{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.8368200836820083,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 6.802199983596802,
      "epoch": 0.0041841004184100415,
      "grad_norm": 1.4453125,
      "learning_rate": 2e-06,
      "loss": 6.4445,
      "mean_token_accuracy": 0.10309398248791694,
      "num_tokens": 9848.0,
      "step": 5
    },
    {
      "entropy": 6.861799955368042,
      "epoch": 0.008368200836820083,
      "grad_norm": 1.453125,
      "learning_rate": 4.5e-06,
      "loss": 6.4842,
      "mean_token_accuracy": 0.10960042849183083,
      "num_tokens": 18196.0,
      "step": 10
    },
    {
      "entropy": 6.8262049674987795,
      "epoch": 0.012552301255230125,
      "grad_norm": 1.2421875,
      "learning_rate": 7e-06,
      "loss": 6.4037,
      "mean_token_accuracy": 0.11181848868727684,
      "num_tokens": 26927.0,
      "step": 15
    },
    {
      "entropy": 6.852504396438599,
      "epoch": 0.016736401673640166,
      "grad_norm": 1.3828125,
      "learning_rate": 9.5e-06,
      "loss": 6.4858,
      "mean_token_accuracy": 0.10887380167841912,
      "num_tokens": 34987.0,
      "step": 20
    },
    {
      "entropy": 6.775377893447876,
      "epoch": 0.02092050209205021,
      "grad_norm": 1.0625,
      "learning_rate": 1.2e-05,
      "loss": 6.4787,
      "mean_token_accuracy": 0.10919283404946327,
      "num_tokens": 44936.0,
      "step": 25
    },
    {
      "entropy": 6.746171236038208,
      "epoch": 0.02510460251046025,
      "grad_norm": 1.25,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 6.5022,
      "mean_token_accuracy": 0.10517153143882751,
      "num_tokens": 54562.0,
      "step": 30
    },
    {
      "entropy": 6.737930631637573,
      "epoch": 0.029288702928870293,
      "grad_norm": 1.078125,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 6.5538,
      "mean_token_accuracy": 0.1072025939822197,
      "num_tokens": 63844.0,
      "step": 35
    },
    {
      "entropy": 6.756887722015381,
      "epoch": 0.03347280334728033,
      "grad_norm": 1.265625,
      "learning_rate": 1.95e-05,
      "loss": 6.543,
      "mean_token_accuracy": 0.11064185202121735,
      "num_tokens": 74331.0,
      "step": 40
    },
    {
      "entropy": 6.818012380599976,
      "epoch": 0.03765690376569038,
      "grad_norm": 1.375,
      "learning_rate": 2.2e-05,
      "loss": 6.5329,
      "mean_token_accuracy": 0.1078970804810524,
      "num_tokens": 83028.0,
      "step": 45
    },
    {
      "entropy": 6.826114511489868,
      "epoch": 0.04184100418410042,
      "grad_norm": 1.21875,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 6.4529,
      "mean_token_accuracy": 0.11463268399238587,
      "num_tokens": 91395.0,
      "step": 50
    },
    {
      "entropy": 6.710237836837768,
      "epoch": 0.04602510460251046,
      "grad_norm": 0.98046875,
      "learning_rate": 2.7e-05,
      "loss": 6.3486,
      "mean_token_accuracy": 0.11258858814835548,
      "num_tokens": 101868.0,
      "step": 55
    },
    {
      "entropy": 6.751042079925537,
      "epoch": 0.0502092050209205,
      "grad_norm": 1.1171875,
      "learning_rate": 2.95e-05,
      "loss": 6.4061,
      "mean_token_accuracy": 0.11649479269981385,
      "num_tokens": 110403.0,
      "step": 60
    },
    {
      "entropy": 6.72459921836853,
      "epoch": 0.05439330543933055,
      "grad_norm": 1.078125,
      "learning_rate": 3.2e-05,
      "loss": 6.4011,
      "mean_token_accuracy": 0.11154339239001274,
      "num_tokens": 119557.0,
      "step": 65
    },
    {
      "entropy": 6.770873069763184,
      "epoch": 0.058577405857740586,
      "grad_norm": 1.3125,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 6.4342,
      "mean_token_accuracy": 0.11136164590716362,
      "num_tokens": 128903.0,
      "step": 70
    },
    {
      "entropy": 6.825613069534302,
      "epoch": 0.06276150627615062,
      "grad_norm": 1.140625,
      "learning_rate": 3.7e-05,
      "loss": 6.4633,
      "mean_token_accuracy": 0.11030099019408227,
      "num_tokens": 137630.0,
      "step": 75
    },
    {
      "entropy": 6.702863502502441,
      "epoch": 0.06694560669456066,
      "grad_norm": 1.1953125,
      "learning_rate": 3.95e-05,
      "loss": 6.2417,
      "mean_token_accuracy": 0.11854962259531021,
      "num_tokens": 146335.0,
      "step": 80
    },
    {
      "entropy": 6.701785278320313,
      "epoch": 0.07112970711297072,
      "grad_norm": 1.0859375,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 6.3738,
      "mean_token_accuracy": 0.1200543224811554,
      "num_tokens": 155766.0,
      "step": 85
    },
    {
      "entropy": 6.6252336502075195,
      "epoch": 0.07531380753138076,
      "grad_norm": 1.0859375,
      "learning_rate": 4.45e-05,
      "loss": 6.42,
      "mean_token_accuracy": 0.11267679929733276,
      "num_tokens": 165844.0,
      "step": 90
    },
    {
      "entropy": 6.688562870025635,
      "epoch": 0.0794979079497908,
      "grad_norm": 1.359375,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 6.4734,
      "mean_token_accuracy": 0.10707445964217185,
      "num_tokens": 174858.0,
      "step": 95
    },
    {
      "entropy": 6.8517701625823975,
      "epoch": 0.08368200836820083,
      "grad_norm": 1.28125,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 6.4229,
      "mean_token_accuracy": 0.11569512039422988,
      "num_tokens": 184020.0,
      "step": 100
    },
    {
      "entropy": 6.7154582977294925,
      "epoch": 0.08786610878661087,
      "grad_norm": 1.203125,
      "learning_rate": 5.2e-05,
      "loss": 6.3312,
      "mean_token_accuracy": 0.11155860349535943,
      "num_tokens": 193714.0,
      "step": 105
    },
    {
      "entropy": 6.720305871963501,
      "epoch": 0.09205020920502092,
      "grad_norm": 1.15625,
      "learning_rate": 5.45e-05,
      "loss": 6.5139,
      "mean_token_accuracy": 0.1125582680106163,
      "num_tokens": 203139.0,
      "step": 110
    },
    {
      "entropy": 6.8431463718414305,
      "epoch": 0.09623430962343096,
      "grad_norm": 1.2265625,
      "learning_rate": 5.7e-05,
      "loss": 6.4414,
      "mean_token_accuracy": 0.11265396922826768,
      "num_tokens": 212446.0,
      "step": 115
    },
    {
      "entropy": 6.769783306121826,
      "epoch": 0.100418410041841,
      "grad_norm": 1.1953125,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 6.3793,
      "mean_token_accuracy": 0.12033891826868057,
      "num_tokens": 220910.0,
      "step": 120
    },
    {
      "entropy": 6.643674516677857,
      "epoch": 0.10460251046025104,
      "grad_norm": 1.1015625,
      "learning_rate": 6.2e-05,
      "loss": 6.4122,
      "mean_token_accuracy": 0.11757991164922714,
      "num_tokens": 230522.0,
      "step": 125
    },
    {
      "entropy": 6.830571556091309,
      "epoch": 0.1087866108786611,
      "grad_norm": 1.1640625,
      "learning_rate": 6.450000000000001e-05,
      "loss": 6.4109,
      "mean_token_accuracy": 0.11237624883651734,
      "num_tokens": 240092.0,
      "step": 130
    },
    {
      "entropy": 6.689486694335938,
      "epoch": 0.11297071129707113,
      "grad_norm": 1.078125,
      "learning_rate": 6.7e-05,
      "loss": 6.4836,
      "mean_token_accuracy": 0.11546132564544678,
      "num_tokens": 249478.0,
      "step": 135
    },
    {
      "entropy": 6.702459669113159,
      "epoch": 0.11715481171548117,
      "grad_norm": 1.3515625,
      "learning_rate": 6.950000000000001e-05,
      "loss": 6.3239,
      "mean_token_accuracy": 0.1283320315182209,
      "num_tokens": 257714.0,
      "step": 140
    },
    {
      "entropy": 6.72728910446167,
      "epoch": 0.12133891213389121,
      "grad_norm": 1.21875,
      "learning_rate": 7.2e-05,
      "loss": 6.3811,
      "mean_token_accuracy": 0.12412762865424157,
      "num_tokens": 266569.0,
      "step": 145
    },
    {
      "entropy": 6.742079639434815,
      "epoch": 0.12552301255230125,
      "grad_norm": 1.1484375,
      "learning_rate": 7.45e-05,
      "loss": 6.3738,
      "mean_token_accuracy": 0.11696210950613022,
      "num_tokens": 276688.0,
      "step": 150
    },
    {
      "entropy": 6.649025583267212,
      "epoch": 0.1297071129707113,
      "grad_norm": 1.1953125,
      "learning_rate": 7.7e-05,
      "loss": 6.256,
      "mean_token_accuracy": 0.1157757207751274,
      "num_tokens": 285430.0,
      "step": 155
    },
    {
      "entropy": 6.6111983299255375,
      "epoch": 0.13389121338912133,
      "grad_norm": 1.09375,
      "learning_rate": 7.950000000000001e-05,
      "loss": 6.4268,
      "mean_token_accuracy": 0.11469644829630851,
      "num_tokens": 295727.0,
      "step": 160
    },
    {
      "entropy": 6.73095850944519,
      "epoch": 0.13807531380753138,
      "grad_norm": 1.265625,
      "learning_rate": 8.2e-05,
      "loss": 6.3704,
      "mean_token_accuracy": 0.12031615525484085,
      "num_tokens": 304854.0,
      "step": 165
    },
    {
      "entropy": 6.7522295951843265,
      "epoch": 0.14225941422594143,
      "grad_norm": 1.3359375,
      "learning_rate": 8.450000000000001e-05,
      "loss": 6.4069,
      "mean_token_accuracy": 0.12297785431146621,
      "num_tokens": 313579.0,
      "step": 170
    },
    {
      "entropy": 6.770452976226807,
      "epoch": 0.14644351464435146,
      "grad_norm": 1.2734375,
      "learning_rate": 8.7e-05,
      "loss": 6.5024,
      "mean_token_accuracy": 0.11297214776277542,
      "num_tokens": 323310.0,
      "step": 175
    },
    {
      "entropy": 6.7838945388793945,
      "epoch": 0.1506276150627615,
      "grad_norm": 1.34375,
      "learning_rate": 8.95e-05,
      "loss": 6.368,
      "mean_token_accuracy": 0.1218890331685543,
      "num_tokens": 331720.0,
      "step": 180
    },
    {
      "entropy": 6.694746971130371,
      "epoch": 0.15481171548117154,
      "grad_norm": 1.3125,
      "learning_rate": 9.2e-05,
      "loss": 6.3376,
      "mean_token_accuracy": 0.12067973613739014,
      "num_tokens": 340331.0,
      "step": 185
    },
    {
      "entropy": 6.690021705627442,
      "epoch": 0.1589958158995816,
      "grad_norm": 1.21875,
      "learning_rate": 9.45e-05,
      "loss": 6.3738,
      "mean_token_accuracy": 0.11946730315685272,
      "num_tokens": 349677.0,
      "step": 190
    },
    {
      "entropy": 6.797479248046875,
      "epoch": 0.16317991631799164,
      "grad_norm": 1.1796875,
      "learning_rate": 9.7e-05,
      "loss": 6.3639,
      "mean_token_accuracy": 0.11213221102952957,
      "num_tokens": 358674.0,
      "step": 195
    },
    {
      "entropy": 6.5850146293640135,
      "epoch": 0.16736401673640167,
      "grad_norm": 1.234375,
      "learning_rate": 9.95e-05,
      "loss": 6.4279,
      "mean_token_accuracy": 0.11513617783784866,
      "num_tokens": 368258.0,
      "step": 200
    },
    {
      "entropy": 6.760379457473755,
      "epoch": 0.17154811715481172,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000102,
      "loss": 6.3138,
      "mean_token_accuracy": 0.12832038179039956,
      "num_tokens": 377083.0,
      "step": 205
    },
    {
      "entropy": 6.707645225524902,
      "epoch": 0.17573221757322174,
      "grad_norm": 1.28125,
      "learning_rate": 0.00010449999999999999,
      "loss": 6.3858,
      "mean_token_accuracy": 0.11404623836278915,
      "num_tokens": 385456.0,
      "step": 210
    },
    {
      "entropy": 6.636844539642334,
      "epoch": 0.1799163179916318,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000107,
      "loss": 6.3039,
      "mean_token_accuracy": 0.12129815816879272,
      "num_tokens": 395295.0,
      "step": 215
    },
    {
      "entropy": 6.709037256240845,
      "epoch": 0.18410041841004185,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001095,
      "loss": 6.3763,
      "mean_token_accuracy": 0.11955561861395836,
      "num_tokens": 405023.0,
      "step": 220
    },
    {
      "entropy": 6.674328136444092,
      "epoch": 0.18828451882845187,
      "grad_norm": 1.21875,
      "learning_rate": 0.000112,
      "loss": 6.4302,
      "mean_token_accuracy": 0.11591066569089889,
      "num_tokens": 414224.0,
      "step": 225
    },
    {
      "entropy": 6.727414226531982,
      "epoch": 0.19246861924686193,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001145,
      "loss": 6.2902,
      "mean_token_accuracy": 0.11443310081958771,
      "num_tokens": 423951.0,
      "step": 230
    },
    {
      "entropy": 6.675467681884766,
      "epoch": 0.19665271966527198,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011700000000000001,
      "loss": 6.3461,
      "mean_token_accuracy": 0.1151287503540516,
      "num_tokens": 432427.0,
      "step": 235
    },
    {
      "entropy": 6.5947874069213865,
      "epoch": 0.200836820083682,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011949999999999999,
      "loss": 6.307,
      "mean_token_accuracy": 0.12089343965053559,
      "num_tokens": 441560.0,
      "step": 240
    },
    {
      "entropy": 6.724042892456055,
      "epoch": 0.20502092050209206,
      "grad_norm": 1.34375,
      "learning_rate": 0.000122,
      "loss": 6.387,
      "mean_token_accuracy": 0.11083290576934815,
      "num_tokens": 450364.0,
      "step": 245
    },
    {
      "entropy": 6.667903852462769,
      "epoch": 0.20920502092050208,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001245,
      "loss": 6.3592,
      "mean_token_accuracy": 0.11416003108024597,
      "num_tokens": 460076.0,
      "step": 250
    },
    {
      "entropy": 6.653390169143677,
      "epoch": 0.21338912133891214,
      "grad_norm": 1.296875,
      "learning_rate": 0.000127,
      "loss": 6.2896,
      "mean_token_accuracy": 0.12183637991547584,
      "num_tokens": 468718.0,
      "step": 255
    },
    {
      "entropy": 6.727795171737671,
      "epoch": 0.2175732217573222,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001295,
      "loss": 6.39,
      "mean_token_accuracy": 0.11337619721889496,
      "num_tokens": 478436.0,
      "step": 260
    },
    {
      "entropy": 6.739923620223999,
      "epoch": 0.2217573221757322,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000132,
      "loss": 6.3767,
      "mean_token_accuracy": 0.11208422631025314,
      "num_tokens": 486833.0,
      "step": 265
    },
    {
      "entropy": 6.713146734237671,
      "epoch": 0.22594142259414227,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013450000000000002,
      "loss": 6.2926,
      "mean_token_accuracy": 0.12398196682333947,
      "num_tokens": 495119.0,
      "step": 270
    },
    {
      "entropy": 6.724122428894043,
      "epoch": 0.2301255230125523,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013700000000000002,
      "loss": 6.475,
      "mean_token_accuracy": 0.11590266525745392,
      "num_tokens": 505396.0,
      "step": 275
    },
    {
      "entropy": 6.622205924987793,
      "epoch": 0.23430962343096234,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001395,
      "loss": 6.3366,
      "mean_token_accuracy": 0.11645356938242912,
      "num_tokens": 515259.0,
      "step": 280
    },
    {
      "entropy": 6.795593929290772,
      "epoch": 0.2384937238493724,
      "grad_norm": 1.265625,
      "learning_rate": 0.00014199999999999998,
      "loss": 6.3565,
      "mean_token_accuracy": 0.1146524854004383,
      "num_tokens": 523608.0,
      "step": 285
    },
    {
      "entropy": 6.573353290557861,
      "epoch": 0.24267782426778242,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001445,
      "loss": 6.3411,
      "mean_token_accuracy": 0.11209621205925942,
      "num_tokens": 534613.0,
      "step": 290
    },
    {
      "entropy": 6.745756149291992,
      "epoch": 0.24686192468619247,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000147,
      "loss": 6.4111,
      "mean_token_accuracy": 0.119757891446352,
      "num_tokens": 543347.0,
      "step": 295
    },
    {
      "entropy": 6.671696519851684,
      "epoch": 0.2510460251046025,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001495,
      "loss": 6.3369,
      "mean_token_accuracy": 0.11111167520284652,
      "num_tokens": 552371.0,
      "step": 300
    },
    {
      "entropy": 6.657607221603394,
      "epoch": 0.25523012552301255,
      "grad_norm": 1.359375,
      "learning_rate": 0.000152,
      "loss": 6.3096,
      "mean_token_accuracy": 0.11953857839107514,
      "num_tokens": 560742.0,
      "step": 305
    },
    {
      "entropy": 6.732522964477539,
      "epoch": 0.2594142259414226,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015450000000000001,
      "loss": 6.4478,
      "mean_token_accuracy": 0.11639050245285035,
      "num_tokens": 569960.0,
      "step": 310
    },
    {
      "entropy": 6.726744174957275,
      "epoch": 0.26359832635983266,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000157,
      "loss": 6.3837,
      "mean_token_accuracy": 0.11821903139352799,
      "num_tokens": 578859.0,
      "step": 315
    },
    {
      "entropy": 6.730755567550659,
      "epoch": 0.26778242677824265,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001595,
      "loss": 6.3894,
      "mean_token_accuracy": 0.1184967190027237,
      "num_tokens": 588146.0,
      "step": 320
    },
    {
      "entropy": 6.5525407791137695,
      "epoch": 0.2719665271966527,
      "grad_norm": 1.390625,
      "learning_rate": 0.000162,
      "loss": 6.3298,
      "mean_token_accuracy": 0.11562250927090645,
      "num_tokens": 597268.0,
      "step": 325
    },
    {
      "entropy": 6.6291014671325685,
      "epoch": 0.27615062761506276,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.288,
      "mean_token_accuracy": 0.12023762464523316,
      "num_tokens": 606199.0,
      "step": 330
    },
    {
      "entropy": 6.786021327972412,
      "epoch": 0.2803347280334728,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.4049,
      "mean_token_accuracy": 0.10941854566335678,
      "num_tokens": 615966.0,
      "step": 335
    },
    {
      "entropy": 6.667136526107788,
      "epoch": 0.28451882845188287,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.3727,
      "mean_token_accuracy": 0.11674070581793786,
      "num_tokens": 625689.0,
      "step": 340
    },
    {
      "entropy": 6.698928308486939,
      "epoch": 0.28870292887029286,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.4012,
      "mean_token_accuracy": 0.11340371370315552,
      "num_tokens": 634877.0,
      "step": 345
    },
    {
      "entropy": 6.716418790817261,
      "epoch": 0.2928870292887029,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.3268,
      "mean_token_accuracy": 0.12364484667778015,
      "num_tokens": 644042.0,
      "step": 350
    },
    {
      "entropy": 6.608269596099854,
      "epoch": 0.29707112970711297,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000177,
      "loss": 6.3555,
      "mean_token_accuracy": 0.11423439830541611,
      "num_tokens": 653072.0,
      "step": 355
    },
    {
      "entropy": 6.664682340621948,
      "epoch": 0.301255230125523,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001795,
      "loss": 6.3614,
      "mean_token_accuracy": 0.11689013764262199,
      "num_tokens": 663686.0,
      "step": 360
    },
    {
      "entropy": 6.632741975784302,
      "epoch": 0.3054393305439331,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000182,
      "loss": 6.2951,
      "mean_token_accuracy": 0.11744664832949639,
      "num_tokens": 672293.0,
      "step": 365
    },
    {
      "entropy": 6.489195203781128,
      "epoch": 0.30962343096234307,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001845,
      "loss": 6.266,
      "mean_token_accuracy": 0.11952990591526032,
      "num_tokens": 681758.0,
      "step": 370
    },
    {
      "entropy": 6.6848828315734865,
      "epoch": 0.3138075313807531,
      "grad_norm": 1.3125,
      "learning_rate": 0.000187,
      "loss": 6.3038,
      "mean_token_accuracy": 0.11904869601130486,
      "num_tokens": 690727.0,
      "step": 375
    },
    {
      "entropy": 6.636202001571656,
      "epoch": 0.3179916317991632,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001895,
      "loss": 6.2574,
      "mean_token_accuracy": 0.11928709298372268,
      "num_tokens": 701210.0,
      "step": 380
    },
    {
      "entropy": 6.494536590576172,
      "epoch": 0.32217573221757323,
      "grad_norm": 1.15625,
      "learning_rate": 0.000192,
      "loss": 6.2645,
      "mean_token_accuracy": 0.12418703511357307,
      "num_tokens": 710869.0,
      "step": 385
    },
    {
      "entropy": 6.6857581615448,
      "epoch": 0.3263598326359833,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001945,
      "loss": 6.405,
      "mean_token_accuracy": 0.11158787459135056,
      "num_tokens": 721455.0,
      "step": 390
    },
    {
      "entropy": 6.52554292678833,
      "epoch": 0.3305439330543933,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.2772,
      "mean_token_accuracy": 0.1241541288793087,
      "num_tokens": 730917.0,
      "step": 395
    },
    {
      "entropy": 6.651937246322632,
      "epoch": 0.33472803347280333,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.3144,
      "mean_token_accuracy": 0.11836001500487328,
      "num_tokens": 739302.0,
      "step": 400
    },
    {
      "entropy": 6.575543546676636,
      "epoch": 0.3389121338912134,
      "grad_norm": 1.46875,
      "learning_rate": 0.000202,
      "loss": 6.2823,
      "mean_token_accuracy": 0.1181989461183548,
      "num_tokens": 748785.0,
      "step": 405
    },
    {
      "entropy": 6.55746021270752,
      "epoch": 0.34309623430962344,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.2125,
      "mean_token_accuracy": 0.12301687523722649,
      "num_tokens": 757215.0,
      "step": 410
    },
    {
      "entropy": 6.495482349395752,
      "epoch": 0.3472803347280335,
      "grad_norm": 1.34375,
      "learning_rate": 0.000207,
      "loss": 6.158,
      "mean_token_accuracy": 0.12710630893707275,
      "num_tokens": 765682.0,
      "step": 415
    },
    {
      "entropy": 6.671501731872558,
      "epoch": 0.3514644351464435,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002095,
      "loss": 6.3598,
      "mean_token_accuracy": 0.1148804098367691,
      "num_tokens": 774814.0,
      "step": 420
    },
    {
      "entropy": 6.54171838760376,
      "epoch": 0.35564853556485354,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000212,
      "loss": 6.2819,
      "mean_token_accuracy": 0.1194837212562561,
      "num_tokens": 784019.0,
      "step": 425
    },
    {
      "entropy": 6.477924489974976,
      "epoch": 0.3598326359832636,
      "grad_norm": 1.25,
      "learning_rate": 0.0002145,
      "loss": 6.1587,
      "mean_token_accuracy": 0.12297921627759933,
      "num_tokens": 793237.0,
      "step": 430
    },
    {
      "entropy": 6.627668190002441,
      "epoch": 0.36401673640167365,
      "grad_norm": 1.375,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.2292,
      "mean_token_accuracy": 0.11881406381726264,
      "num_tokens": 801548.0,
      "step": 435
    },
    {
      "entropy": 6.450384140014648,
      "epoch": 0.3682008368200837,
      "grad_norm": 1.25,
      "learning_rate": 0.0002195,
      "loss": 6.2893,
      "mean_token_accuracy": 0.12606984302401542,
      "num_tokens": 810974.0,
      "step": 440
    },
    {
      "entropy": 6.513573360443115,
      "epoch": 0.3723849372384937,
      "grad_norm": 1.46875,
      "learning_rate": 0.000222,
      "loss": 6.1682,
      "mean_token_accuracy": 0.12752986550331116,
      "num_tokens": 820023.0,
      "step": 445
    },
    {
      "entropy": 6.611561870574951,
      "epoch": 0.37656903765690375,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002245,
      "loss": 6.2589,
      "mean_token_accuracy": 0.12134803086519241,
      "num_tokens": 828297.0,
      "step": 450
    },
    {
      "entropy": 6.553285312652588,
      "epoch": 0.3807531380753138,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.3043,
      "mean_token_accuracy": 0.1155056394636631,
      "num_tokens": 838268.0,
      "step": 455
    },
    {
      "entropy": 6.6202904224395756,
      "epoch": 0.38493723849372385,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.2264,
      "mean_token_accuracy": 0.1235147014260292,
      "num_tokens": 846491.0,
      "step": 460
    },
    {
      "entropy": 6.374619865417481,
      "epoch": 0.3891213389121339,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.1882,
      "mean_token_accuracy": 0.12715737968683244,
      "num_tokens": 856501.0,
      "step": 465
    },
    {
      "entropy": 6.529414176940918,
      "epoch": 0.39330543933054396,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.2449,
      "mean_token_accuracy": 0.12165890857577324,
      "num_tokens": 867054.0,
      "step": 470
    },
    {
      "entropy": 6.44644775390625,
      "epoch": 0.39748953974895396,
      "grad_norm": 1.375,
      "learning_rate": 0.000237,
      "loss": 6.271,
      "mean_token_accuracy": 0.12292861416935921,
      "num_tokens": 875956.0,
      "step": 475
    },
    {
      "entropy": 6.492046117782593,
      "epoch": 0.401673640167364,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002395,
      "loss": 6.1897,
      "mean_token_accuracy": 0.12475509718060493,
      "num_tokens": 884427.0,
      "step": 480
    },
    {
      "entropy": 6.615246963500977,
      "epoch": 0.40585774058577406,
      "grad_norm": 1.171875,
      "learning_rate": 0.000242,
      "loss": 6.2791,
      "mean_token_accuracy": 0.12459831088781356,
      "num_tokens": 894181.0,
      "step": 485
    },
    {
      "entropy": 6.427358102798462,
      "epoch": 0.4100418410041841,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002445,
      "loss": 6.1205,
      "mean_token_accuracy": 0.12992214635014535,
      "num_tokens": 904076.0,
      "step": 490
    },
    {
      "entropy": 6.470092678070069,
      "epoch": 0.41422594142259417,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000247,
      "loss": 6.2517,
      "mean_token_accuracy": 0.12282974123954774,
      "num_tokens": 913915.0,
      "step": 495
    },
    {
      "entropy": 6.594865226745606,
      "epoch": 0.41841004184100417,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002495,
      "loss": 6.3694,
      "mean_token_accuracy": 0.11399232223629951,
      "num_tokens": 922876.0,
      "step": 500
    },
    {
      "epoch": 0.41841004184100417,
      "eval_entropy": 6.477647836954316,
      "eval_loss": 6.225774765014648,
      "eval_mean_token_accuracy": 0.1297860539809692,
      "eval_num_tokens": 922876.0,
      "eval_runtime": 2.8824,
      "eval_samples_per_second": 1355.489,
      "eval_steps_per_second": 169.653,
      "step": 500
    },
    {
      "entropy": 6.409206914901733,
      "epoch": 0.4225941422594142,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000252,
      "loss": 6.1393,
      "mean_token_accuracy": 0.1315084345638752,
      "num_tokens": 931998.0,
      "step": 505
    },
    {
      "entropy": 6.55022029876709,
      "epoch": 0.42677824267782427,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002545,
      "loss": 6.214,
      "mean_token_accuracy": 0.12237880975008011,
      "num_tokens": 941387.0,
      "step": 510
    },
    {
      "entropy": 6.381158638000488,
      "epoch": 0.4309623430962343,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000257,
      "loss": 6.2484,
      "mean_token_accuracy": 0.11762547865509987,
      "num_tokens": 951866.0,
      "step": 515
    },
    {
      "entropy": 6.425309085845948,
      "epoch": 0.4351464435146444,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002595,
      "loss": 6.2003,
      "mean_token_accuracy": 0.12336372956633568,
      "num_tokens": 961252.0,
      "step": 520
    },
    {
      "entropy": 6.572003746032715,
      "epoch": 0.4393305439330544,
      "grad_norm": 1.53125,
      "learning_rate": 0.000262,
      "loss": 6.2033,
      "mean_token_accuracy": 0.1296585388481617,
      "num_tokens": 969857.0,
      "step": 525
    },
    {
      "entropy": 6.4825043201446535,
      "epoch": 0.4435146443514644,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.1998,
      "mean_token_accuracy": 0.12262558117508889,
      "num_tokens": 978754.0,
      "step": 530
    },
    {
      "entropy": 6.42857813835144,
      "epoch": 0.4476987447698745,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.1703,
      "mean_token_accuracy": 0.11900547742843628,
      "num_tokens": 987935.0,
      "step": 535
    },
    {
      "entropy": 6.492337369918824,
      "epoch": 0.45188284518828453,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.1494,
      "mean_token_accuracy": 0.12749148234725,
      "num_tokens": 996873.0,
      "step": 540
    },
    {
      "entropy": 6.484167146682739,
      "epoch": 0.4560669456066946,
      "grad_norm": 1.25,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.2565,
      "mean_token_accuracy": 0.12450389862060547,
      "num_tokens": 1006487.0,
      "step": 545
    },
    {
      "entropy": 6.306475353240967,
      "epoch": 0.4602510460251046,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002745,
      "loss": 6.0848,
      "mean_token_accuracy": 0.1302729830145836,
      "num_tokens": 1015715.0,
      "step": 550
    },
    {
      "entropy": 6.4527686595916744,
      "epoch": 0.46443514644351463,
      "grad_norm": 1.390625,
      "learning_rate": 0.000277,
      "loss": 6.2025,
      "mean_token_accuracy": 0.12450708001852036,
      "num_tokens": 1025091.0,
      "step": 555
    },
    {
      "entropy": 6.397480440139771,
      "epoch": 0.4686192468619247,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002795,
      "loss": 6.1394,
      "mean_token_accuracy": 0.12280101329088211,
      "num_tokens": 1034174.0,
      "step": 560
    },
    {
      "entropy": 6.512916564941406,
      "epoch": 0.47280334728033474,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.3333,
      "mean_token_accuracy": 0.11398612335324287,
      "num_tokens": 1043777.0,
      "step": 565
    },
    {
      "entropy": 6.476423072814941,
      "epoch": 0.4769874476987448,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002845,
      "loss": 6.1838,
      "mean_token_accuracy": 0.12991034761071205,
      "num_tokens": 1052749.0,
      "step": 570
    },
    {
      "entropy": 6.485992193222046,
      "epoch": 0.4811715481171548,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000287,
      "loss": 6.207,
      "mean_token_accuracy": 0.12597287371754645,
      "num_tokens": 1061564.0,
      "step": 575
    },
    {
      "entropy": 6.389427852630615,
      "epoch": 0.48535564853556484,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002895,
      "loss": 6.1204,
      "mean_token_accuracy": 0.13354995474219322,
      "num_tokens": 1070006.0,
      "step": 580
    },
    {
      "entropy": 6.363674736022949,
      "epoch": 0.4895397489539749,
      "grad_norm": 1.1875,
      "learning_rate": 0.000292,
      "loss": 6.1847,
      "mean_token_accuracy": 0.12432906851172447,
      "num_tokens": 1079799.0,
      "step": 585
    },
    {
      "entropy": 6.432363891601563,
      "epoch": 0.49372384937238495,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002945,
      "loss": 6.0386,
      "mean_token_accuracy": 0.13422225415706635,
      "num_tokens": 1088842.0,
      "step": 590
    },
    {
      "entropy": 6.306155967712402,
      "epoch": 0.497907949790795,
      "grad_norm": 1.375,
      "learning_rate": 0.000297,
      "loss": 6.1119,
      "mean_token_accuracy": 0.13670141398906707,
      "num_tokens": 1097960.0,
      "step": 595
    },
    {
      "entropy": 6.31667046546936,
      "epoch": 0.502092050209205,
      "grad_norm": 1.375,
      "learning_rate": 0.0002995,
      "loss": 6.0895,
      "mean_token_accuracy": 0.12390791401267051,
      "num_tokens": 1106797.0,
      "step": 600
    },
    {
      "entropy": 6.38304967880249,
      "epoch": 0.5062761506276151,
      "grad_norm": 1.3125,
      "learning_rate": 0.000302,
      "loss": 6.2594,
      "mean_token_accuracy": 0.13380709663033485,
      "num_tokens": 1116384.0,
      "step": 605
    },
    {
      "entropy": 6.254551887512207,
      "epoch": 0.5104602510460251,
      "grad_norm": 1.375,
      "learning_rate": 0.0003045,
      "loss": 6.012,
      "mean_token_accuracy": 0.13452231362462044,
      "num_tokens": 1125206.0,
      "step": 610
    },
    {
      "entropy": 6.527569246292114,
      "epoch": 0.5146443514644351,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000307,
      "loss": 6.2528,
      "mean_token_accuracy": 0.12866818457841872,
      "num_tokens": 1134110.0,
      "step": 615
    },
    {
      "entropy": 6.497180366516114,
      "epoch": 0.5188284518828452,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003095,
      "loss": 6.2468,
      "mean_token_accuracy": 0.12335871756076813,
      "num_tokens": 1142833.0,
      "step": 620
    },
    {
      "entropy": 6.333167219161988,
      "epoch": 0.5230125523012552,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000312,
      "loss": 6.1645,
      "mean_token_accuracy": 0.13038308247923852,
      "num_tokens": 1152006.0,
      "step": 625
    },
    {
      "entropy": 6.3656152248382565,
      "epoch": 0.5271966527196653,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003145,
      "loss": 6.0956,
      "mean_token_accuracy": 0.13017264977097512,
      "num_tokens": 1161061.0,
      "step": 630
    },
    {
      "entropy": 6.417209196090698,
      "epoch": 0.5313807531380753,
      "grad_norm": 1.1875,
      "learning_rate": 0.000317,
      "loss": 6.1234,
      "mean_token_accuracy": 0.1287168890237808,
      "num_tokens": 1170754.0,
      "step": 635
    },
    {
      "entropy": 6.204028367996216,
      "epoch": 0.5355648535564853,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003195,
      "loss": 6.0824,
      "mean_token_accuracy": 0.1296873278915882,
      "num_tokens": 1180854.0,
      "step": 640
    },
    {
      "entropy": 6.387170028686524,
      "epoch": 0.5397489539748954,
      "grad_norm": 1.265625,
      "learning_rate": 0.000322,
      "loss": 6.1625,
      "mean_token_accuracy": 0.12682752832770347,
      "num_tokens": 1190567.0,
      "step": 645
    },
    {
      "entropy": 6.313226795196533,
      "epoch": 0.5439330543933054,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.0333,
      "mean_token_accuracy": 0.13330961391329765,
      "num_tokens": 1200247.0,
      "step": 650
    },
    {
      "entropy": 6.270791339874267,
      "epoch": 0.5481171548117155,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.1335,
      "mean_token_accuracy": 0.12965440675616263,
      "num_tokens": 1209358.0,
      "step": 655
    },
    {
      "entropy": 6.285220718383789,
      "epoch": 0.5523012552301255,
      "grad_norm": 1.25,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.1296,
      "mean_token_accuracy": 0.13011690378189086,
      "num_tokens": 1217746.0,
      "step": 660
    },
    {
      "entropy": 6.305118894577026,
      "epoch": 0.5564853556485355,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.1036,
      "mean_token_accuracy": 0.13556399419903756,
      "num_tokens": 1226829.0,
      "step": 665
    },
    {
      "entropy": 6.277094221115112,
      "epoch": 0.5606694560669456,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.0266,
      "mean_token_accuracy": 0.1367674969136715,
      "num_tokens": 1236642.0,
      "step": 670
    },
    {
      "entropy": 6.425677394866943,
      "epoch": 0.5648535564853556,
      "grad_norm": 1.5,
      "learning_rate": 0.000337,
      "loss": 6.1464,
      "mean_token_accuracy": 0.13671540543437005,
      "num_tokens": 1244934.0,
      "step": 675
    },
    {
      "entropy": 6.204587936401367,
      "epoch": 0.5690376569037657,
      "grad_norm": 1.375,
      "learning_rate": 0.0003395,
      "loss": 6.0569,
      "mean_token_accuracy": 0.1287803091108799,
      "num_tokens": 1254786.0,
      "step": 680
    },
    {
      "entropy": 6.193703556060791,
      "epoch": 0.5732217573221757,
      "grad_norm": 1.21875,
      "learning_rate": 0.000342,
      "loss": 6.0179,
      "mean_token_accuracy": 0.13714393973350525,
      "num_tokens": 1263745.0,
      "step": 685
    },
    {
      "entropy": 6.277493715286255,
      "epoch": 0.5774058577405857,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.1458,
      "mean_token_accuracy": 0.13557302355766296,
      "num_tokens": 1273101.0,
      "step": 690
    },
    {
      "entropy": 6.420682430267334,
      "epoch": 0.5815899581589958,
      "grad_norm": 1.46875,
      "learning_rate": 0.000347,
      "loss": 6.1451,
      "mean_token_accuracy": 0.12692318111658096,
      "num_tokens": 1282547.0,
      "step": 695
    },
    {
      "entropy": 6.242686367034912,
      "epoch": 0.5857740585774058,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003495,
      "loss": 6.1568,
      "mean_token_accuracy": 0.1301800861954689,
      "num_tokens": 1291594.0,
      "step": 700
    },
    {
      "entropy": 6.311301374435425,
      "epoch": 0.5899581589958159,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000352,
      "loss": 6.0677,
      "mean_token_accuracy": 0.13865734487771988,
      "num_tokens": 1300282.0,
      "step": 705
    },
    {
      "entropy": 6.283356428146362,
      "epoch": 0.5941422594142259,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003545,
      "loss": 6.1457,
      "mean_token_accuracy": 0.13872819766402245,
      "num_tokens": 1309304.0,
      "step": 710
    },
    {
      "entropy": 6.22599310874939,
      "epoch": 0.5983263598326359,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000357,
      "loss": 5.9905,
      "mean_token_accuracy": 0.13430829271674155,
      "num_tokens": 1317638.0,
      "step": 715
    },
    {
      "entropy": 6.374803638458252,
      "epoch": 0.602510460251046,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003595,
      "loss": 6.1976,
      "mean_token_accuracy": 0.12840964049100875,
      "num_tokens": 1327242.0,
      "step": 720
    },
    {
      "entropy": 6.2666154384613035,
      "epoch": 0.606694560669456,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000362,
      "loss": 6.014,
      "mean_token_accuracy": 0.13483576700091363,
      "num_tokens": 1335880.0,
      "step": 725
    },
    {
      "entropy": 6.198787117004395,
      "epoch": 0.6108786610878661,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003645,
      "loss": 6.0882,
      "mean_token_accuracy": 0.13066002652049064,
      "num_tokens": 1344909.0,
      "step": 730
    },
    {
      "entropy": 6.220889329910278,
      "epoch": 0.6150627615062761,
      "grad_norm": 1.25,
      "learning_rate": 0.000367,
      "loss": 5.9637,
      "mean_token_accuracy": 0.1385305680334568,
      "num_tokens": 1353997.0,
      "step": 735
    },
    {
      "entropy": 6.250986480712891,
      "epoch": 0.6192468619246861,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003695,
      "loss": 6.1325,
      "mean_token_accuracy": 0.13334286585450172,
      "num_tokens": 1363293.0,
      "step": 740
    },
    {
      "entropy": 6.397185230255127,
      "epoch": 0.6234309623430963,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000372,
      "loss": 6.1387,
      "mean_token_accuracy": 0.12498829737305642,
      "num_tokens": 1372174.0,
      "step": 745
    },
    {
      "entropy": 6.285424041748047,
      "epoch": 0.6276150627615062,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003745,
      "loss": 6.0244,
      "mean_token_accuracy": 0.13294555693864823,
      "num_tokens": 1381067.0,
      "step": 750
    },
    {
      "entropy": 6.165482044219971,
      "epoch": 0.6317991631799164,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000377,
      "loss": 6.0349,
      "mean_token_accuracy": 0.13717807978391647,
      "num_tokens": 1390072.0,
      "step": 755
    },
    {
      "entropy": 6.254514551162719,
      "epoch": 0.6359832635983264,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003795,
      "loss": 6.017,
      "mean_token_accuracy": 0.1379196561872959,
      "num_tokens": 1399391.0,
      "step": 760
    },
    {
      "entropy": 6.182499074935913,
      "epoch": 0.6401673640167364,
      "grad_norm": 1.28125,
      "learning_rate": 0.000382,
      "loss": 6.0082,
      "mean_token_accuracy": 0.1376515731215477,
      "num_tokens": 1408054.0,
      "step": 765
    },
    {
      "entropy": 6.243570137023926,
      "epoch": 0.6443514644351465,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003845,
      "loss": 6.0371,
      "mean_token_accuracy": 0.12797121033072473,
      "num_tokens": 1417406.0,
      "step": 770
    },
    {
      "entropy": 6.195934247970581,
      "epoch": 0.6485355648535565,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.1035,
      "mean_token_accuracy": 0.12684772536158562,
      "num_tokens": 1426907.0,
      "step": 775
    },
    {
      "entropy": 6.2838667869567875,
      "epoch": 0.6527196652719666,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.1754,
      "mean_token_accuracy": 0.12340578138828277,
      "num_tokens": 1436085.0,
      "step": 780
    },
    {
      "entropy": 6.26796145439148,
      "epoch": 0.6569037656903766,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.0634,
      "mean_token_accuracy": 0.1289584718644619,
      "num_tokens": 1444823.0,
      "step": 785
    },
    {
      "entropy": 6.289790201187134,
      "epoch": 0.6610878661087866,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.0986,
      "mean_token_accuracy": 0.13113027885556222,
      "num_tokens": 1454621.0,
      "step": 790
    },
    {
      "entropy": 6.109712553024292,
      "epoch": 0.6652719665271967,
      "grad_norm": 1.328125,
      "learning_rate": 0.00039700000000000005,
      "loss": 5.9769,
      "mean_token_accuracy": 0.1314921833574772,
      "num_tokens": 1463784.0,
      "step": 795
    },
    {
      "entropy": 6.236892223358154,
      "epoch": 0.6694560669456067,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003995,
      "loss": 6.0932,
      "mean_token_accuracy": 0.13157428726553916,
      "num_tokens": 1473257.0,
      "step": 800
    },
    {
      "entropy": 6.230193567276001,
      "epoch": 0.6736401673640168,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000402,
      "loss": 6.0429,
      "mean_token_accuracy": 0.13767861425876618,
      "num_tokens": 1482195.0,
      "step": 805
    },
    {
      "entropy": 6.138327884674072,
      "epoch": 0.6778242677824268,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004045,
      "loss": 6.0855,
      "mean_token_accuracy": 0.1348243311047554,
      "num_tokens": 1491590.0,
      "step": 810
    },
    {
      "entropy": 6.219961166381836,
      "epoch": 0.6820083682008368,
      "grad_norm": 1.5,
      "learning_rate": 0.00040699999999999997,
      "loss": 5.9732,
      "mean_token_accuracy": 0.14189569428563117,
      "num_tokens": 1500208.0,
      "step": 815
    },
    {
      "entropy": 6.153802299499512,
      "epoch": 0.6861924686192469,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004095,
      "loss": 6.1078,
      "mean_token_accuracy": 0.13705730587244033,
      "num_tokens": 1509412.0,
      "step": 820
    },
    {
      "entropy": 6.214185428619385,
      "epoch": 0.6903765690376569,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000412,
      "loss": 6.1048,
      "mean_token_accuracy": 0.13078184500336648,
      "num_tokens": 1518426.0,
      "step": 825
    },
    {
      "entropy": 6.275689888000488,
      "epoch": 0.694560669456067,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004145,
      "loss": 6.0391,
      "mean_token_accuracy": 0.1352381370961666,
      "num_tokens": 1528605.0,
      "step": 830
    },
    {
      "entropy": 6.123213243484497,
      "epoch": 0.698744769874477,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000417,
      "loss": 6.0004,
      "mean_token_accuracy": 0.14104130268096923,
      "num_tokens": 1537449.0,
      "step": 835
    },
    {
      "entropy": 6.1678873062133786,
      "epoch": 0.702928870292887,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004195,
      "loss": 6.0386,
      "mean_token_accuracy": 0.13681078255176543,
      "num_tokens": 1546743.0,
      "step": 840
    },
    {
      "entropy": 6.144597291946411,
      "epoch": 0.7071129707112971,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000422,
      "loss": 6.0213,
      "mean_token_accuracy": 0.13631223142147064,
      "num_tokens": 1555691.0,
      "step": 845
    },
    {
      "entropy": 6.11121015548706,
      "epoch": 0.7112970711297071,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004245,
      "loss": 5.9676,
      "mean_token_accuracy": 0.14060806781053542,
      "num_tokens": 1564957.0,
      "step": 850
    },
    {
      "entropy": 6.265472888946533,
      "epoch": 0.7154811715481172,
      "grad_norm": 1.46875,
      "learning_rate": 0.000427,
      "loss": 6.0478,
      "mean_token_accuracy": 0.13608204945921898,
      "num_tokens": 1573901.0,
      "step": 855
    },
    {
      "entropy": 6.0549863338470455,
      "epoch": 0.7196652719665272,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004295,
      "loss": 6.0369,
      "mean_token_accuracy": 0.13866082727909088,
      "num_tokens": 1582644.0,
      "step": 860
    },
    {
      "entropy": 6.237826824188232,
      "epoch": 0.7238493723849372,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000432,
      "loss": 6.1671,
      "mean_token_accuracy": 0.12615143060684203,
      "num_tokens": 1592363.0,
      "step": 865
    },
    {
      "entropy": 6.219017791748047,
      "epoch": 0.7280334728033473,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004345,
      "loss": 6.1152,
      "mean_token_accuracy": 0.131891880184412,
      "num_tokens": 1602012.0,
      "step": 870
    },
    {
      "entropy": 6.147130346298217,
      "epoch": 0.7322175732217573,
      "grad_norm": 1.25,
      "learning_rate": 0.000437,
      "loss": 6.0066,
      "mean_token_accuracy": 0.1359176442027092,
      "num_tokens": 1611625.0,
      "step": 875
    },
    {
      "entropy": 6.027130603790283,
      "epoch": 0.7364016736401674,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004395,
      "loss": 5.96,
      "mean_token_accuracy": 0.13448768407106398,
      "num_tokens": 1620941.0,
      "step": 880
    },
    {
      "entropy": 6.1633179664611815,
      "epoch": 0.7405857740585774,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000442,
      "loss": 6.0589,
      "mean_token_accuracy": 0.13061976209282875,
      "num_tokens": 1629225.0,
      "step": 885
    },
    {
      "entropy": 6.152882194519043,
      "epoch": 0.7447698744769874,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004445,
      "loss": 6.0312,
      "mean_token_accuracy": 0.12992663234472274,
      "num_tokens": 1639551.0,
      "step": 890
    },
    {
      "entropy": 6.138622665405274,
      "epoch": 0.7489539748953975,
      "grad_norm": 1.140625,
      "learning_rate": 0.000447,
      "loss": 6.02,
      "mean_token_accuracy": 0.12998297885060311,
      "num_tokens": 1648963.0,
      "step": 895
    },
    {
      "entropy": 6.0763167381286625,
      "epoch": 0.7531380753138075,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044950000000000003,
      "loss": 5.9294,
      "mean_token_accuracy": 0.14245674312114714,
      "num_tokens": 1657735.0,
      "step": 900
    },
    {
      "entropy": 6.254350471496582,
      "epoch": 0.7573221757322176,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045200000000000004,
      "loss": 5.9799,
      "mean_token_accuracy": 0.13656511679291725,
      "num_tokens": 1666349.0,
      "step": 905
    },
    {
      "entropy": 5.909399604797363,
      "epoch": 0.7615062761506276,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045450000000000004,
      "loss": 5.9225,
      "mean_token_accuracy": 0.13355625793337822,
      "num_tokens": 1675920.0,
      "step": 910
    },
    {
      "entropy": 6.054324007034301,
      "epoch": 0.7656903765690377,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045700000000000005,
      "loss": 5.9577,
      "mean_token_accuracy": 0.13852019906044005,
      "num_tokens": 1685626.0,
      "step": 915
    },
    {
      "entropy": 6.240289211273193,
      "epoch": 0.7698744769874477,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.0816,
      "mean_token_accuracy": 0.14210813418030738,
      "num_tokens": 1694972.0,
      "step": 920
    },
    {
      "entropy": 6.101596355438232,
      "epoch": 0.7740585774058577,
      "grad_norm": 1.203125,
      "learning_rate": 0.000462,
      "loss": 5.9615,
      "mean_token_accuracy": 0.14294997751712799,
      "num_tokens": 1703806.0,
      "step": 925
    },
    {
      "entropy": 6.082038736343383,
      "epoch": 0.7782426778242678,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004645,
      "loss": 6.0631,
      "mean_token_accuracy": 0.1384358860552311,
      "num_tokens": 1712949.0,
      "step": 930
    },
    {
      "entropy": 6.252063655853272,
      "epoch": 0.7824267782426778,
      "grad_norm": 1.203125,
      "learning_rate": 0.000467,
      "loss": 6.1782,
      "mean_token_accuracy": 0.134084989130497,
      "num_tokens": 1722328.0,
      "step": 935
    },
    {
      "entropy": 6.226525402069091,
      "epoch": 0.7866108786610879,
      "grad_norm": 1.25,
      "learning_rate": 0.0004695,
      "loss": 6.1574,
      "mean_token_accuracy": 0.1339714154601097,
      "num_tokens": 1731880.0,
      "step": 940
    },
    {
      "entropy": 6.010696268081665,
      "epoch": 0.7907949790794979,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000472,
      "loss": 5.9706,
      "mean_token_accuracy": 0.140121129155159,
      "num_tokens": 1740719.0,
      "step": 945
    },
    {
      "entropy": 6.144752645492554,
      "epoch": 0.7949790794979079,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004745,
      "loss": 6.0063,
      "mean_token_accuracy": 0.13252055421471595,
      "num_tokens": 1749753.0,
      "step": 950
    },
    {
      "entropy": 6.193329572677612,
      "epoch": 0.799163179916318,
      "grad_norm": 1.125,
      "learning_rate": 0.000477,
      "loss": 6.0296,
      "mean_token_accuracy": 0.12951601520180703,
      "num_tokens": 1758913.0,
      "step": 955
    },
    {
      "entropy": 6.127861022949219,
      "epoch": 0.803347280334728,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004795,
      "loss": 6.047,
      "mean_token_accuracy": 0.13796046003699303,
      "num_tokens": 1768536.0,
      "step": 960
    },
    {
      "entropy": 6.048326778411865,
      "epoch": 0.8075313807531381,
      "grad_norm": 1.25,
      "learning_rate": 0.000482,
      "loss": 5.9733,
      "mean_token_accuracy": 0.13804301396012306,
      "num_tokens": 1777291.0,
      "step": 965
    },
    {
      "entropy": 6.0986556053161625,
      "epoch": 0.8117154811715481,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004845,
      "loss": 6.0696,
      "mean_token_accuracy": 0.13444095477461815,
      "num_tokens": 1787186.0,
      "step": 970
    },
    {
      "entropy": 6.042343235015869,
      "epoch": 0.8158995815899581,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000487,
      "loss": 5.8767,
      "mean_token_accuracy": 0.14351157248020172,
      "num_tokens": 1795581.0,
      "step": 975
    },
    {
      "entropy": 5.956916570663452,
      "epoch": 0.8200836820083682,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004895,
      "loss": 5.8884,
      "mean_token_accuracy": 0.14488404095172883,
      "num_tokens": 1804627.0,
      "step": 980
    },
    {
      "entropy": 6.110140657424926,
      "epoch": 0.8242677824267782,
      "grad_norm": 1.203125,
      "learning_rate": 0.000492,
      "loss": 6.0162,
      "mean_token_accuracy": 0.13789800107479094,
      "num_tokens": 1813808.0,
      "step": 985
    },
    {
      "entropy": 6.012312602996826,
      "epoch": 0.8284518828451883,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004945,
      "loss": 6.008,
      "mean_token_accuracy": 0.13816680759191513,
      "num_tokens": 1823422.0,
      "step": 990
    },
    {
      "entropy": 6.114294052124023,
      "epoch": 0.8326359832635983,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000497,
      "loss": 6.0489,
      "mean_token_accuracy": 0.13757719099521637,
      "num_tokens": 1832437.0,
      "step": 995
    },
    {
      "entropy": 6.102473211288452,
      "epoch": 0.8368200836820083,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995,
      "loss": 5.97,
      "mean_token_accuracy": 0.13490361198782921,
      "num_tokens": 1842063.0,
      "step": 1000
    },
    {
      "epoch": 0.8368200836820083,
      "eval_entropy": 5.8732611679592015,
      "eval_loss": 6.004635334014893,
      "eval_mean_token_accuracy": 0.1470446902458653,
      "eval_num_tokens": 1842063.0,
      "eval_runtime": 2.7509,
      "eval_samples_per_second": 1420.266,
      "eval_steps_per_second": 177.76,
      "step": 1000
    }
  ],
  "logging_steps": 5,
  "max_steps": 11950,
  "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": 2662990405632000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}