247228093933,
      "loss": 6.6975,
      "mean_token_accuracy": 0.15941915661096573,
      "num_tokens": 4233699.0,
      "step": 1700
    },
    {
      "entropy": 6.703160762786865,
      "epoch": 1.9941520467836256,
      "grad_norm": 1.0,
      "learning_rate": 0.0004903896599904878,
      "loss": 6.7079,
      "mean_token_accuracy": 0.1484083577990532,
      "num_tokens": 4247669.0,
      "step": 1705
    },
    {
      "entropy": 6.775749015808105,
      "epoch": 2.0,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004902536623650125,
      "loss": 6.7558,
      "mean_token_accuracy": 0.1439180076122284,
      "num_tokens": 4258968.0,
      "step": 1710
    },
    {
      "entropy": 6.73444766998291,
      "epoch": 2.0058479532163744,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004901167305232068,
      "loss": 6.4691,
      "mean_token_accuracy": 0.15124359130859374,
      "num_tokens": 4270684.0,
      "step": 1715
    },
    {
      "entropy": 6.663851499557495,
      "epoch": 2.0116959064327484,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004899788650593648,
      "loss": 6.3956,
      "mean_token_accuracy": 0.1556681588292122,
      "num_tokens": 4282567.0,
      "step": 1720
    },
    {
      "entropy": 6.703156185150147,
      "epoch": 2.017543859649123,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004898400665718325,
      "loss": 6.4372,
      "mean_token_accuracy": 0.15561432689428328,
      "num_tokens": 4294803.0,
      "step": 1725
    },
    {
      "entropy": 6.656558609008789,
      "epoch": 2.023391812865497,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004897003356630052,
      "loss": 6.4204,
      "mean_token_accuracy": 0.15574801713228226,
      "num_tokens": 4305898.0,
      "step": 1730
    },
    {
      "entropy": 6.6269762992858885,
      "epoch": 2.0292397660818713,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004895596729393252,
      "loss": 6.3562,
      "mean_token_accuracy": 0.15742129534482957,
      "num_tokens": 4317985.0,
      "step": 1735
    },
    {
      "entropy": 6.54496283531189,
      "epoch": 2.0350877192982457,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004894180790112787,
      "loss": 6.4188,
      "mean_token_accuracy": 0.15664637386798858,
      "num_tokens": 4330482.0,
      "step": 1740
    },
    {
      "entropy": 6.694279479980469,
      "epoch": 2.0409356725146197,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004892755544933934,
      "loss": 6.4747,
      "mean_token_accuracy": 0.1461329534649849,
      "num_tokens": 4342782.0,
      "step": 1745
    },
    {
      "entropy": 6.626242208480835,
      "epoch": 2.046783625730994,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004891321000042361,
      "loss": 6.4792,
      "mean_token_accuracy": 0.1464425027370453,
      "num_tokens": 4356204.0,
      "step": 1750
    },
    {
      "entropy": 6.624731397628784,
      "epoch": 2.0526315789473686,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004889877161664096,
      "loss": 6.3913,
      "mean_token_accuracy": 0.1517579272389412,
      "num_tokens": 4370003.0,
      "step": 1755
    },
    {
      "entropy": 6.635585975646973,
      "epoch": 2.0584795321637426,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048884240360655,
      "loss": 6.45,
      "mean_token_accuracy": 0.15283022224903106,
      "num_tokens": 4381972.0,
      "step": 1760
    },
    {
      "entropy": 6.579027986526489,
      "epoch": 2.064327485380117,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004886961629553242,
      "loss": 6.3762,
      "mean_token_accuracy": 0.15750732719898225,
      "num_tokens": 4394033.0,
      "step": 1765
    },
    {
      "entropy": 6.599072217941284,
      "epoch": 2.0701754385964914,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004885489948474272,
      "loss": 6.4731,
      "mean_token_accuracy": 0.1535159945487976,
      "num_tokens": 4406799.0,
      "step": 1770
    },
    {
      "entropy": 6.641034030914307,
      "epoch": 2.0760233918128654,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004884008999215792,
      "loss": 6.4766,
      "mean_token_accuracy": 0.148863784968853,
      "num_tokens": 4420182.0,
      "step": 1775
    },
    {
      "entropy": 6.611565637588501,
      "epoch": 2.08187134502924,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004882518788205225,
      "loss": 6.4328,
      "mean_token_accuracy": 0.15162279605865478,
      "num_tokens": 4432014.0,
      "step": 1780
    },
    {
      "entropy": 6.6640540599823,
      "epoch": 2.087719298245614,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004881019321910198,
      "loss": 6.4501,
      "mean_token_accuracy": 0.15250293016433716,
      "num_tokens": 4444755.0,
      "step": 1785
    },
    {
      "entropy": 6.577320575714111,
      "epoch": 2.0935672514619883,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048795106068384985,
      "loss": 6.5076,
      "mean_token_accuracy": 0.1541823074221611,
      "num_tokens": 4456390.0,
      "step": 1790
    },
    {
      "entropy": 6.621748447418213,
      "epoch": 2.0994152046783627,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000487799264953806,
      "loss": 6.4261,
      "mean_token_accuracy": 0.15415703356266022,
      "num_tokens": 4468567.0,
      "step": 1795
    },
    {
      "entropy": 6.576422262191772,
      "epoch": 2.1052631578947367,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004876465456596926,
      "loss": 6.3937,
      "mean_token_accuracy": 0.16213200390338897,
      "num_tokens": 4479553.0,
      "step": 1800
    },
    {
      "entropy": 6.578837108612061,
      "epoch": 2.111111111111111,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004874929034643223,
      "loss": 6.4095,
      "mean_token_accuracy": 0.1652977555990219,
      "num_tokens": 4491505.0,
      "step": 1805
    },
    {
      "entropy": 6.571292304992676,
      "epoch": 2.116959064327485,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004873383390345132,
      "loss": 6.4702,
      "mean_token_accuracy": 0.15317193418741226,
      "num_tokens": 4504896.0,
      "step": 1810
    },
    {
      "entropy": 6.558498668670654,
      "epoch": 2.1228070175438596,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048718285304108613,
      "loss": 6.4574,
      "mean_token_accuracy": 0.1517552226781845,
      "num_tokens": 4516319.0,
      "step": 1815
    },
    {
      "entropy": 6.611506986618042,
      "epoch": 2.128654970760234,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048702644615886137,
      "loss": 6.4825,
      "mean_token_accuracy": 0.14925179556012153,
      "num_tokens": 4529132.0,
      "step": 1820
    },
    {
      "entropy": 6.613365411758423,
      "epoch": 2.134502923976608,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048686911906665597,
      "loss": 6.4493,
      "mean_token_accuracy": 0.14987133592367172,
      "num_tokens": 4541080.0,
      "step": 1825
    },
    {
      "entropy": 6.573991918563843,
      "epoch": 2.1403508771929824,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048671087244728093,
      "loss": 6.4614,
      "mean_token_accuracy": 0.15233668237924575,
      "num_tokens": 4553442.0,
      "step": 1830
    },
    {
      "entropy": 6.486274766921997,
      "epoch": 2.146198830409357,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048655170698753774,
      "loss": 6.3905,
      "mean_token_accuracy": 0.16419739574193953,
      "num_tokens": 4565617.0,
      "step": 1835
    },
    {
      "entropy": 6.626535081863404,
      "epoch": 2.152046783625731,
      "grad_norm": 1.09375,
      "learning_rate": 0.000486391623378216,
      "loss": 6.4744,
      "mean_token_accuracy": 0.14740555733442307,
      "num_tokens": 4577418.0,
      "step": 1840
    },
    {
      "entropy": 6.598245096206665,
      "epoch": 2.1578947368421053,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004862306223140901,
      "loss": 6.4,
      "mean_token_accuracy": 0.1555834747850895,
      "num_tokens": 4589763.0,
      "step": 1845
    },
    {
      "entropy": 6.601409673690796,
      "epoch": 2.1637426900584797,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048606870449391604,
      "loss": 6.5003,
      "mean_token_accuracy": 0.15487492382526397,
      "num_tokens": 4602522.0,
      "step": 1850
    },
    {
      "entropy": 6.6073479652404785,
      "epoch": 2.1695906432748537,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048590587062042907,
      "loss": 6.4661,
      "mean_token_accuracy": 0.15708144158124923,
      "num_tokens": 4614680.0,
      "step": 1855
    },
    {
      "entropy": 6.566403198242187,
      "epoch": 2.175438596491228,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048574212140033965,
      "loss": 6.4809,
      "mean_token_accuracy": 0.15187129229307175,
      "num_tokens": 4628166.0,
      "step": 1860
    },
    {
      "entropy": 6.573957204818726,
      "epoch": 2.181286549707602,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004855774575443311,
      "loss": 6.5018,
      "mean_token_accuracy": 0.15662527829408646,
      "num_tokens": 4640200.0,
      "step": 1865
    },
    {
      "entropy": 6.616477537155151,
      "epoch": 2.1871345029239766,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048541187976705645,
      "loss": 6.4426,
      "mean_token_accuracy": 0.15040336698293685,
      "num_tokens": 4652968.0,
      "step": 1870
    },
    {
      "entropy": 6.576403617858887,
      "epoch": 2.192982456140351,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048524538878713517,
      "loss": 6.4527,
      "mean_token_accuracy": 0.14977433681488037,
      "num_tokens": 4665741.0,
      "step": 1875
    },
    {
      "entropy": 6.52663288116455,
      "epoch": 2.198830409356725,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00048507798532715005,
      "loss": 6.4553,
      "mean_token_accuracy": 0.15197168439626693,
      "num_tokens": 4680074.0,
      "step": 1880
    },
    {
      "entropy": 6.625212478637695,
      "epoch": 2.2046783625730995,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048490967011364394,
      "loss": 6.4871,
      "mean_token_accuracy": 0.15389255732297896,
      "num_tokens": 4692340.0,
      "step": 1885
    },
    {
      "entropy": 6.529519033432007,
      "epoch": 2.2105263157894735,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004847404438771172,
      "loss": 6.4935,
      "mean_token_accuracy": 0.15022647008299828,
      "num_tokens": 4704470.0,
      "step": 1890
    },
    {
      "entropy": 6.648282909393311,
      "epoch": 2.216374269005848,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004845703073520236,
      "loss": 6.5072,
      "mean_token_accuracy": 0.14370284005999565,
      "num_tokens": 4716253.0,
      "step": 1895
    },
    {
      "entropy": 6.664510059356689,
      "epoch": 2.2222222222222223,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048439926127676796,
      "loss": 6.6323,
      "mean_token_accuracy": 0.14200105741620064,
      "num_tokens": 4730083.0,
      "step": 1900
    },
    {
      "entropy": 6.622454452514648,
      "epoch": 2.2280701754385963,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048422730639370256,
      "loss": 6.3976,
      "mean_token_accuracy": 0.1607450395822525,
      "num_tokens": 4742770.0,
      "step": 1905
    },
    {
      "entropy": 6.517550230026245,
      "epoch": 2.2339181286549707,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004840544434491238,
      "loss": 6.5229,
      "mean_token_accuracy": 0.15023039728403093,
      "num_tokens": 4754299.0,
      "step": 1910
    },
    {
      "entropy": 6.603322982788086,
      "epoch": 2.239766081871345,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048388067319326937,
      "loss": 6.477,
      "mean_token_accuracy": 0.14898111671209335,
      "num_tokens": 4766964.0,
      "step": 1915
    },
    {
      "entropy": 6.506717586517334,
      "epoch": 2.245614035087719,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048370599638031464,
      "loss": 6.5044,
      "mean_token_accuracy": 0.14812409728765488,
      "num_tokens": 4779921.0,
      "step": 1920
    },
    {
      "entropy": 6.654232549667358,
      "epoch": 2.2514619883040936,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048353041376836937,
      "loss": 6.5058,
      "mean_token_accuracy": 0.14764770194888116,
      "num_tokens": 4792233.0,
      "step": 1925
    },
    {
      "entropy": 6.529701566696167,
      "epoch": 2.257309941520468,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004833539261194749,
      "loss": 6.4803,
      "mean_token_accuracy": 0.15641626417636872,
      "num_tokens": 4803348.0,
      "step": 1930
    },
    {
      "entropy": 6.634726905822754,
      "epoch": 2.263157894736842,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048317653419960007,
      "loss": 6.5588,
      "mean_token_accuracy": 0.14596108347177505,
      "num_tokens": 4816574.0,
      "step": 1935
    },
    {
      "entropy": 6.476966142654419,
      "epoch": 2.2690058479532165,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048299823877863857,
      "loss": 6.4772,
      "mean_token_accuracy": 0.1571604609489441,
      "num_tokens": 4828008.0,
      "step": 1940
    },
    {
      "entropy": 6.661566209793091,
      "epoch": 2.2748538011695905,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004828190406304054,
      "loss": 6.5338,
      "mean_token_accuracy": 0.14808987975120544,
      "num_tokens": 4840621.0,
      "step": 1945
    },
    {
      "entropy": 6.566416645050049,
      "epoch": 2.280701754385965,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004826389405326334,
      "loss": 6.4336,
      "mean_token_accuracy": 0.1514001801609993,
      "num_tokens": 4853503.0,
      "step": 1950
    },
    {
      "entropy": 6.560876512527466,
      "epoch": 2.2865497076023393,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004824579392669698,
      "loss": 6.4614,
      "mean_token_accuracy": 0.1499699056148529,
      "num_tokens": 4865925.0,
      "step": 1955
    },
    {
      "entropy": 6.561740493774414,
      "epoch": 2.2923976608187133,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048227603761897316,
      "loss": 6.527,
      "mean_token_accuracy": 0.1484415277838707,
      "num_tokens": 4878704.0,
      "step": 1960
    },
    {
      "entropy": 6.646590280532837,
      "epoch": 2.2982456140350878,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004820932363781096,
      "loss": 6.4919,
      "mean_token_accuracy": 0.14958563446998596,
      "num_tokens": 4891750.0,
      "step": 1965
    },
    {
      "entropy": 6.548250484466553,
      "epoch": 2.3040935672514617,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004819095363377496,
      "loss": 6.4752,
      "mean_token_accuracy": 0.15743751972913742,
      "num_tokens": 4904539.0,
      "step": 1970
    },
    {
      "entropy": 6.508108806610108,
      "epoch": 2.309941520467836,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048172493829516465,
      "loss": 6.3696,
      "mean_token_accuracy": 0.15684462636709212,
      "num_tokens": 4916210.0,
      "step": 1975
    },
    {
      "entropy": 6.562199449539184,
      "epoch": 2.3157894736842106,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004815394430515234,
      "loss": 6.4354,
      "mean_token_accuracy": 0.15617517828941346,
      "num_tokens": 4929011.0,
      "step": 1980
    },
    {
      "entropy": 6.486643314361572,
      "epoch": 2.3216374269005846,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004813530514118884,
      "loss": 6.4113,
      "mean_token_accuracy": 0.15723690688610076,
      "num_tokens": 4942484.0,
      "step": 1985
    },
    {
      "entropy": 6.480406904220581,
      "epoch": 2.327485380116959,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004811657641852129,
      "loss": 6.4178,
      "mean_token_accuracy": 0.15218075662851333,
      "num_tokens": 4955395.0,
      "step": 1990
    },
    {
      "entropy": 6.5748659610748295,
      "epoch": 2.3333333333333335,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048097758218433687,
      "loss": 6.4423,
      "mean_token_accuracy": 0.15172057300806047,
      "num_tokens": 4966853.0,
      "step": 1995
    },
    {
      "entropy": 6.3933188915252686,
      "epoch": 2.3391812865497075,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048078850622598355,
      "loss": 6.3266,
      "mean_token_accuracy": 0.16453724801540376,
      "num_tokens": 4978514.0,
      "step": 2000
    },
    {
      "epoch": 2.3391812865497075,
      "eval_entropy": 6.446461832277191,
      "eval_loss": 6.858452796936035,
      "eval_mean_token_accuracy": 0.1379934970240018,
      "eval_num_tokens": 4978514.0,
      "eval_runtime": 2.2419,
      "eval_samples_per_second": 1207.444,
      "eval_steps_per_second": 151.209,
      "step": 2000
    }
  ],
  "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": 7178272881408000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}