jpn-jpan-100mb-100mb_seed10/checkpoint-21000/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.423542989036353,
  "eval_steps": 3000,
  "global_step": 21000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.692008876800537,
      "epoch": 0.0005770340450086555,
      "grad_norm": 11.4375,
      "learning_rate": 2e-06,
      "loss": 10.8655,
      "mean_token_accuracy": 0.0,
      "num_tokens": 14330.0,
      "step": 5
    },
    {
      "entropy": 10.691947269439698,
      "epoch": 0.001154068090017311,
      "grad_norm": 11.8125,
      "learning_rate": 4.5e-06,
      "loss": 10.7846,
      "mean_token_accuracy": 0.0006837096530944109,
      "num_tokens": 26284.0,
      "step": 10
    },
    {
      "entropy": 10.690341567993164,
      "epoch": 0.0017311021350259665,
      "grad_norm": 9.6875,
      "learning_rate": 7e-06,
      "loss": 10.5317,
      "mean_token_accuracy": 0.03542185090482235,
      "num_tokens": 38394.0,
      "step": 15
    },
    {
      "entropy": 10.66337537765503,
      "epoch": 0.002308136180034622,
      "grad_norm": 5.96875,
      "learning_rate": 9.5e-06,
      "loss": 10.2064,
      "mean_token_accuracy": 0.0529673807322979,
      "num_tokens": 51341.0,
      "step": 20
    },
    {
      "entropy": 10.496922397613526,
      "epoch": 0.0028851702250432777,
      "grad_norm": 3.84375,
      "learning_rate": 1.2e-05,
      "loss": 9.9191,
      "mean_token_accuracy": 0.05405417382717133,
      "num_tokens": 64559.0,
      "step": 25
    },
    {
      "entropy": 10.455469608306885,
      "epoch": 0.003462204270051933,
      "grad_norm": 3.46875,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 9.8404,
      "mean_token_accuracy": 0.053551160171628,
      "num_tokens": 76688.0,
      "step": 30
    },
    {
      "entropy": 10.447058010101319,
      "epoch": 0.004039238315060588,
      "grad_norm": 2.9375,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 9.7034,
      "mean_token_accuracy": 0.055309614166617393,
      "num_tokens": 88392.0,
      "step": 35
    },
    {
      "entropy": 10.431883907318115,
      "epoch": 0.004616272360069244,
      "grad_norm": 2.734375,
      "learning_rate": 1.95e-05,
      "loss": 9.7162,
      "mean_token_accuracy": 0.053266770020127294,
      "num_tokens": 100542.0,
      "step": 40
    },
    {
      "entropy": 10.33115463256836,
      "epoch": 0.0051933064050779,
      "grad_norm": 3.09375,
      "learning_rate": 2.2e-05,
      "loss": 9.5797,
      "mean_token_accuracy": 0.07032700143754482,
      "num_tokens": 113190.0,
      "step": 45
    },
    {
      "entropy": 10.309079837799072,
      "epoch": 0.005770340450086555,
      "grad_norm": 2.484375,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.5553,
      "mean_token_accuracy": 0.0662531953305006,
      "num_tokens": 125554.0,
      "step": 50
    },
    {
      "entropy": 10.260576152801514,
      "epoch": 0.006347374495095211,
      "grad_norm": 2.234375,
      "learning_rate": 2.7e-05,
      "loss": 9.514,
      "mean_token_accuracy": 0.07373123690485954,
      "num_tokens": 138527.0,
      "step": 55
    },
    {
      "entropy": 10.316655826568603,
      "epoch": 0.006924408540103866,
      "grad_norm": 2.828125,
      "learning_rate": 2.95e-05,
      "loss": 9.5893,
      "mean_token_accuracy": 0.06340348795056343,
      "num_tokens": 150404.0,
      "step": 60
    },
    {
      "entropy": 10.265781021118164,
      "epoch": 0.0075014425851125215,
      "grad_norm": 2.09375,
      "learning_rate": 3.2e-05,
      "loss": 9.4827,
      "mean_token_accuracy": 0.0759716846048832,
      "num_tokens": 163743.0,
      "step": 65
    },
    {
      "entropy": 10.188546752929687,
      "epoch": 0.008078476630121177,
      "grad_norm": 2.03125,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.4526,
      "mean_token_accuracy": 0.07428089305758476,
      "num_tokens": 175657.0,
      "step": 70
    },
    {
      "entropy": 10.151709461212159,
      "epoch": 0.008655510675129832,
      "grad_norm": 1.8671875,
      "learning_rate": 3.7e-05,
      "loss": 9.3966,
      "mean_token_accuracy": 0.07655606269836426,
      "num_tokens": 188756.0,
      "step": 75
    },
    {
      "entropy": 10.180650806427002,
      "epoch": 0.009232544720138488,
      "grad_norm": 2.15625,
      "learning_rate": 3.95e-05,
      "loss": 9.3212,
      "mean_token_accuracy": 0.07740879319608211,
      "num_tokens": 200548.0,
      "step": 80
    },
    {
      "entropy": 10.025209522247314,
      "epoch": 0.009809578765147143,
      "grad_norm": 2.0,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.2239,
      "mean_token_accuracy": 0.08616762235760689,
      "num_tokens": 213307.0,
      "step": 85
    },
    {
      "entropy": 10.14529628753662,
      "epoch": 0.0103866128101558,
      "grad_norm": 1.78125,
      "learning_rate": 4.45e-05,
      "loss": 9.2001,
      "mean_token_accuracy": 0.07975278049707413,
      "num_tokens": 226863.0,
      "step": 90
    },
    {
      "entropy": 10.057593822479248,
      "epoch": 0.010963646855164455,
      "grad_norm": 1.7265625,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 9.1455,
      "mean_token_accuracy": 0.08079247176647186,
      "num_tokens": 238655.0,
      "step": 95
    },
    {
      "entropy": 9.97190818786621,
      "epoch": 0.01154068090017311,
      "grad_norm": 1.78125,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 9.0904,
      "mean_token_accuracy": 0.08307576477527619,
      "num_tokens": 251118.0,
      "step": 100
    },
    {
      "entropy": 9.988478946685792,
      "epoch": 0.012117714945181766,
      "grad_norm": 1.84375,
      "learning_rate": 5.2e-05,
      "loss": 8.9828,
      "mean_token_accuracy": 0.08939878344535827,
      "num_tokens": 263665.0,
      "step": 105
    },
    {
      "entropy": 9.928913402557374,
      "epoch": 0.012694748990190421,
      "grad_norm": 1.9375,
      "learning_rate": 5.45e-05,
      "loss": 8.9232,
      "mean_token_accuracy": 0.08705427050590515,
      "num_tokens": 276531.0,
      "step": 110
    },
    {
      "entropy": 9.803722476959228,
      "epoch": 0.013271783035199077,
      "grad_norm": 1.6171875,
      "learning_rate": 5.7e-05,
      "loss": 8.8978,
      "mean_token_accuracy": 0.08507851585745811,
      "num_tokens": 289878.0,
      "step": 115
    },
    {
      "entropy": 9.810256195068359,
      "epoch": 0.013848817080207732,
      "grad_norm": 1.953125,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.8167,
      "mean_token_accuracy": 0.08298168927431107,
      "num_tokens": 302705.0,
      "step": 120
    },
    {
      "entropy": 9.63891897201538,
      "epoch": 0.014425851125216388,
      "grad_norm": 2.1875,
      "learning_rate": 6.2e-05,
      "loss": 8.6927,
      "mean_token_accuracy": 0.09015745520591736,
      "num_tokens": 314775.0,
      "step": 125
    },
    {
      "entropy": 9.604006576538087,
      "epoch": 0.015002885170225043,
      "grad_norm": 1.4609375,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.7246,
      "mean_token_accuracy": 0.08897400721907615,
      "num_tokens": 328563.0,
      "step": 130
    },
    {
      "entropy": 9.50387487411499,
      "epoch": 0.015579919215233698,
      "grad_norm": 1.40625,
      "learning_rate": 6.7e-05,
      "loss": 8.5573,
      "mean_token_accuracy": 0.09593394622206688,
      "num_tokens": 341820.0,
      "step": 135
    },
    {
      "entropy": 9.313443374633788,
      "epoch": 0.016156953260242354,
      "grad_norm": 1.1796875,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.5692,
      "mean_token_accuracy": 0.1013730101287365,
      "num_tokens": 354994.0,
      "step": 140
    },
    {
      "entropy": 9.28393840789795,
      "epoch": 0.01673398730525101,
      "grad_norm": 1.2265625,
      "learning_rate": 7.2e-05,
      "loss": 8.4769,
      "mean_token_accuracy": 0.09589103534817696,
      "num_tokens": 365852.0,
      "step": 145
    },
    {
      "entropy": 9.20916519165039,
      "epoch": 0.017311021350259664,
      "grad_norm": 1.7890625,
      "learning_rate": 7.45e-05,
      "loss": 8.4884,
      "mean_token_accuracy": 0.0926565632224083,
      "num_tokens": 379147.0,
      "step": 150
    },
    {
      "entropy": 9.129295921325683,
      "epoch": 0.01788805539526832,
      "grad_norm": 1.4140625,
      "learning_rate": 7.7e-05,
      "loss": 8.472,
      "mean_token_accuracy": 0.08658494129776954,
      "num_tokens": 392257.0,
      "step": 155
    },
    {
      "entropy": 9.01758165359497,
      "epoch": 0.018465089440276975,
      "grad_norm": 1.7890625,
      "learning_rate": 7.950000000000001e-05,
      "loss": 8.4153,
      "mean_token_accuracy": 0.08768738433718681,
      "num_tokens": 405249.0,
      "step": 160
    },
    {
      "entropy": 8.947623825073242,
      "epoch": 0.019042123485285632,
      "grad_norm": 1.375,
      "learning_rate": 8.2e-05,
      "loss": 8.389,
      "mean_token_accuracy": 0.09221537113189697,
      "num_tokens": 417819.0,
      "step": 165
    },
    {
      "entropy": 8.845267486572265,
      "epoch": 0.019619157530294286,
      "grad_norm": 1.3515625,
      "learning_rate": 8.450000000000001e-05,
      "loss": 8.2969,
      "mean_token_accuracy": 0.09195446819067002,
      "num_tokens": 429364.0,
      "step": 170
    },
    {
      "entropy": 8.706775093078614,
      "epoch": 0.020196191575302943,
      "grad_norm": 1.546875,
      "learning_rate": 8.7e-05,
      "loss": 8.3186,
      "mean_token_accuracy": 0.1003533847630024,
      "num_tokens": 441804.0,
      "step": 175
    },
    {
      "entropy": 8.740819454193115,
      "epoch": 0.0207732256203116,
      "grad_norm": 1.8125,
      "learning_rate": 8.95e-05,
      "loss": 8.2603,
      "mean_token_accuracy": 0.10172701776027679,
      "num_tokens": 453886.0,
      "step": 180
    },
    {
      "entropy": 8.652677536010742,
      "epoch": 0.021350259665320254,
      "grad_norm": 1.0703125,
      "learning_rate": 9.2e-05,
      "loss": 8.2733,
      "mean_token_accuracy": 0.09830424413084984,
      "num_tokens": 465828.0,
      "step": 185
    },
    {
      "entropy": 8.614245128631591,
      "epoch": 0.02192729371032891,
      "grad_norm": 1.4609375,
      "learning_rate": 9.45e-05,
      "loss": 8.225,
      "mean_token_accuracy": 0.10284605026245117,
      "num_tokens": 478783.0,
      "step": 190
    },
    {
      "entropy": 8.633440303802491,
      "epoch": 0.022504327755337564,
      "grad_norm": 1.234375,
      "learning_rate": 9.7e-05,
      "loss": 8.2896,
      "mean_token_accuracy": 0.0949929729104042,
      "num_tokens": 491167.0,
      "step": 195
    },
    {
      "entropy": 8.589673137664795,
      "epoch": 0.02308136180034622,
      "grad_norm": 1.78125,
      "learning_rate": 9.95e-05,
      "loss": 8.2636,
      "mean_token_accuracy": 0.09787702932953835,
      "num_tokens": 503511.0,
      "step": 200
    },
    {
      "entropy": 8.475285720825195,
      "epoch": 0.023658395845354875,
      "grad_norm": 1.25,
      "learning_rate": 0.000102,
      "loss": 8.1585,
      "mean_token_accuracy": 0.10205093994736672,
      "num_tokens": 515816.0,
      "step": 205
    },
    {
      "entropy": 8.644243526458741,
      "epoch": 0.024235429890363532,
      "grad_norm": 1.203125,
      "learning_rate": 0.00010449999999999999,
      "loss": 8.2575,
      "mean_token_accuracy": 0.09533519297838211,
      "num_tokens": 528919.0,
      "step": 210
    },
    {
      "entropy": 8.444662475585938,
      "epoch": 0.024812463935372186,
      "grad_norm": 1.109375,
      "learning_rate": 0.000107,
      "loss": 8.1863,
      "mean_token_accuracy": 0.10425740778446198,
      "num_tokens": 540973.0,
      "step": 215
    },
    {
      "entropy": 8.465065479278564,
      "epoch": 0.025389497980380843,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001095,
      "loss": 8.1542,
      "mean_token_accuracy": 0.11338075250387192,
      "num_tokens": 553265.0,
      "step": 220
    },
    {
      "entropy": 8.467651462554931,
      "epoch": 0.025966532025389497,
      "grad_norm": 1.046875,
      "learning_rate": 0.000112,
      "loss": 8.2439,
      "mean_token_accuracy": 0.10033949688076974,
      "num_tokens": 567534.0,
      "step": 225
    },
    {
      "entropy": 8.495089054107666,
      "epoch": 0.026543566070398154,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001145,
      "loss": 8.1763,
      "mean_token_accuracy": 0.10402834340929985,
      "num_tokens": 579297.0,
      "step": 230
    },
    {
      "entropy": 8.415564250946044,
      "epoch": 0.027120600115406807,
      "grad_norm": 1.125,
      "learning_rate": 0.00011700000000000001,
      "loss": 8.1508,
      "mean_token_accuracy": 0.10633992031216621,
      "num_tokens": 592554.0,
      "step": 235
    },
    {
      "entropy": 8.438064002990723,
      "epoch": 0.027697634160415464,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00011949999999999999,
      "loss": 8.1491,
      "mean_token_accuracy": 0.10545074120163918,
      "num_tokens": 605038.0,
      "step": 240
    },
    {
      "entropy": 8.381492137908936,
      "epoch": 0.02827466820542412,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000122,
      "loss": 8.163,
      "mean_token_accuracy": 0.1005297139286995,
      "num_tokens": 617172.0,
      "step": 245
    },
    {
      "entropy": 8.391489028930664,
      "epoch": 0.028851702250432775,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001245,
      "loss": 8.1113,
      "mean_token_accuracy": 0.1038644127547741,
      "num_tokens": 629243.0,
      "step": 250
    },
    {
      "entropy": 8.385975456237793,
      "epoch": 0.029428736295441432,
      "grad_norm": 1.15625,
      "learning_rate": 0.000127,
      "loss": 8.1635,
      "mean_token_accuracy": 0.10401127263903617,
      "num_tokens": 641487.0,
      "step": 255
    },
    {
      "entropy": 8.447768592834473,
      "epoch": 0.030005770340450086,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001295,
      "loss": 8.1661,
      "mean_token_accuracy": 0.09748540446162224,
      "num_tokens": 653214.0,
      "step": 260
    },
    {
      "entropy": 8.27870225906372,
      "epoch": 0.030582804385458743,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000132,
      "loss": 8.1277,
      "mean_token_accuracy": 0.11644875779747962,
      "num_tokens": 666549.0,
      "step": 265
    },
    {
      "entropy": 8.414948844909668,
      "epoch": 0.031159838430467397,
      "grad_norm": 1.375,
      "learning_rate": 0.00013450000000000002,
      "loss": 8.1369,
      "mean_token_accuracy": 0.1066818192601204,
      "num_tokens": 679443.0,
      "step": 270
    },
    {
      "entropy": 8.385531330108643,
      "epoch": 0.031736872475476054,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00013700000000000002,
      "loss": 8.1374,
      "mean_token_accuracy": 0.1021499790251255,
      "num_tokens": 692929.0,
      "step": 275
    },
    {
      "entropy": 8.398066520690918,
      "epoch": 0.03231390652048471,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001395,
      "loss": 8.0569,
      "mean_token_accuracy": 0.10605769827961922,
      "num_tokens": 705605.0,
      "step": 280
    },
    {
      "entropy": 8.333725929260254,
      "epoch": 0.03289094056549336,
      "grad_norm": 0.96875,
      "learning_rate": 0.00014199999999999998,
      "loss": 8.1521,
      "mean_token_accuracy": 0.09875996261835099,
      "num_tokens": 719064.0,
      "step": 285
    },
    {
      "entropy": 8.305869674682617,
      "epoch": 0.03346797461050202,
      "grad_norm": 1.046875,
      "learning_rate": 0.0001445,
      "loss": 8.0664,
      "mean_token_accuracy": 0.10697638019919395,
      "num_tokens": 732572.0,
      "step": 290
    },
    {
      "entropy": 8.336920833587646,
      "epoch": 0.034045008655510675,
      "grad_norm": 1.140625,
      "learning_rate": 0.000147,
      "loss": 8.0627,
      "mean_token_accuracy": 0.10701068118214607,
      "num_tokens": 745161.0,
      "step": 295
    },
    {
      "entropy": 8.26529426574707,
      "epoch": 0.03462204270051933,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001495,
      "loss": 8.0361,
      "mean_token_accuracy": 0.11285013034939766,
      "num_tokens": 756180.0,
      "step": 300
    },
    {
      "entropy": 8.337069225311279,
      "epoch": 0.03519907674552799,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000152,
      "loss": 8.0545,
      "mean_token_accuracy": 0.11181231290102005,
      "num_tokens": 768568.0,
      "step": 305
    },
    {
      "entropy": 8.28462495803833,
      "epoch": 0.03577611079053664,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00015450000000000001,
      "loss": 8.0859,
      "mean_token_accuracy": 0.11087586879730224,
      "num_tokens": 781580.0,
      "step": 310
    },
    {
      "entropy": 8.21790370941162,
      "epoch": 0.0363531448355453,
      "grad_norm": 1.1875,
      "learning_rate": 0.000157,
      "loss": 7.9589,
      "mean_token_accuracy": 0.12124751061201096,
      "num_tokens": 793486.0,
      "step": 315
    },
    {
      "entropy": 8.240262222290038,
      "epoch": 0.03693017888055395,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0001595,
      "loss": 8.0306,
      "mean_token_accuracy": 0.11093810424208642,
      "num_tokens": 805997.0,
      "step": 320
    },
    {
      "entropy": 8.202302837371827,
      "epoch": 0.03750721292556261,
      "grad_norm": 1.484375,
      "learning_rate": 0.000162,
      "loss": 7.9966,
      "mean_token_accuracy": 0.11470017209649086,
      "num_tokens": 817681.0,
      "step": 325
    },
    {
      "entropy": 8.279125785827636,
      "epoch": 0.038084246970571264,
      "grad_norm": 1.375,
      "learning_rate": 0.00016450000000000001,
      "loss": 8.0151,
      "mean_token_accuracy": 0.1159965842962265,
      "num_tokens": 830027.0,
      "step": 330
    },
    {
      "entropy": 8.221384429931641,
      "epoch": 0.03866128101557992,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00016700000000000002,
      "loss": 8.0353,
      "mean_token_accuracy": 0.11259357258677483,
      "num_tokens": 842862.0,
      "step": 335
    },
    {
      "entropy": 8.216114330291749,
      "epoch": 0.03923831506058857,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.9479,
      "mean_token_accuracy": 0.11576894521713257,
      "num_tokens": 856548.0,
      "step": 340
    },
    {
      "entropy": 8.159035778045654,
      "epoch": 0.03981534910559723,
      "grad_norm": 1.0625,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.9681,
      "mean_token_accuracy": 0.11400512978434563,
      "num_tokens": 869651.0,
      "step": 345
    },
    {
      "entropy": 8.185454750061036,
      "epoch": 0.040392383150605886,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.9938,
      "mean_token_accuracy": 0.1144215151667595,
      "num_tokens": 880717.0,
      "step": 350
    },
    {
      "entropy": 8.218715572357178,
      "epoch": 0.04096941719561454,
      "grad_norm": 1.15625,
      "learning_rate": 0.000177,
      "loss": 7.9851,
      "mean_token_accuracy": 0.11426586061716079,
      "num_tokens": 894066.0,
      "step": 355
    },
    {
      "entropy": 8.077167415618897,
      "epoch": 0.0415464512406232,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001795,
      "loss": 7.8762,
      "mean_token_accuracy": 0.11363315507769585,
      "num_tokens": 905287.0,
      "step": 360
    },
    {
      "entropy": 8.201995038986206,
      "epoch": 0.042123485285631854,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000182,
      "loss": 7.9672,
      "mean_token_accuracy": 0.1200343832373619,
      "num_tokens": 917512.0,
      "step": 365
    },
    {
      "entropy": 8.16402292251587,
      "epoch": 0.04270051933064051,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001845,
      "loss": 7.9206,
      "mean_token_accuracy": 0.11295781657099724,
      "num_tokens": 930263.0,
      "step": 370
    },
    {
      "entropy": 8.1646258354187,
      "epoch": 0.04327755337564916,
      "grad_norm": 1.28125,
      "learning_rate": 0.000187,
      "loss": 7.9761,
      "mean_token_accuracy": 0.11365053132176399,
      "num_tokens": 942330.0,
      "step": 375
    },
    {
      "entropy": 8.16142988204956,
      "epoch": 0.04385458742065782,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0001895,
      "loss": 7.9558,
      "mean_token_accuracy": 0.11354472935199737,
      "num_tokens": 954005.0,
      "step": 380
    },
    {
      "entropy": 8.192191314697265,
      "epoch": 0.044431621465666475,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000192,
      "loss": 7.9435,
      "mean_token_accuracy": 0.11344988718628883,
      "num_tokens": 966725.0,
      "step": 385
    },
    {
      "entropy": 8.170158672332764,
      "epoch": 0.04500865551067513,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001945,
      "loss": 7.9346,
      "mean_token_accuracy": 0.11046166941523552,
      "num_tokens": 978021.0,
      "step": 390
    },
    {
      "entropy": 8.039873075485229,
      "epoch": 0.04558568955568378,
      "grad_norm": 1.140625,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.9217,
      "mean_token_accuracy": 0.11965367197990417,
      "num_tokens": 991321.0,
      "step": 395
    },
    {
      "entropy": 8.160380363464355,
      "epoch": 0.04616272360069244,
      "grad_norm": 1.140625,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.9176,
      "mean_token_accuracy": 0.11871105283498765,
      "num_tokens": 1004172.0,
      "step": 400
    },
    {
      "entropy": 8.075436639785767,
      "epoch": 0.0467397576457011,
      "grad_norm": 1.078125,
      "learning_rate": 0.000202,
      "loss": 7.8652,
      "mean_token_accuracy": 0.12333920821547509,
      "num_tokens": 1016513.0,
      "step": 405
    },
    {
      "entropy": 8.040183401107788,
      "epoch": 0.04731679169070975,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020449999999999998,
      "loss": 7.867,
      "mean_token_accuracy": 0.11452069729566575,
      "num_tokens": 1028536.0,
      "step": 410
    },
    {
      "entropy": 8.143816423416137,
      "epoch": 0.04789382573571841,
      "grad_norm": 1.421875,
      "learning_rate": 0.000207,
      "loss": 7.7972,
      "mean_token_accuracy": 0.12931199073791505,
      "num_tokens": 1040534.0,
      "step": 415
    },
    {
      "entropy": 7.921964454650879,
      "epoch": 0.048470859780727064,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002095,
      "loss": 7.9327,
      "mean_token_accuracy": 0.1181611604988575,
      "num_tokens": 1054995.0,
      "step": 420
    },
    {
      "entropy": 8.17293095588684,
      "epoch": 0.04904789382573572,
      "grad_norm": 1.140625,
      "learning_rate": 0.000212,
      "loss": 7.9133,
      "mean_token_accuracy": 0.11562721282243729,
      "num_tokens": 1067722.0,
      "step": 425
    },
    {
      "entropy": 8.100650596618653,
      "epoch": 0.04962492787074437,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002145,
      "loss": 7.8862,
      "mean_token_accuracy": 0.111819938570261,
      "num_tokens": 1079788.0,
      "step": 430
    },
    {
      "entropy": 8.100652170181274,
      "epoch": 0.05020196191575303,
      "grad_norm": 1.1875,
      "learning_rate": 0.00021700000000000002,
      "loss": 7.885,
      "mean_token_accuracy": 0.11935903802514077,
      "num_tokens": 1092016.0,
      "step": 435
    },
    {
      "entropy": 8.106025123596192,
      "epoch": 0.050778995960761686,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002195,
      "loss": 7.901,
      "mean_token_accuracy": 0.11573071926832199,
      "num_tokens": 1104522.0,
      "step": 440
    },
    {
      "entropy": 8.052202367782593,
      "epoch": 0.05135603000577034,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000222,
      "loss": 7.9235,
      "mean_token_accuracy": 0.12538544684648514,
      "num_tokens": 1117239.0,
      "step": 445
    },
    {
      "entropy": 8.048323249816894,
      "epoch": 0.05193306405077899,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002245,
      "loss": 7.8395,
      "mean_token_accuracy": 0.12211849763989449,
      "num_tokens": 1129061.0,
      "step": 450
    },
    {
      "entropy": 8.052234315872193,
      "epoch": 0.052510098095787654,
      "grad_norm": 1.140625,
      "learning_rate": 0.00022700000000000002,
      "loss": 7.828,
      "mean_token_accuracy": 0.1126674197614193,
      "num_tokens": 1141409.0,
      "step": 455
    },
    {
      "entropy": 7.968437433242798,
      "epoch": 0.05308713214079631,
      "grad_norm": 1.140625,
      "learning_rate": 0.00022950000000000002,
      "loss": 7.8315,
      "mean_token_accuracy": 0.11563115417957306,
      "num_tokens": 1153623.0,
      "step": 460
    },
    {
      "entropy": 8.061407089233398,
      "epoch": 0.05366416618580496,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023200000000000003,
      "loss": 7.8652,
      "mean_token_accuracy": 0.11920796036720276,
      "num_tokens": 1166877.0,
      "step": 465
    },
    {
      "entropy": 8.019161272048951,
      "epoch": 0.054241200230813615,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00023449999999999998,
      "loss": 7.8554,
      "mean_token_accuracy": 0.11903531998395919,
      "num_tokens": 1179625.0,
      "step": 470
    },
    {
      "entropy": 7.995475769042969,
      "epoch": 0.054818234275822275,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000237,
      "loss": 7.8848,
      "mean_token_accuracy": 0.11980156600475311,
      "num_tokens": 1193586.0,
      "step": 475
    },
    {
      "entropy": 8.04285216331482,
      "epoch": 0.05539526832083093,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002395,
      "loss": 7.8237,
      "mean_token_accuracy": 0.12064610123634338,
      "num_tokens": 1207086.0,
      "step": 480
    },
    {
      "entropy": 7.9670654296875,
      "epoch": 0.05597230236583958,
      "grad_norm": 1.203125,
      "learning_rate": 0.000242,
      "loss": 7.8129,
      "mean_token_accuracy": 0.12205322310328484,
      "num_tokens": 1219389.0,
      "step": 485
    },
    {
      "entropy": 7.9849029064178465,
      "epoch": 0.05654933641084824,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002445,
      "loss": 7.7593,
      "mean_token_accuracy": 0.12576186433434486,
      "num_tokens": 1232080.0,
      "step": 490
    },
    {
      "entropy": 7.957043218612671,
      "epoch": 0.0571263704558569,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000247,
      "loss": 7.8116,
      "mean_token_accuracy": 0.1276965133845806,
      "num_tokens": 1246775.0,
      "step": 495
    },
    {
      "entropy": 7.927796268463135,
      "epoch": 0.05770340450086555,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002495,
      "loss": 7.8609,
      "mean_token_accuracy": 0.11414324194192886,
      "num_tokens": 1258839.0,
      "step": 500
    },
    {
      "entropy": 7.979376649856567,
      "epoch": 0.058280438545874204,
      "grad_norm": 1.078125,
      "learning_rate": 0.000252,
      "loss": 7.7701,
      "mean_token_accuracy": 0.12441443428397178,
      "num_tokens": 1270710.0,
      "step": 505
    },
    {
      "entropy": 7.965421342849732,
      "epoch": 0.058857472590882864,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002545,
      "loss": 7.7192,
      "mean_token_accuracy": 0.12516708746552468,
      "num_tokens": 1281930.0,
      "step": 510
    },
    {
      "entropy": 7.895893859863281,
      "epoch": 0.05943450663589152,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000257,
      "loss": 7.8013,
      "mean_token_accuracy": 0.11694373786449433,
      "num_tokens": 1294548.0,
      "step": 515
    },
    {
      "entropy": 7.908454704284668,
      "epoch": 0.06001154068090017,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002595,
      "loss": 7.7466,
      "mean_token_accuracy": 0.11997460946440697,
      "num_tokens": 1307588.0,
      "step": 520
    },
    {
      "entropy": 7.922945499420166,
      "epoch": 0.060588574725908825,
      "grad_norm": 1.0625,
      "learning_rate": 0.000262,
      "loss": 7.7252,
      "mean_token_accuracy": 0.12717078104615212,
      "num_tokens": 1319478.0,
      "step": 525
    },
    {
      "entropy": 7.9428062915802,
      "epoch": 0.061165608770917486,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026450000000000003,
      "loss": 7.8123,
      "mean_token_accuracy": 0.11468452289700508,
      "num_tokens": 1332950.0,
      "step": 530
    },
    {
      "entropy": 7.8876683712005615,
      "epoch": 0.06174264281592614,
      "grad_norm": 1.0625,
      "learning_rate": 0.00026700000000000004,
      "loss": 7.6891,
      "mean_token_accuracy": 0.12758101224899293,
      "num_tokens": 1345510.0,
      "step": 535
    },
    {
      "entropy": 7.72719554901123,
      "epoch": 0.06231967686093479,
      "grad_norm": 2.171875,
      "learning_rate": 0.00026950000000000005,
      "loss": 7.7493,
      "mean_token_accuracy": 0.1305077627301216,
      "num_tokens": 1359388.0,
      "step": 540
    },
    {
      "entropy": 7.881310319900512,
      "epoch": 0.06289671090594345,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027200000000000005,
      "loss": 7.7725,
      "mean_token_accuracy": 0.1230618417263031,
      "num_tokens": 1373067.0,
      "step": 545
    },
    {
      "entropy": 7.837239837646484,
      "epoch": 0.06347374495095211,
      "grad_norm": 0.953125,
      "learning_rate": 0.0002745,
      "loss": 7.7374,
      "mean_token_accuracy": 0.1261225961148739,
      "num_tokens": 1385718.0,
      "step": 550
    },
    {
      "entropy": 8.000623750686646,
      "epoch": 0.06405077899596076,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000277,
      "loss": 7.7379,
      "mean_token_accuracy": 0.12663036063313485,
      "num_tokens": 1398788.0,
      "step": 555
    },
    {
      "entropy": 7.7889786720275875,
      "epoch": 0.06462781304096941,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002795,
      "loss": 7.7291,
      "mean_token_accuracy": 0.13189933598041534,
      "num_tokens": 1411752.0,
      "step": 560
    },
    {
      "entropy": 7.891875267028809,
      "epoch": 0.06520484708597807,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00028199999999999997,
      "loss": 7.7763,
      "mean_token_accuracy": 0.12392229437828065,
      "num_tokens": 1423979.0,
      "step": 565
    },
    {
      "entropy": 7.946758174896241,
      "epoch": 0.06578188113098672,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002845,
      "loss": 7.828,
      "mean_token_accuracy": 0.1189330443739891,
      "num_tokens": 1437009.0,
      "step": 570
    },
    {
      "entropy": 7.795953941345215,
      "epoch": 0.06635891517599539,
      "grad_norm": 1.296875,
      "learning_rate": 0.000287,
      "loss": 7.6487,
      "mean_token_accuracy": 0.12712273001670837,
      "num_tokens": 1448220.0,
      "step": 575
    },
    {
      "entropy": 7.787666893005371,
      "epoch": 0.06693594922100404,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002895,
      "loss": 7.6992,
      "mean_token_accuracy": 0.12536668553948402,
      "num_tokens": 1460024.0,
      "step": 580
    },
    {
      "entropy": 8.011966657638549,
      "epoch": 0.0675129832660127,
      "grad_norm": 1.03125,
      "learning_rate": 0.000292,
      "loss": 7.7021,
      "mean_token_accuracy": 0.11621926128864288,
      "num_tokens": 1472283.0,
      "step": 585
    },
    {
      "entropy": 7.664339017868042,
      "epoch": 0.06809001731102135,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0002945,
      "loss": 7.7415,
      "mean_token_accuracy": 0.12770119607448577,
      "num_tokens": 1486565.0,
      "step": 590
    },
    {
      "entropy": 7.810535287857055,
      "epoch": 0.06866705135603,
      "grad_norm": 1.046875,
      "learning_rate": 0.000297,
      "loss": 7.7145,
      "mean_token_accuracy": 0.1263989932835102,
      "num_tokens": 1498708.0,
      "step": 595
    },
    {
      "entropy": 7.804998970031738,
      "epoch": 0.06924408540103866,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002995,
      "loss": 7.6492,
      "mean_token_accuracy": 0.1294543094933033,
      "num_tokens": 1511318.0,
      "step": 600
    },
    {
      "entropy": 7.817861795425415,
      "epoch": 0.06982111944604731,
      "grad_norm": 0.984375,
      "learning_rate": 0.000302,
      "loss": 7.6533,
      "mean_token_accuracy": 0.13287195861339568,
      "num_tokens": 1525415.0,
      "step": 605
    },
    {
      "entropy": 7.86554970741272,
      "epoch": 0.07039815349105598,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003045,
      "loss": 7.6892,
      "mean_token_accuracy": 0.12285038828849792,
      "num_tokens": 1538757.0,
      "step": 610
    },
    {
      "entropy": 7.643300628662109,
      "epoch": 0.07097518753606463,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000307,
      "loss": 7.671,
      "mean_token_accuracy": 0.12466116324067116,
      "num_tokens": 1550619.0,
      "step": 615
    },
    {
      "entropy": 7.822739982604981,
      "epoch": 0.07155222158107329,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003095,
      "loss": 7.7067,
      "mean_token_accuracy": 0.12112424820661545,
      "num_tokens": 1562588.0,
      "step": 620
    },
    {
      "entropy": 7.776826238632202,
      "epoch": 0.07212925562608194,
      "grad_norm": 1.046875,
      "learning_rate": 0.000312,
      "loss": 7.6167,
      "mean_token_accuracy": 0.1296200931072235,
      "num_tokens": 1574892.0,
      "step": 625
    },
    {
      "entropy": 7.708193683624268,
      "epoch": 0.0727062896710906,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003145,
      "loss": 7.6351,
      "mean_token_accuracy": 0.1296207919716835,
      "num_tokens": 1587525.0,
      "step": 630
    },
    {
      "entropy": 7.8489649295806885,
      "epoch": 0.07328332371609925,
      "grad_norm": 1.15625,
      "learning_rate": 0.000317,
      "loss": 7.6498,
      "mean_token_accuracy": 0.11944840252399444,
      "num_tokens": 1600350.0,
      "step": 635
    },
    {
      "entropy": 7.635661983489991,
      "epoch": 0.0738603577611079,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0003195,
      "loss": 7.6298,
      "mean_token_accuracy": 0.13184600844979286,
      "num_tokens": 1613318.0,
      "step": 640
    },
    {
      "entropy": 7.869474792480469,
      "epoch": 0.07443739180611655,
      "grad_norm": 1.15625,
      "learning_rate": 0.000322,
      "loss": 7.7555,
      "mean_token_accuracy": 0.11981712207198143,
      "num_tokens": 1626757.0,
      "step": 645
    },
    {
      "entropy": 7.726322650909424,
      "epoch": 0.07501442585112522,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032450000000000003,
      "loss": 7.5653,
      "mean_token_accuracy": 0.1351322740316391,
      "num_tokens": 1639185.0,
      "step": 650
    },
    {
      "entropy": 7.705139207839966,
      "epoch": 0.07559145989613388,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032700000000000003,
      "loss": 7.6865,
      "mean_token_accuracy": 0.12470589056611062,
      "num_tokens": 1652009.0,
      "step": 655
    },
    {
      "entropy": 7.701083326339722,
      "epoch": 0.07616849394114253,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032950000000000004,
      "loss": 7.6647,
      "mean_token_accuracy": 0.12210506573319435,
      "num_tokens": 1665604.0,
      "step": 660
    },
    {
      "entropy": 7.796258974075317,
      "epoch": 0.07674552798615118,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00033200000000000005,
      "loss": 7.6319,
      "mean_token_accuracy": 0.12362316846847535,
      "num_tokens": 1679036.0,
      "step": 665
    },
    {
      "entropy": 7.649466943740845,
      "epoch": 0.07732256203115984,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033450000000000005,
      "loss": 7.6317,
      "mean_token_accuracy": 0.12975930497050286,
      "num_tokens": 1691512.0,
      "step": 670
    },
    {
      "entropy": 7.84274697303772,
      "epoch": 0.07789959607616849,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000337,
      "loss": 7.6818,
      "mean_token_accuracy": 0.12658526599407197,
      "num_tokens": 1704589.0,
      "step": 675
    },
    {
      "entropy": 7.613417482376098,
      "epoch": 0.07847663012117714,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003395,
      "loss": 7.6798,
      "mean_token_accuracy": 0.11807645931839943,
      "num_tokens": 1718612.0,
      "step": 680
    },
    {
      "entropy": 7.623522710800171,
      "epoch": 0.07905366416618581,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000342,
      "loss": 7.5748,
      "mean_token_accuracy": 0.12185449972748756,
      "num_tokens": 1730472.0,
      "step": 685
    },
    {
      "entropy": 7.759632921218872,
      "epoch": 0.07963069821119446,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034449999999999997,
      "loss": 7.7018,
      "mean_token_accuracy": 0.12128116115927697,
      "num_tokens": 1742535.0,
      "step": 690
    },
    {
      "entropy": 7.77196855545044,
      "epoch": 0.08020773225620312,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000347,
      "loss": 7.6766,
      "mean_token_accuracy": 0.11788606792688369,
      "num_tokens": 1753877.0,
      "step": 695
    },
    {
      "entropy": 7.763478708267212,
      "epoch": 0.08078476630121177,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003495,
      "loss": 7.5707,
      "mean_token_accuracy": 0.12885044515132904,
      "num_tokens": 1766413.0,
      "step": 700
    },
    {
      "entropy": 7.513013744354248,
      "epoch": 0.08136180034622043,
      "grad_norm": 1.25,
      "learning_rate": 0.000352,
      "loss": 7.4623,
      "mean_token_accuracy": 0.13055645897984505,
      "num_tokens": 1779719.0,
      "step": 705
    },
    {
      "entropy": 7.673223543167114,
      "epoch": 0.08193883439122908,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003545,
      "loss": 7.5946,
      "mean_token_accuracy": 0.12953749522566796,
      "num_tokens": 1791448.0,
      "step": 710
    },
    {
      "entropy": 7.66050238609314,
      "epoch": 0.08251586843623773,
      "grad_norm": 1.015625,
      "learning_rate": 0.000357,
      "loss": 7.603,
      "mean_token_accuracy": 0.12697791904211045,
      "num_tokens": 1803802.0,
      "step": 715
    },
    {
      "entropy": 7.657990217208862,
      "epoch": 0.0830929024812464,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003595,
      "loss": 7.6079,
      "mean_token_accuracy": 0.12700373977422713,
      "num_tokens": 1817545.0,
      "step": 720
    },
    {
      "entropy": 7.667493295669556,
      "epoch": 0.08366993652625505,
      "grad_norm": 0.90625,
      "learning_rate": 0.000362,
      "loss": 7.5869,
      "mean_token_accuracy": 0.12516693994402886,
      "num_tokens": 1830115.0,
      "step": 725
    },
    {
      "entropy": 7.611949396133423,
      "epoch": 0.08424697057126371,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003645,
      "loss": 7.5303,
      "mean_token_accuracy": 0.1249528095126152,
      "num_tokens": 1844707.0,
      "step": 730
    },
    {
      "entropy": 7.6815104484558105,
      "epoch": 0.08482400461627236,
      "grad_norm": 1.0625,
      "learning_rate": 0.000367,
      "loss": 7.6093,
      "mean_token_accuracy": 0.12226399332284928,
      "num_tokens": 1858719.0,
      "step": 735
    },
    {
      "entropy": 7.659861135482788,
      "epoch": 0.08540103866128101,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003695,
      "loss": 7.5551,
      "mean_token_accuracy": 0.1302394263446331,
      "num_tokens": 1870421.0,
      "step": 740
    },
    {
      "entropy": 7.612351417541504,
      "epoch": 0.08597807270628967,
      "grad_norm": 1.03125,
      "learning_rate": 0.000372,
      "loss": 7.6212,
      "mean_token_accuracy": 0.11757046580314637,
      "num_tokens": 1883792.0,
      "step": 745
    },
    {
      "entropy": 7.606675052642823,
      "epoch": 0.08655510675129832,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003745,
      "loss": 7.598,
      "mean_token_accuracy": 0.12712156772613525,
      "num_tokens": 1896209.0,
      "step": 750
    },
    {
      "entropy": 7.588115072250366,
      "epoch": 0.08713214079630698,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000377,
      "loss": 7.5331,
      "mean_token_accuracy": 0.12872409150004388,
      "num_tokens": 1908520.0,
      "step": 755
    },
    {
      "entropy": 7.640614938735962,
      "epoch": 0.08770917484131564,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003795,
      "loss": 7.5346,
      "mean_token_accuracy": 0.12833384424448013,
      "num_tokens": 1922489.0,
      "step": 760
    },
    {
      "entropy": 7.590028858184814,
      "epoch": 0.0882862088863243,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000382,
      "loss": 7.5305,
      "mean_token_accuracy": 0.12443057373166085,
      "num_tokens": 1935206.0,
      "step": 765
    },
    {
      "entropy": 7.588243770599365,
      "epoch": 0.08886324293133295,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003845,
      "loss": 7.5068,
      "mean_token_accuracy": 0.1315755218267441,
      "num_tokens": 1948798.0,
      "step": 770
    },
    {
      "entropy": 7.5549592018127445,
      "epoch": 0.0894402769763416,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038700000000000003,
      "loss": 7.4928,
      "mean_token_accuracy": 0.13577663153409958,
      "num_tokens": 1961095.0,
      "step": 775
    },
    {
      "entropy": 7.596438980102539,
      "epoch": 0.09001731102135026,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038950000000000003,
      "loss": 7.5125,
      "mean_token_accuracy": 0.1260148160159588,
      "num_tokens": 1974229.0,
      "step": 780
    },
    {
      "entropy": 7.607128286361695,
      "epoch": 0.09059434506635891,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00039200000000000004,
      "loss": 7.602,
      "mean_token_accuracy": 0.12650037407875062,
      "num_tokens": 1986334.0,
      "step": 785
    },
    {
      "entropy": 7.584266185760498,
      "epoch": 0.09117137911136756,
      "grad_norm": 0.921875,
      "learning_rate": 0.00039450000000000005,
      "loss": 7.5522,
      "mean_token_accuracy": 0.1282362699508667,
      "num_tokens": 2000337.0,
      "step": 790
    },
    {
      "entropy": 7.498491811752319,
      "epoch": 0.09174841315637623,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039700000000000005,
      "loss": 7.4676,
      "mean_token_accuracy": 0.13595186024904252,
      "num_tokens": 2012048.0,
      "step": 795
    },
    {
      "entropy": 7.588355207443238,
      "epoch": 0.09232544720138489,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003995,
      "loss": 7.497,
      "mean_token_accuracy": 0.13125768825411796,
      "num_tokens": 2025244.0,
      "step": 800
    },
    {
      "entropy": 7.5243137836456295,
      "epoch": 0.09290248124639354,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000402,
      "loss": 7.5419,
      "mean_token_accuracy": 0.13079462423920632,
      "num_tokens": 2036542.0,
      "step": 805
    },
    {
      "entropy": 7.59242262840271,
      "epoch": 0.0934795152914022,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004045,
      "loss": 7.484,
      "mean_token_accuracy": 0.13099602833390236,
      "num_tokens": 2049358.0,
      "step": 810
    },
    {
      "entropy": 7.438653898239136,
      "epoch": 0.09405654933641085,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040699999999999997,
      "loss": 7.4969,
      "mean_token_accuracy": 0.13118922114372253,
      "num_tokens": 2062114.0,
      "step": 815
    },
    {
      "entropy": 7.643336772918701,
      "epoch": 0.0946335833814195,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004095,
      "loss": 7.4629,
      "mean_token_accuracy": 0.1343265689909458,
      "num_tokens": 2074236.0,
      "step": 820
    },
    {
      "entropy": 7.525158262252807,
      "epoch": 0.09521061742642815,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000412,
      "loss": 7.3893,
      "mean_token_accuracy": 0.135568980127573,
      "num_tokens": 2086320.0,
      "step": 825
    },
    {
      "entropy": 7.457302284240723,
      "epoch": 0.09578765147143682,
      "grad_norm": 1.0,
      "learning_rate": 0.0004145,
      "loss": 7.4473,
      "mean_token_accuracy": 0.13218377977609636,
      "num_tokens": 2099119.0,
      "step": 830
    },
    {
      "entropy": 7.532028770446777,
      "epoch": 0.09636468551644548,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000417,
      "loss": 7.4827,
      "mean_token_accuracy": 0.13144656643271446,
      "num_tokens": 2111795.0,
      "step": 835
    },
    {
      "entropy": 7.413727283477783,
      "epoch": 0.09694171956145413,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004195,
      "loss": 7.4238,
      "mean_token_accuracy": 0.12527592554688455,
      "num_tokens": 2126621.0,
      "step": 840
    },
    {
      "entropy": 7.471304702758789,
      "epoch": 0.09751875360646278,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000422,
      "loss": 7.3631,
      "mean_token_accuracy": 0.14259372502565384,
      "num_tokens": 2139770.0,
      "step": 845
    },
    {
      "entropy": 7.440826320648194,
      "epoch": 0.09809578765147144,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004245,
      "loss": 7.4611,
      "mean_token_accuracy": 0.13174420446157456,
      "num_tokens": 2151683.0,
      "step": 850
    },
    {
      "entropy": 7.430302906036377,
      "epoch": 0.09867282169648009,
      "grad_norm": 0.96875,
      "learning_rate": 0.000427,
      "loss": 7.421,
      "mean_token_accuracy": 0.13163666427135468,
      "num_tokens": 2163178.0,
      "step": 855
    },
    {
      "entropy": 7.475621461868286,
      "epoch": 0.09924985574148874,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004295,
      "loss": 7.5021,
      "mean_token_accuracy": 0.1217731073498726,
      "num_tokens": 2176312.0,
      "step": 860
    },
    {
      "entropy": 7.50553822517395,
      "epoch": 0.0998268897864974,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000432,
      "loss": 7.4503,
      "mean_token_accuracy": 0.12792299836874008,
      "num_tokens": 2189045.0,
      "step": 865
    },
    {
      "entropy": 7.380361652374267,
      "epoch": 0.10040392383150606,
      "grad_norm": 1.125,
      "learning_rate": 0.0004345,
      "loss": 7.4504,
      "mean_token_accuracy": 0.1204420991241932,
      "num_tokens": 2202083.0,
      "step": 870
    },
    {
      "entropy": 7.474799299240113,
      "epoch": 0.10098095787651472,
      "grad_norm": 1.15625,
      "learning_rate": 0.000437,
      "loss": 7.4886,
      "mean_token_accuracy": 0.12477894797921181,
      "num_tokens": 2215775.0,
      "step": 875
    },
    {
      "entropy": 7.536693000793457,
      "epoch": 0.10155799192152337,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004395,
      "loss": 7.4308,
      "mean_token_accuracy": 0.12408801987767219,
      "num_tokens": 2228387.0,
      "step": 880
    },
    {
      "entropy": 7.429737758636475,
      "epoch": 0.10213502596653203,
      "grad_norm": 1.015625,
      "learning_rate": 0.000442,
      "loss": 7.442,
      "mean_token_accuracy": 0.13106903955340385,
      "num_tokens": 2241491.0,
      "step": 885
    },
    {
      "entropy": 7.284702110290527,
      "epoch": 0.10271206001154068,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004445,
      "loss": 7.3722,
      "mean_token_accuracy": 0.13908286914229392,
      "num_tokens": 2255282.0,
      "step": 890
    },
    {
      "entropy": 7.476779127120972,
      "epoch": 0.10328909405654933,
      "grad_norm": 1.03125,
      "learning_rate": 0.000447,
      "loss": 7.3868,
      "mean_token_accuracy": 0.1294294334948063,
      "num_tokens": 2268081.0,
      "step": 895
    },
    {
      "entropy": 7.410451793670655,
      "epoch": 0.10386612810155799,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044950000000000003,
      "loss": 7.3463,
      "mean_token_accuracy": 0.1367862269282341,
      "num_tokens": 2281069.0,
      "step": 900
    },
    {
      "entropy": 7.424945640563965,
      "epoch": 0.10444316214656665,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045200000000000004,
      "loss": 7.4009,
      "mean_token_accuracy": 0.12864845171570777,
      "num_tokens": 2293606.0,
      "step": 905
    },
    {
      "entropy": 7.475167083740234,
      "epoch": 0.10502019619157531,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045450000000000004,
      "loss": 7.3905,
      "mean_token_accuracy": 0.13015836849808693,
      "num_tokens": 2304973.0,
      "step": 910
    },
    {
      "entropy": 7.403925657272339,
      "epoch": 0.10559723023658396,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00045700000000000005,
      "loss": 7.426,
      "mean_token_accuracy": 0.12412591949105263,
      "num_tokens": 2319016.0,
      "step": 915
    },
    {
      "entropy": 7.38353796005249,
      "epoch": 0.10617426428159261,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045950000000000006,
      "loss": 7.402,
      "mean_token_accuracy": 0.1354122206568718,
      "num_tokens": 2330599.0,
      "step": 920
    },
    {
      "entropy": 7.44637622833252,
      "epoch": 0.10675129832660127,
      "grad_norm": 1.03125,
      "learning_rate": 0.000462,
      "loss": 7.4054,
      "mean_token_accuracy": 0.12546970024704934,
      "num_tokens": 2343153.0,
      "step": 925
    },
    {
      "entropy": 7.42939600944519,
      "epoch": 0.10732833237160992,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004645,
      "loss": 7.4317,
      "mean_token_accuracy": 0.12410960122942924,
      "num_tokens": 2356762.0,
      "step": 930
    },
    {
      "entropy": 7.285542058944702,
      "epoch": 0.10790536641661858,
      "grad_norm": 1.171875,
      "learning_rate": 0.000467,
      "loss": 7.4542,
      "mean_token_accuracy": 0.1292865566909313,
      "num_tokens": 2369581.0,
      "step": 935
    },
    {
      "entropy": 7.475341939926148,
      "epoch": 0.10848240046162723,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004695,
      "loss": 7.3583,
      "mean_token_accuracy": 0.12862989977002143,
      "num_tokens": 2381419.0,
      "step": 940
    },
    {
      "entropy": 7.332505321502685,
      "epoch": 0.1090594345066359,
      "grad_norm": 1.015625,
      "learning_rate": 0.000472,
      "loss": 7.3724,
      "mean_token_accuracy": 0.13676412105560304,
      "num_tokens": 2393586.0,
      "step": 945
    },
    {
      "entropy": 7.300458240509033,
      "epoch": 0.10963646855164455,
      "grad_norm": 1.0,
      "learning_rate": 0.0004745,
      "loss": 7.36,
      "mean_token_accuracy": 0.1377156250178814,
      "num_tokens": 2407720.0,
      "step": 950
    },
    {
      "entropy": 7.401245880126953,
      "epoch": 0.1102135025966532,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000477,
      "loss": 7.4057,
      "mean_token_accuracy": 0.12623452842235566,
      "num_tokens": 2421089.0,
      "step": 955
    },
    {
      "entropy": 7.3679827690124515,
      "epoch": 0.11079053664166186,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004795,
      "loss": 7.3399,
      "mean_token_accuracy": 0.1357784502208233,
      "num_tokens": 2432859.0,
      "step": 960
    },
    {
      "entropy": 7.317124748229981,
      "epoch": 0.11136757068667051,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000482,
      "loss": 7.3411,
      "mean_token_accuracy": 0.12867245748639106,
      "num_tokens": 2445072.0,
      "step": 965
    },
    {
      "entropy": 7.359953451156616,
      "epoch": 0.11194460473167916,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004845,
      "loss": 7.3532,
      "mean_token_accuracy": 0.13644537180662156,
      "num_tokens": 2458724.0,
      "step": 970
    },
    {
      "entropy": 7.275381755828858,
      "epoch": 0.11252163877668782,
      "grad_norm": 1.015625,
      "learning_rate": 0.000487,
      "loss": 7.4164,
      "mean_token_accuracy": 0.12978286817669868,
      "num_tokens": 2472976.0,
      "step": 975
    },
    {
      "entropy": 7.473965215682983,
      "epoch": 0.11309867282169649,
      "grad_norm": 1.0,
      "learning_rate": 0.0004895,
      "loss": 7.371,
      "mean_token_accuracy": 0.12792500779032706,
      "num_tokens": 2486164.0,
      "step": 980
    },
    {
      "entropy": 7.275361776351929,
      "epoch": 0.11367570686670514,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000492,
      "loss": 7.3328,
      "mean_token_accuracy": 0.13169763311743737,
      "num_tokens": 2499395.0,
      "step": 985
    },
    {
      "entropy": 7.323497009277344,
      "epoch": 0.1142527409117138,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004945,
      "loss": 7.3271,
      "mean_token_accuracy": 0.13341858312487603,
      "num_tokens": 2512163.0,
      "step": 990
    },
    {
      "entropy": 7.386171340942383,
      "epoch": 0.11482977495672245,
      "grad_norm": 1.015625,
      "learning_rate": 0.000497,
      "loss": 7.3263,
      "mean_token_accuracy": 0.12971674054861068,
      "num_tokens": 2523634.0,
      "step": 995
    },
    {
      "entropy": 7.3099339485168455,
      "epoch": 0.1154068090017311,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995,
      "loss": 7.362,
      "mean_token_accuracy": 0.135257351398468,
      "num_tokens": 2536237.0,
      "step": 1000
    },
    {
      "entropy": 7.379648447036743,
      "epoch": 0.11598384304673975,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999999975783157,
      "loss": 7.3765,
      "mean_token_accuracy": 0.12056859806180001,
      "num_tokens": 2548035.0,
      "step": 1005
    },
    {
      "entropy": 7.317667388916016,
      "epoch": 0.11656087709174841,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999999877402236,
      "loss": 7.3159,
      "mean_token_accuracy": 0.13096051216125487,
      "num_tokens": 2561670.0,
      "step": 1010
    },
    {
      "entropy": 7.273052406311035,
      "epoch": 0.11713791113675708,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999999703343686,
      "loss": 7.3317,
      "mean_token_accuracy": 0.13321302831172943,
      "num_tokens": 2573471.0,
      "step": 1015
    },
    {
      "entropy": 7.3107616901397705,
      "epoch": 0.11771494518176573,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999999453607515,
      "loss": 7.282,
      "mean_token_accuracy": 0.13478338941931725,
      "num_tokens": 2586534.0,
      "step": 1020
    },
    {
      "entropy": 7.346148490905762,
      "epoch": 0.11829197922677438,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999999128193729,
      "loss": 7.3461,
      "mean_token_accuracy": 0.12729120627045631,
      "num_tokens": 2597970.0,
      "step": 1025
    },
    {
      "entropy": 7.172978067398072,
      "epoch": 0.11886901327178304,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000499999872710234,
      "loss": 7.2802,
      "mean_token_accuracy": 0.13685771971940994,
      "num_tokens": 2610924.0,
      "step": 1030
    },
    {
      "entropy": 7.363209962844849,
      "epoch": 0.11944604731679169,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999998250333361,
      "loss": 7.3474,
      "mean_token_accuracy": 0.13248208165168762,
      "num_tokens": 2623571.0,
      "step": 1035
    },
    {
      "entropy": 7.261402225494384,
      "epoch": 0.12002308136180034,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499999769788681,
      "loss": 7.3617,
      "mean_token_accuracy": 0.1302515096962452,
      "num_tokens": 2636063.0,
      "step": 1040
    },
    {
      "entropy": 7.279189252853394,
      "epoch": 0.120600115406809,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999997069762703,
      "loss": 7.3021,
      "mean_token_accuracy": 0.13231852501630784,
      "num_tokens": 2648621.0,
      "step": 1045
    },
    {
      "entropy": 7.2357665538787845,
      "epoch": 0.12117714945181765,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999996365961062,
      "loss": 7.2375,
      "mean_token_accuracy": 0.13813495635986328,
      "num_tokens": 2661794.0,
      "step": 1050
    },
    {
      "entropy": 7.320775318145752,
      "epoch": 0.12175418349682632,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999995586481912,
      "loss": 7.3105,
      "mean_token_accuracy": 0.12649577111005783,
      "num_tokens": 2674048.0,
      "step": 1055
    },
    {
      "entropy": 7.145959424972534,
      "epoch": 0.12233121754183497,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999994731325276,
      "loss": 7.2144,
      "mean_token_accuracy": 0.13311017528176308,
      "num_tokens": 2685510.0,
      "step": 1060
    },
    {
      "entropy": 7.320645141601562,
      "epoch": 0.12290825158684363,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004999993800491187,
      "loss": 7.3029,
      "mean_token_accuracy": 0.13155850619077683,
      "num_tokens": 2698414.0,
      "step": 1065
    },
    {
      "entropy": 7.2215817928314205,
      "epoch": 0.12348528563185228,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999992793979672,
      "loss": 7.301,
      "mean_token_accuracy": 0.13250066041946412,
      "num_tokens": 2711343.0,
      "step": 1070
    },
    {
      "entropy": 7.241006183624267,
      "epoch": 0.12406231967686093,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999991711790769,
      "loss": 7.1878,
      "mean_token_accuracy": 0.14295068979263306,
      "num_tokens": 2723123.0,
      "step": 1075
    },
    {
      "entropy": 7.133680200576782,
      "epoch": 0.12463935372186959,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999990553924511,
      "loss": 7.2722,
      "mean_token_accuracy": 0.1344312846660614,
      "num_tokens": 2736015.0,
      "step": 1080
    },
    {
      "entropy": 7.196067523956299,
      "epoch": 0.12521638776687824,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999989320380939,
      "loss": 7.221,
      "mean_token_accuracy": 0.1397103875875473,
      "num_tokens": 2747561.0,
      "step": 1085
    },
    {
      "entropy": 7.292922401428223,
      "epoch": 0.1257934218118869,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999988011160094,
      "loss": 7.2605,
      "mean_token_accuracy": 0.1328883707523346,
      "num_tokens": 2758992.0,
      "step": 1090
    },
    {
      "entropy": 7.279149007797241,
      "epoch": 0.12637045585689555,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499998662626202,
      "loss": 7.2514,
      "mean_token_accuracy": 0.13322949036955833,
      "num_tokens": 2771164.0,
      "step": 1095
    },
    {
      "entropy": 7.044620227813721,
      "epoch": 0.12694748990190421,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999985165686764,
      "loss": 7.2315,
      "mean_token_accuracy": 0.1337655045092106,
      "num_tokens": 2784400.0,
      "step": 1100
    },
    {
      "entropy": 7.231164073944091,
      "epoch": 0.12752452394691285,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999983629434373,
      "loss": 7.1957,
      "mean_token_accuracy": 0.13645816892385482,
      "num_tokens": 2795729.0,
      "step": 1105
    },
    {
      "entropy": 7.156470489501953,
      "epoch": 0.12810155799192152,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999982017504901,
      "loss": 7.2668,
      "mean_token_accuracy": 0.13351675048470496,
      "num_tokens": 2808118.0,
      "step": 1110
    },
    {
      "entropy": 7.241878986358643,
      "epoch": 0.1286785920369302,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999980329898401,
      "loss": 7.227,
      "mean_token_accuracy": 0.1321822077035904,
      "num_tokens": 2821118.0,
      "step": 1115
    },
    {
      "entropy": 7.150159502029419,
      "epoch": 0.12925562608193883,
      "grad_norm": 1.296875,
      "learning_rate": 0.000499997856661493,
      "loss": 7.2141,
      "mean_token_accuracy": 0.1326787978410721,
      "num_tokens": 2833636.0,
      "step": 1120
    },
    {
      "entropy": 7.185760831832885,
      "epoch": 0.1298326601269475,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499997672765455,
      "loss": 7.2273,
      "mean_token_accuracy": 0.13806429505348206,
      "num_tokens": 2846617.0,
      "step": 1125
    },
    {
      "entropy": 7.1420997142791744,
      "epoch": 0.13040969417195614,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999974813017318,
      "loss": 7.1761,
      "mean_token_accuracy": 0.14047402441501616,
      "num_tokens": 2859191.0,
      "step": 1130
    },
    {
      "entropy": 7.267583703994751,
      "epoch": 0.1309867282169648,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999972822703301,
      "loss": 7.2047,
      "mean_token_accuracy": 0.13429140374064447,
      "num_tokens": 2872592.0,
      "step": 1135
    },
    {
      "entropy": 7.123564100265503,
      "epoch": 0.13156376226197344,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999970756712567,
      "loss": 7.1782,
      "mean_token_accuracy": 0.13929741457104683,
      "num_tokens": 2884094.0,
      "step": 1140
    },
    {
      "entropy": 7.225417804718018,
      "epoch": 0.1321407963069821,
      "grad_norm": 0.875,
      "learning_rate": 0.0004999968615045183,
      "loss": 7.2401,
      "mean_token_accuracy": 0.13268111422657966,
      "num_tokens": 2896947.0,
      "step": 1145
    },
    {
      "entropy": 7.190128946304322,
      "epoch": 0.13271783035199078,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999966397701222,
      "loss": 7.2378,
      "mean_token_accuracy": 0.13291095346212387,
      "num_tokens": 2909029.0,
      "step": 1150
    },
    {
      "entropy": 7.222978162765503,
      "epoch": 0.13329486439699942,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999964104680759,
      "loss": 7.2202,
      "mean_token_accuracy": 0.1370919555425644,
      "num_tokens": 2920778.0,
      "step": 1155
    },
    {
      "entropy": 7.057752561569214,
      "epoch": 0.13387189844200809,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999961735983871,
      "loss": 7.1546,
      "mean_token_accuracy": 0.1401558116078377,
      "num_tokens": 2933754.0,
      "step": 1160
    },
    {
      "entropy": 7.285988759994507,
      "epoch": 0.13444893248701673,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999959291610639,
      "loss": 7.2337,
      "mean_token_accuracy": 0.1410619392991066,
      "num_tokens": 2946019.0,
      "step": 1165
    },
    {
      "entropy": 7.11617488861084,
      "epoch": 0.1350259665320254,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999956771561143,
      "loss": 7.1804,
      "mean_token_accuracy": 0.13369629979133607,
      "num_tokens": 2959688.0,
      "step": 1170
    },
    {
      "entropy": 7.10788106918335,
      "epoch": 0.13560300057703403,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999954175835469,
      "loss": 7.0996,
      "mean_token_accuracy": 0.1448257327079773,
      "num_tokens": 2971211.0,
      "step": 1175
    },
    {
      "entropy": 7.1773261547088625,
      "epoch": 0.1361800346220427,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999951504433703,
      "loss": 7.2158,
      "mean_token_accuracy": 0.13084226623177528,
      "num_tokens": 2983559.0,
      "step": 1180
    },
    {
      "entropy": 7.126594400405883,
      "epoch": 0.13675706866705137,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999948757355935,
      "loss": 7.168,
      "mean_token_accuracy": 0.13138265088200568,
      "num_tokens": 2995053.0,
      "step": 1185
    },
    {
      "entropy": 7.131065607070923,
      "epoch": 0.13733410271206,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499994593460226,
      "loss": 7.1255,
      "mean_token_accuracy": 0.13711200654506683,
      "num_tokens": 3006143.0,
      "step": 1190
    },
    {
      "entropy": 7.1007500171661375,
      "epoch": 0.13791113675706868,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999943036172771,
      "loss": 7.1097,
      "mean_token_accuracy": 0.1373543344438076,
      "num_tokens": 3018793.0,
      "step": 1195
    },
    {
      "entropy": 7.153094959259033,
      "epoch": 0.13848817080207732,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999940062067565,
      "loss": 7.1354,
      "mean_token_accuracy": 0.13795246854424476,
      "num_tokens": 3031547.0,
      "step": 1200
    },
    {
      "entropy": 7.04004921913147,
      "epoch": 0.13906520484708598,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999937012286742,
      "loss": 7.1537,
      "mean_token_accuracy": 0.13225894793868065,
      "num_tokens": 3044035.0,
      "step": 1205
    },
    {
      "entropy": 7.132466888427734,
      "epoch": 0.13964223889209462,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999933886830405,
      "loss": 7.1779,
      "mean_token_accuracy": 0.14098356738686563,
      "num_tokens": 3056238.0,
      "step": 1210
    },
    {
      "entropy": 7.119182538986206,
      "epoch": 0.1402192729371033,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499993068569866,
      "loss": 7.1719,
      "mean_token_accuracy": 0.1401755228638649,
      "num_tokens": 3069250.0,
      "step": 1215
    },
    {
      "entropy": 7.180674600601196,
      "epoch": 0.14079630698211196,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999927408891614,
      "loss": 7.1522,
      "mean_token_accuracy": 0.13382151871919631,
      "num_tokens": 3082048.0,
      "step": 1220
    },
    {
      "entropy": 7.034345865249634,
      "epoch": 0.1413733410271206,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999924056409378,
      "loss": 7.2487,
      "mean_token_accuracy": 0.13513218089938164,
      "num_tokens": 3094588.0,
      "step": 1225
    },
    {
      "entropy": 7.2652326107025145,
      "epoch": 0.14195037507212926,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999920628252063,
      "loss": 7.2225,
      "mean_token_accuracy": 0.12983713001012803,
      "num_tokens": 3108067.0,
      "step": 1230
    },
    {
      "entropy": 7.027235317230224,
      "epoch": 0.1425274091171379,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999917124419785,
      "loss": 7.1671,
      "mean_token_accuracy": 0.14140997901558877,
      "num_tokens": 3120524.0,
      "step": 1235
    },
    {
      "entropy": 7.2310967445373535,
      "epoch": 0.14310444316214657,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999913544912664,
      "loss": 7.2371,
      "mean_token_accuracy": 0.12993111908435823,
      "num_tokens": 3133533.0,
      "step": 1240
    },
    {
      "entropy": 7.224728441238403,
      "epoch": 0.1436814772071552,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999909889730817,
      "loss": 7.1937,
      "mean_token_accuracy": 0.13602788001298904,
      "num_tokens": 3146183.0,
      "step": 1245
    },
    {
      "entropy": 7.092638444900513,
      "epoch": 0.14425851125216388,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999906158874368,
      "loss": 7.1575,
      "mean_token_accuracy": 0.1345686912536621,
      "num_tokens": 3158436.0,
      "step": 1250
    },
    {
      "entropy": 7.0961510181427006,
      "epoch": 0.14483554529717255,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999902352343444,
      "loss": 7.1168,
      "mean_token_accuracy": 0.14122596234083176,
      "num_tokens": 3170426.0,
      "step": 1255
    },
    {
      "entropy": 7.156266593933106,
      "epoch": 0.1454125793421812,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999898470138171,
      "loss": 7.1086,
      "mean_token_accuracy": 0.1393971025943756,
      "num_tokens": 3182865.0,
      "step": 1260
    },
    {
      "entropy": 7.043174409866333,
      "epoch": 0.14598961338718985,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999894512258681,
      "loss": 7.1419,
      "mean_token_accuracy": 0.13401855304837226,
      "num_tokens": 3195674.0,
      "step": 1265
    },
    {
      "entropy": 7.135420846939087,
      "epoch": 0.1465666474321985,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999890478705107,
      "loss": 7.0969,
      "mean_token_accuracy": 0.14015892669558525,
      "num_tokens": 3208848.0,
      "step": 1270
    },
    {
      "entropy": 7.114065217971802,
      "epoch": 0.14714368147720716,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999886369477585,
      "loss": 7.1859,
      "mean_token_accuracy": 0.13531602919101715,
      "num_tokens": 3221179.0,
      "step": 1275
    },
    {
      "entropy": 7.15838098526001,
      "epoch": 0.1477207155222158,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999882184576251,
      "loss": 7.2691,
      "mean_token_accuracy": 0.12990250065922737,
      "num_tokens": 3233671.0,
      "step": 1280
    },
    {
      "entropy": 7.2466796875,
      "epoch": 0.14829774956722447,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999877924001248,
      "loss": 7.1856,
      "mean_token_accuracy": 0.1305832877755165,
      "num_tokens": 3247219.0,
      "step": 1285
    },
    {
      "entropy": 7.049865818023681,
      "epoch": 0.1488747836122331,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999873587752718,
      "loss": 7.1495,
      "mean_token_accuracy": 0.13346763998270034,
      "num_tokens": 3259444.0,
      "step": 1290
    },
    {
      "entropy": 6.969862937927246,
      "epoch": 0.14945181765724178,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999869175830808,
      "loss": 7.0034,
      "mean_token_accuracy": 0.14465287700295448,
      "num_tokens": 3272092.0,
      "step": 1295
    },
    {
      "entropy": 7.140999794006348,
      "epoch": 0.15002885170225044,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999864688235666,
      "loss": 7.1902,
      "mean_token_accuracy": 0.12994412407279016,
      "num_tokens": 3284260.0,
      "step": 1300
    },
    {
      "entropy": 7.1015129566192625,
      "epoch": 0.15060588574725908,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999860124967442,
      "loss": 7.1399,
      "mean_token_accuracy": 0.1303492307662964,
      "num_tokens": 3296662.0,
      "step": 1305
    },
    {
      "entropy": 7.115274858474732,
      "epoch": 0.15118291979226775,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999855486026291,
      "loss": 7.098,
      "mean_token_accuracy": 0.1382736712694168,
      "num_tokens": 3307782.0,
      "step": 1310
    },
    {
      "entropy": 7.117005920410156,
      "epoch": 0.1517599538372764,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999850771412369,
      "loss": 7.1608,
      "mean_token_accuracy": 0.13215686306357383,
      "num_tokens": 3320519.0,
      "step": 1315
    },
    {
      "entropy": 7.089527368545532,
      "epoch": 0.15233698788228506,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999845981125832,
      "loss": 7.0264,
      "mean_token_accuracy": 0.14284324571490287,
      "num_tokens": 3331722.0,
      "step": 1320
    },
    {
      "entropy": 7.069760704040528,
      "epoch": 0.1529140219272937,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999841115166844,
      "loss": 7.1303,
      "mean_token_accuracy": 0.13235283344984056,
      "num_tokens": 3345143.0,
      "step": 1325
    },
    {
      "entropy": 7.030734205245972,
      "epoch": 0.15349105597230236,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999836173535568,
      "loss": 7.0524,
      "mean_token_accuracy": 0.14473133385181428,
      "num_tokens": 3358946.0,
      "step": 1330
    },
    {
      "entropy": 7.128999328613281,
      "epoch": 0.15406809001731103,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999831156232169,
      "loss": 7.1604,
      "mean_token_accuracy": 0.1378837525844574,
      "num_tokens": 3372465.0,
      "step": 1335
    },
    {
      "entropy": 7.039134931564331,
      "epoch": 0.15464512406231967,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999826063256818,
      "loss": 7.0408,
      "mean_token_accuracy": 0.14155885428190232,
      "num_tokens": 3384392.0,
      "step": 1340
    },
    {
      "entropy": 7.1258001804351805,
      "epoch": 0.15522215810732834,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999820894609683,
      "loss": 7.1384,
      "mean_token_accuracy": 0.1228943333029747,
      "num_tokens": 3398137.0,
      "step": 1345
    },
    {
      "entropy": 7.124754333496094,
      "epoch": 0.15579919215233698,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999815650290941,
      "loss": 7.0811,
      "mean_token_accuracy": 0.14019765704870224,
      "num_tokens": 3410353.0,
      "step": 1350
    },
    {
      "entropy": 7.00805253982544,
      "epoch": 0.15637622619734565,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999810330300766,
      "loss": 7.0326,
      "mean_token_accuracy": 0.13864756077528,
      "num_tokens": 3423206.0,
      "step": 1355
    },
    {
      "entropy": 7.053879833221435,
      "epoch": 0.1569532602423543,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000499980493463934,
      "loss": 7.1112,
      "mean_token_accuracy": 0.1342466413974762,
      "num_tokens": 3436780.0,
      "step": 1360
    },
    {
      "entropy": 7.1501076221466064,
      "epoch": 0.15753029428736295,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999799463306841,
      "loss": 7.1779,
      "mean_token_accuracy": 0.13237905949354173,
      "num_tokens": 3450285.0,
      "step": 1365
    },
    {
      "entropy": 7.058084535598755,
      "epoch": 0.15810732833237162,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999793916303455,
      "loss": 7.0564,
      "mean_token_accuracy": 0.14176423102617264,
      "num_tokens": 3462157.0,
      "step": 1370
    },
    {
      "entropy": 6.970040607452392,
      "epoch": 0.15868436237738026,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999788293629368,
      "loss": 7.0133,
      "mean_token_accuracy": 0.1461627095937729,
      "num_tokens": 3475027.0,
      "step": 1375
    },
    {
      "entropy": 7.019241905212402,
      "epoch": 0.15926139642238893,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499978259528477,
      "loss": 7.1074,
      "mean_token_accuracy": 0.13902512937784195,
      "num_tokens": 3489111.0,
      "step": 1380
    },
    {
      "entropy": 7.062412309646606,
      "epoch": 0.15983843046739757,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999776821269852,
      "loss": 7.0485,
      "mean_token_accuracy": 0.1354866899549961,
      "num_tokens": 3501673.0,
      "step": 1385
    },
    {
      "entropy": 6.9125723361969,
      "epoch": 0.16041546451240624,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999770971584807,
      "loss": 6.972,
      "mean_token_accuracy": 0.14499091804027558,
      "num_tokens": 3513511.0,
      "step": 1390
    },
    {
      "entropy": 7.126408863067627,
      "epoch": 0.16099249855741488,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999765046229834,
      "loss": 7.1008,
      "mean_token_accuracy": 0.13360448479652404,
      "num_tokens": 3525218.0,
      "step": 1395
    },
    {
      "entropy": 7.0275006771087645,
      "epoch": 0.16156953260242354,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499975904520513,
      "loss": 6.9918,
      "mean_token_accuracy": 0.1486022636294365,
      "num_tokens": 3537006.0,
      "step": 1400
    },
    {
      "entropy": 7.0024169921875,
      "epoch": 0.1621465666474322,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00049997529685109,
      "loss": 7.0915,
      "mean_token_accuracy": 0.13563401028513908,
      "num_tokens": 3549873.0,
      "step": 1405
    },
    {
      "entropy": 7.1586450099945065,
      "epoch": 0.16272360069244085,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999746816147344,
      "loss": 7.1743,
      "mean_token_accuracy": 0.1260405257344246,
      "num_tokens": 3563542.0,
      "step": 1410
    },
    {
      "entropy": 6.972772789001465,
      "epoch": 0.16330063473744952,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999740588114673,
      "loss": 7.0563,
      "mean_token_accuracy": 0.13719577193260193,
      "num_tokens": 3575679.0,
      "step": 1415
    },
    {
      "entropy": 7.14780478477478,
      "epoch": 0.16387766878245816,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999734284413095,
      "loss": 7.0748,
      "mean_token_accuracy": 0.1393434651196003,
      "num_tokens": 3588605.0,
      "step": 1420
    },
    {
      "entropy": 6.959795999526977,
      "epoch": 0.16445470282746683,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499972790504282,
      "loss": 7.075,
      "mean_token_accuracy": 0.14719273895025253,
      "num_tokens": 3601894.0,
      "step": 1425
    },
    {
      "entropy": 6.970956611633301,
      "epoch": 0.16503173687247547,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004999721450004067,
      "loss": 7.0138,
      "mean_token_accuracy": 0.14499847069382668,
      "num_tokens": 3614436.0,
      "step": 1430
    },
    {
      "entropy": 7.0098734378814695,
      "epoch": 0.16560877091748413,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999714919297049,
      "loss": 6.9708,
      "mean_token_accuracy": 0.14259159415960312,
      "num_tokens": 3627590.0,
      "step": 1435
    },
    {
      "entropy": 7.064096879959107,
      "epoch": 0.1661858049624928,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004999708312921987,
      "loss": 7.1157,
      "mean_token_accuracy": 0.13143253847956657,
      "num_tokens": 3640834.0,
      "step": 1440
    },
    {
      "entropy": 6.9391326904296875,
      "epoch": 0.16676283900750144,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999701630879103,
      "loss": 6.9743,
      "mean_token_accuracy": 0.1408776670694351,
      "num_tokens": 3653024.0,
      "step": 1445
    },
    {
      "entropy": 7.055517578125,
      "epoch": 0.1673398730525101,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999694873168623,
      "loss": 7.0344,
      "mean_token_accuracy": 0.1418595716357231,
      "num_tokens": 3665311.0,
      "step": 1450
    },
    {
      "entropy": 6.973097419738769,
      "epoch": 0.16791690709751875,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999688039790773,
      "loss": 7.0708,
      "mean_token_accuracy": 0.13803807273507118,
      "num_tokens": 3679102.0,
      "step": 1455
    },
    {
      "entropy": 7.027426385879517,
      "epoch": 0.16849394114252741,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999681130745784,
      "loss": 6.9891,
      "mean_token_accuracy": 0.1482289746403694,
      "num_tokens": 3690046.0,
      "step": 1460
    },
    {
      "entropy": 6.917978143692016,
      "epoch": 0.16907097518753605,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999674146033888,
      "loss": 7.015,
      "mean_token_accuracy": 0.1324767976999283,
      "num_tokens": 3703668.0,
      "step": 1465
    },
    {
      "entropy": 7.034569263458252,
      "epoch": 0.16964800923254472,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004999667085655318,
      "loss": 7.0303,
      "mean_token_accuracy": 0.14069233685731888,
      "num_tokens": 3715852.0,
      "step": 1470
    },
    {
      "entropy": 6.9502979755401615,
      "epoch": 0.1702250432775534,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999659949610313,
      "loss": 6.9468,
      "mean_token_accuracy": 0.14053553566336632,
      "num_tokens": 3728310.0,
      "step": 1475
    },
    {
      "entropy": 7.0809468746185305,
      "epoch": 0.17080207732256203,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999652737899114,
      "loss": 7.0286,
      "mean_token_accuracy": 0.1336084134876728,
      "num_tokens": 3741323.0,
      "step": 1480
    },
    {
      "entropy": 6.936049604415894,
      "epoch": 0.1713791113675707,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999645450521963,
      "loss": 7.0266,
      "mean_token_accuracy": 0.13839906752109526,
      "num_tokens": 3752667.0,
      "step": 1485
    },
    {
      "entropy": 7.046130800247193,
      "epoch": 0.17195614541257934,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999638087479104,
      "loss": 7.0423,
      "mean_token_accuracy": 0.1376166321337223,
      "num_tokens": 3764282.0,
      "step": 1490
    },
    {
      "entropy": 6.975060272216797,
      "epoch": 0.172533179457588,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999630648770786,
      "loss": 7.0681,
      "mean_token_accuracy": 0.13423443585634232,
      "num_tokens": 3778280.0,
      "step": 1495
    },
    {
      "entropy": 6.946654510498047,
      "epoch": 0.17311021350259664,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999623134397257,
      "loss": 7.0722,
      "mean_token_accuracy": 0.1414807364344597,
      "num_tokens": 3791010.0,
      "step": 1500
    },
    {
      "entropy": 7.078482675552368,
      "epoch": 0.1736872475476053,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999615544358774,
      "loss": 7.0158,
      "mean_token_accuracy": 0.13697310537099838,
      "num_tokens": 3802738.0,
      "step": 1505
    },
    {
      "entropy": 6.945161008834839,
      "epoch": 0.17426428159261395,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999607878655587,
      "loss": 7.0416,
      "mean_token_accuracy": 0.13953944891691208,
      "num_tokens": 3815448.0,
      "step": 1510
    },
    {
      "entropy": 7.016398239135742,
      "epoch": 0.17484131563762262,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004999600137287958,
      "loss": 7.0208,
      "mean_token_accuracy": 0.1362545184791088,
      "num_tokens": 3829833.0,
      "step": 1515
    },
    {
      "entropy": 6.95634651184082,
      "epoch": 0.17541834968263129,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999592320256146,
      "loss": 6.9977,
      "mean_token_accuracy": 0.1375351645052433,
      "num_tokens": 3841544.0,
      "step": 1520
    },
    {
      "entropy": 7.058304262161255,
      "epoch": 0.17599538372763993,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999584427560412,
      "loss": 6.9914,
      "mean_token_accuracy": 0.1338309071958065,
      "num_tokens": 3853886.0,
      "step": 1525
    },
    {
      "entropy": 6.900669813156128,
      "epoch": 0.1765724177726486,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999576459201024,
      "loss": 7.0318,
      "mean_token_accuracy": 0.14436644166707993,
      "num_tokens": 3867942.0,
      "step": 1530
    },
    {
      "entropy": 6.937264537811279,
      "epoch": 0.17714945181765723,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499956841517825,
      "loss": 7.0233,
      "mean_token_accuracy": 0.1413443475961685,
      "num_tokens": 3881764.0,
      "step": 1535
    },
    {
      "entropy": 6.968937110900879,
      "epoch": 0.1777264858626659,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499956029549236,
      "loss": 6.9303,
      "mean_token_accuracy": 0.14205588772892952,
      "num_tokens": 3893979.0,
      "step": 1540
    },
    {
      "entropy": 6.96471152305603,
      "epoch": 0.17830351990767454,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999552100143625,
      "loss": 7.0387,
      "mean_token_accuracy": 0.13929225131869316,
      "num_tokens": 3905934.0,
      "step": 1545
    },
    {
      "entropy": 7.040468406677246,
      "epoch": 0.1788805539526832,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999543829132324,
      "loss": 6.9983,
      "mean_token_accuracy": 0.14421933889389038,
      "num_tokens": 3918348.0,
      "step": 1550
    },
    {
      "entropy": 6.9538825988769535,
      "epoch": 0.17945758799769188,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999535482458734,
      "loss": 6.973,
      "mean_token_accuracy": 0.1384164996445179,
      "num_tokens": 3931863.0,
      "step": 1555
    },
    {
      "entropy": 6.961157894134521,
      "epoch": 0.18003462204270052,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999527060123135,
      "loss": 6.9652,
      "mean_token_accuracy": 0.1431449018418789,
      "num_tokens": 3945108.0,
      "step": 1560
    },
    {
      "entropy": 6.975248908996582,
      "epoch": 0.18061165608770918,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999518562125811,
      "loss": 6.9944,
      "mean_token_accuracy": 0.13739961683750151,
      "num_tokens": 3956695.0,
      "step": 1565
    },
    {
      "entropy": 6.972767305374146,
      "epoch": 0.18118869013271782,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004999509988467047,
      "loss": 6.9805,
      "mean_token_accuracy": 0.1392670303583145,
      "num_tokens": 3970458.0,
      "step": 1570
    },
    {
      "entropy": 6.973606538772583,
      "epoch": 0.1817657241777265,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999501339147132,
      "loss": 6.9695,
      "mean_token_accuracy": 0.13941276222467422,
      "num_tokens": 3981962.0,
      "step": 1575
    },
    {
      "entropy": 7.010713863372803,
      "epoch": 0.18234275822273513,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999492614166357,
      "loss": 6.9822,
      "mean_token_accuracy": 0.13373472541570663,
      "num_tokens": 3993512.0,
      "step": 1580
    },
    {
      "entropy": 6.8247644901275635,
      "epoch": 0.1829197922677438,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999483813525014,
      "loss": 6.875,
      "mean_token_accuracy": 0.15325912684202195,
      "num_tokens": 4005826.0,
      "step": 1585
    },
    {
      "entropy": 6.960187387466431,
      "epoch": 0.18349682631275246,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999474937223403,
      "loss": 6.9475,
      "mean_token_accuracy": 0.13995122984051706,
      "num_tokens": 4018927.0,
      "step": 1590
    },
    {
      "entropy": 6.905779933929443,
      "epoch": 0.1840738603577611,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004999465985261818,
      "loss": 6.9818,
      "mean_token_accuracy": 0.13842669874429703,
      "num_tokens": 4030970.0,
      "step": 1595
    },
    {
      "entropy": 6.920939159393311,
      "epoch": 0.18465089440276977,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999456957640562,
      "loss": 6.9621,
      "mean_token_accuracy": 0.14341214746236802,
      "num_tokens": 4043396.0,
      "step": 1600
    },
    {
      "entropy": 6.979000473022461,
      "epoch": 0.1852279284477784,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999447854359939,
      "loss": 7.0026,
      "mean_token_accuracy": 0.13570645451545715,
      "num_tokens": 4055588.0,
      "step": 1605
    },
    {
      "entropy": 6.951703596115112,
      "epoch": 0.18580496249278708,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999438675420255,
      "loss": 6.9723,
      "mean_token_accuracy": 0.13750323951244353,
      "num_tokens": 4068036.0,
      "step": 1610
    },
    {
      "entropy": 6.912143850326538,
      "epoch": 0.18638199653779572,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999429420821818,
      "loss": 7.0134,
      "mean_token_accuracy": 0.13561285212635993,
      "num_tokens": 4080531.0,
      "step": 1615
    },
    {
      "entropy": 6.996648979187012,
      "epoch": 0.1869590305828044,
      "grad_norm": 0.875,
      "learning_rate": 0.000499942009056494,
      "loss": 6.9186,
      "mean_token_accuracy": 0.14655348733067514,
      "num_tokens": 4093614.0,
      "step": 1620
    },
    {
      "entropy": 6.9176637649536135,
      "epoch": 0.18753606462781305,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999410684649935,
      "loss": 7.0373,
      "mean_token_accuracy": 0.13838913813233375,
      "num_tokens": 4106836.0,
      "step": 1625
    },
    {
      "entropy": 7.0432507514953615,
      "epoch": 0.1881130986728217,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999401203077119,
      "loss": 6.9509,
      "mean_token_accuracy": 0.14006586968898774,
      "num_tokens": 4118739.0,
      "step": 1630
    },
    {
      "entropy": 6.852195692062378,
      "epoch": 0.18869013271783036,
      "grad_norm": 0.921875,
      "learning_rate": 0.000499939164584681,
      "loss": 6.9308,
      "mean_token_accuracy": 0.14347120001912117,
      "num_tokens": 4131174.0,
      "step": 1635
    },
    {
      "entropy": 7.005696773529053,
      "epoch": 0.189267166762839,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999382012959331,
      "loss": 6.9709,
      "mean_token_accuracy": 0.133350370824337,
      "num_tokens": 4143459.0,
      "step": 1640
    },
    {
      "entropy": 6.931919956207276,
      "epoch": 0.18984420080784767,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999372304415005,
      "loss": 6.9939,
      "mean_token_accuracy": 0.13698216155171394,
      "num_tokens": 4155864.0,
      "step": 1645
    },
    {
      "entropy": 6.837768316268921,
      "epoch": 0.1904212348528563,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999362520214159,
      "loss": 6.7786,
      "mean_token_accuracy": 0.14995110183954238,
      "num_tokens": 4168433.0,
      "step": 1650
    },
    {
      "entropy": 6.960832262039185,
      "epoch": 0.19099826889786498,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999352660357122,
      "loss": 6.9764,
      "mean_token_accuracy": 0.13728797063231468,
      "num_tokens": 4180852.0,
      "step": 1655
    },
    {
      "entropy": 6.950531387329102,
      "epoch": 0.19157530294287364,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999342724844226,
      "loss": 6.9155,
      "mean_token_accuracy": 0.14255120754241943,
      "num_tokens": 4194712.0,
      "step": 1660
    },
    {
      "entropy": 6.91253252029419,
      "epoch": 0.19215233698788228,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999332713675804,
      "loss": 6.9909,
      "mean_token_accuracy": 0.14385711327195166,
      "num_tokens": 4207970.0,
      "step": 1665
    },
    {
      "entropy": 6.85596489906311,
      "epoch": 0.19272937103289095,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999322626852193,
      "loss": 6.9123,
      "mean_token_accuracy": 0.14430793896317481,
      "num_tokens": 4220893.0,
      "step": 1670
    },
    {
      "entropy": 6.888790082931519,
      "epoch": 0.1933064050778996,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999312464373734,
      "loss": 6.9127,
      "mean_token_accuracy": 0.14029934853315354,
      "num_tokens": 4232463.0,
      "step": 1675
    },
    {
      "entropy": 6.904188966751098,
      "epoch": 0.19388343912290826,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999302226240767,
      "loss": 6.8553,
      "mean_token_accuracy": 0.13843169063329697,
      "num_tokens": 4244604.0,
      "step": 1680
    },
    {
      "entropy": 6.918503475189209,
      "epoch": 0.1944604731679169,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999291912453637,
      "loss": 6.9512,
      "mean_token_accuracy": 0.14121533036231995,
      "num_tokens": 4257260.0,
      "step": 1685
    },
    {
      "entropy": 6.981924057006836,
      "epoch": 0.19503750721292556,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999281523012691,
      "loss": 6.8856,
      "mean_token_accuracy": 0.14021847546100616,
      "num_tokens": 4269885.0,
      "step": 1690
    },
    {
      "entropy": 6.918470287322998,
      "epoch": 0.1956145412579342,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999271057918278,
      "loss": 7.0177,
      "mean_token_accuracy": 0.137399822473526,
      "num_tokens": 4281560.0,
      "step": 1695
    },
    {
      "entropy": 6.923596143722534,
      "epoch": 0.19619157530294287,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999260517170751,
      "loss": 6.8986,
      "mean_token_accuracy": 0.1467658117413521,
      "num_tokens": 4293746.0,
      "step": 1700
    },
    {
      "entropy": 6.947674751281738,
      "epoch": 0.19676860934795154,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999249900770463,
      "loss": 6.9248,
      "mean_token_accuracy": 0.1379011631011963,
      "num_tokens": 4306595.0,
      "step": 1705
    },
    {
      "entropy": 6.960206651687622,
      "epoch": 0.19734564339296018,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999239208717772,
      "loss": 6.8912,
      "mean_token_accuracy": 0.1443597435951233,
      "num_tokens": 4318705.0,
      "step": 1710
    },
    {
      "entropy": 6.834646701812744,
      "epoch": 0.19792267743796885,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999228441013037,
      "loss": 6.9075,
      "mean_token_accuracy": 0.14063191488385202,
      "num_tokens": 4331321.0,
      "step": 1715
    },
    {
      "entropy": 6.875298166275025,
      "epoch": 0.1984997114829775,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499921759765662,
      "loss": 6.875,
      "mean_token_accuracy": 0.14705403074622153,
      "num_tokens": 4343517.0,
      "step": 1720
    },
    {
      "entropy": 6.939173650741577,
      "epoch": 0.19907674552798615,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999206678648888,
      "loss": 6.9152,
      "mean_token_accuracy": 0.14792341142892837,
      "num_tokens": 4356342.0,
      "step": 1725
    },
    {
      "entropy": 6.834136724472046,
      "epoch": 0.1996537795729948,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999195683990206,
      "loss": 6.9055,
      "mean_token_accuracy": 0.14260428175330161,
      "num_tokens": 4369078.0,
      "step": 1730
    },
    {
      "entropy": 6.930724620819092,
      "epoch": 0.20023081361800346,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999184613680945,
      "loss": 6.8893,
      "mean_token_accuracy": 0.13664604276418685,
      "num_tokens": 4382091.0,
      "step": 1735
    },
    {
      "entropy": 6.834902429580689,
      "epoch": 0.20080784766301213,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999173467721476,
      "loss": 6.916,
      "mean_token_accuracy": 0.13967763930559157,
      "num_tokens": 4393919.0,
      "step": 1740
    },
    {
      "entropy": 6.9305174350738525,
      "epoch": 0.20138488170802077,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004999162246112175,
      "loss": 6.8841,
      "mean_token_accuracy": 0.14394750744104384,
      "num_tokens": 4407054.0,
      "step": 1745
    },
    {
      "entropy": 6.847874784469605,
      "epoch": 0.20196191575302944,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004999150948853419,
      "loss": 6.9343,
      "mean_token_accuracy": 0.13633092492818832,
      "num_tokens": 4420180.0,
      "step": 1750
    },
    {
      "entropy": 7.021355628967285,
      "epoch": 0.20253894979803808,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999139575945587,
      "loss": 6.9167,
      "mean_token_accuracy": 0.15015274733304979,
      "num_tokens": 4433248.0,
      "step": 1755
    },
    {
      "entropy": 6.800896453857422,
      "epoch": 0.20311598384304674,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999128127389065,
      "loss": 6.9057,
      "mean_token_accuracy": 0.14164429455995559,
      "num_tokens": 4446340.0,
      "step": 1760
    },
    {
      "entropy": 6.900873231887817,
      "epoch": 0.20369301788805538,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999116603184233,
      "loss": 6.9154,
      "mean_token_accuracy": 0.14202600717544556,
      "num_tokens": 4459918.0,
      "step": 1765
    },
    {
      "entropy": 6.862395143508911,
      "epoch": 0.20427005193306405,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999105003331482,
      "loss": 6.8724,
      "mean_token_accuracy": 0.14486078545451164,
      "num_tokens": 4472020.0,
      "step": 1770
    },
    {
      "entropy": 6.871878528594971,
      "epoch": 0.20484708597807272,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999093327831202,
      "loss": 6.844,
      "mean_token_accuracy": 0.14406751170754434,
      "num_tokens": 4484269.0,
      "step": 1775
    },
    {
      "entropy": 6.906937932968139,
      "epoch": 0.20542412002308136,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999081576683784,
      "loss": 6.944,
      "mean_token_accuracy": 0.13962593898177147,
      "num_tokens": 4498040.0,
      "step": 1780
    },
    {
      "entropy": 6.899182319641113,
      "epoch": 0.20600115406809003,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004999069749889626,
      "loss": 6.8898,
      "mean_token_accuracy": 0.14287936463952064,
      "num_tokens": 4509701.0,
      "step": 1785
    },
    {
      "entropy": 6.898169612884521,
      "epoch": 0.20657818811309867,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999057847449123,
      "loss": 6.9164,
      "mean_token_accuracy": 0.14265610203146933,
      "num_tokens": 4521840.0,
      "step": 1790
    },
    {
      "entropy": 6.880409812927246,
      "epoch": 0.20715522215810733,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999045869362678,
      "loss": 6.9135,
      "mean_token_accuracy": 0.141028805822134,
      "num_tokens": 4534241.0,
      "step": 1795
    },
    {
      "entropy": 6.891494417190552,
      "epoch": 0.20773225620311597,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499903381563069,
      "loss": 6.8726,
      "mean_token_accuracy": 0.14260033294558525,
      "num_tokens": 4547255.0,
      "step": 1800
    },
    {
      "entropy": 6.858333587646484,
      "epoch": 0.20830929024812464,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999021686253568,
      "loss": 6.8877,
      "mean_token_accuracy": 0.14118088632822037,
      "num_tokens": 4559538.0,
      "step": 1805
    },
    {
      "entropy": 6.860760068893432,
      "epoch": 0.2088863242931333,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999009481231719,
      "loss": 6.8935,
      "mean_token_accuracy": 0.14185215309262275,
      "num_tokens": 4571444.0,
      "step": 1810
    },
    {
      "entropy": 6.703266668319702,
      "epoch": 0.20946335833814195,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004998997200565553,
      "loss": 6.7323,
      "mean_token_accuracy": 0.15689657777547836,
      "num_tokens": 4586294.0,
      "step": 1815
    },
    {
      "entropy": 6.907558965682983,
      "epoch": 0.21004039238315061,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998984844255483,
      "loss": 6.8867,
      "mean_token_accuracy": 0.14380680844187738,
      "num_tokens": 4599661.0,
      "step": 1820
    },
    {
      "entropy": 6.882632923126221,
      "epoch": 0.21061742642815925,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004998972412301925,
      "loss": 6.892,
      "mean_token_accuracy": 0.14093836694955825,
      "num_tokens": 4611305.0,
      "step": 1825
    },
    {
      "entropy": 6.85354471206665,
      "epoch": 0.21119446047316792,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004998959904705297,
      "loss": 6.9001,
      "mean_token_accuracy": 0.13364512771368026,
      "num_tokens": 4623992.0,
      "step": 1830
    },
    {
      "entropy": 6.925228023529053,
      "epoch": 0.21177149451817656,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998947321466021,
      "loss": 6.8803,
      "mean_token_accuracy": 0.13986791148781777,
      "num_tokens": 4636495.0,
      "step": 1835
    },
    {
      "entropy": 6.903949499130249,
      "epoch": 0.21234852856318523,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004998934662584518,
      "loss": 6.9039,
      "mean_token_accuracy": 0.13886573910713196,
      "num_tokens": 4650266.0,
      "step": 1840
    },
    {
      "entropy": 6.853269004821778,
      "epoch": 0.2129255626081939,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998921928061214,
      "loss": 6.859,
      "mean_token_accuracy": 0.14313986301422119,
      "num_tokens": 4663238.0,
      "step": 1845
    },
    {
      "entropy": 6.901078176498413,
      "epoch": 0.21350259665320254,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004998909117896539,
      "loss": 6.8956,
      "mean_token_accuracy": 0.1408998563885689,
      "num_tokens": 4675501.0,
      "step": 1850
    },
    {
      "entropy": 6.818858623504639,
      "epoch": 0.2140796306982112,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998896232090922,
      "loss": 6.8711,
      "mean_token_accuracy": 0.13704144060611725,
      "num_tokens": 4688996.0,
      "step": 1855
    },
    {
      "entropy": 6.915656042098999,
      "epoch": 0.21465666474321984,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004998883270644798,
      "loss": 6.9764,
      "mean_token_accuracy": 0.13644033223390578,
      "num_tokens": 4703226.0,
      "step": 1860
    },
    {
      "entropy": 6.90679931640625,
      "epoch": 0.2152336987882285,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004998870233558602,
      "loss": 6.8537,
      "mean_token_accuracy": 0.13797650784254073,
      "num_tokens": 4715825.0,
      "step": 1865
    },
    {
      "entropy": 6.776429557800293,
      "epoch": 0.21581073283323715,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998857120832774,
      "loss": 6.8472,
      "mean_token_accuracy": 0.14446549713611603,
      "num_tokens": 4727819.0,
      "step": 1870
    },
    {
      "entropy": 6.890902948379517,
      "epoch": 0.21638776687824582,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004998843932467751,
      "loss": 6.8847,
      "mean_token_accuracy": 0.13867277577519416,
      "num_tokens": 4742890.0,
      "step": 1875
    },
    {
      "entropy": 6.844753170013428,
      "epoch": 0.21696480092325446,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998830668463982,
      "loss": 6.868,
      "mean_token_accuracy": 0.1419196143746376,
      "num_tokens": 4755609.0,
      "step": 1880
    },
    {
      "entropy": 6.797444152832031,
      "epoch": 0.21754183496826313,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004998817328821909,
      "loss": 6.8404,
      "mean_token_accuracy": 0.1449926421046257,
      "num_tokens": 4766674.0,
      "step": 1885
    },
    {
      "entropy": 6.869100189208984,
      "epoch": 0.2181188690132718,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998803913541983,
      "loss": 6.8862,
      "mean_token_accuracy": 0.13758110478520394,
      "num_tokens": 4779639.0,
      "step": 1890
    },
    {
      "entropy": 7.015814256668091,
      "epoch": 0.21869590305828043,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998790422624654,
      "loss": 6.9639,
      "mean_token_accuracy": 0.13711576908826828,
      "num_tokens": 4791702.0,
      "step": 1895
    },
    {
      "entropy": 6.843172836303711,
      "epoch": 0.2192729371032891,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004998776856070376,
      "loss": 6.7892,
      "mean_token_accuracy": 0.14893252551555633,
      "num_tokens": 4805756.0,
      "step": 1900
    },
    {
      "entropy": 6.778265953063965,
      "epoch": 0.21984997114829774,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998763213879606,
      "loss": 6.8403,
      "mean_token_accuracy": 0.1457715541124344,
      "num_tokens": 4817941.0,
      "step": 1905
    },
    {
      "entropy": 6.812149238586426,
      "epoch": 0.2204270051933064,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998749496052803,
      "loss": 6.7324,
      "mean_token_accuracy": 0.14796641916036607,
      "num_tokens": 4832298.0,
      "step": 1910
    },
    {
      "entropy": 6.744766473770142,
      "epoch": 0.22100403923831505,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998735702590426,
      "loss": 6.8345,
      "mean_token_accuracy": 0.14761534780263902,
      "num_tokens": 4844814.0,
      "step": 1915
    },
    {
      "entropy": 6.905437231063843,
      "epoch": 0.22158107328332372,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004998721833492942,
      "loss": 6.8911,
      "mean_token_accuracy": 0.14497228264808654,
      "num_tokens": 4857888.0,
      "step": 1920
    },
    {
      "entropy": 6.804248714447022,
      "epoch": 0.22215810732833238,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004998707888760816,
      "loss": 6.7916,
      "mean_token_accuracy": 0.14039909094572067,
      "num_tokens": 4870479.0,
      "step": 1925
    },
    {
      "entropy": 6.816269254684448,
      "epoch": 0.22273514137334102,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004998693868394516,
      "loss": 6.784,
      "mean_token_accuracy": 0.14926855266094208,
      "num_tokens": 4883294.0,
      "step": 1930
    },
    {
      "entropy": 6.94655408859253,
      "epoch": 0.2233121754183497,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004998679772394516,
      "loss": 6.9359,
      "mean_token_accuracy": 0.13703567907214165,
      "num_tokens": 4896331.0,
      "step": 1935
    },
    {
      "entropy": 6.7918534755706785,
      "epoch": 0.22388920946335833,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004998665600761289,
      "loss": 6.8752,
      "mean_token_accuracy": 0.14121760800480843,
      "num_tokens": 4908857.0,
      "step": 1940
    },
    {
      "entropy": 6.851579332351685,
      "epoch": 0.224466243508367,
      "grad_norm": 1.0,
      "learning_rate": 0.000499865135349531,
      "loss": 6.8359,
      "mean_token_accuracy": 0.14525341987609863,
      "num_tokens": 4921564.0,
      "step": 1945
    },
    {
      "entropy": 6.8388258934021,
      "epoch": 0.22504327755337564,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004998637030597061,
      "loss": 6.9006,
      "mean_token_accuracy": 0.1439589023590088,
      "num_tokens": 4934960.0,
      "step": 1950
    },
    {
      "entropy": 6.740708255767823,
      "epoch": 0.2256203115983843,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998622632067022,
      "loss": 6.7348,
      "mean_token_accuracy": 0.14645355492830275,
      "num_tokens": 4946942.0,
      "step": 1955
    },
    {
      "entropy": 6.884095239639282,
      "epoch": 0.22619734564339297,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998608157905678,
      "loss": 6.8547,
      "mean_token_accuracy": 0.137442384660244,
      "num_tokens": 4960751.0,
      "step": 1960
    },
    {
      "entropy": 6.83891282081604,
      "epoch": 0.2267743796884016,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998593608113515,
      "loss": 6.8662,
      "mean_token_accuracy": 0.13838402479887008,
      "num_tokens": 4973551.0,
      "step": 1965
    },
    {
      "entropy": 6.7111108779907225,
      "epoch": 0.22735141373341028,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998578982691024,
      "loss": 6.884,
      "mean_token_accuracy": 0.14016093909740449,
      "num_tokens": 4986224.0,
      "step": 1970
    },
    {
      "entropy": 6.898041009902954,
      "epoch": 0.22792844777841892,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004998564281638694,
      "loss": 6.8416,
      "mean_token_accuracy": 0.14575736969709396,
      "num_tokens": 4999874.0,
      "step": 1975
    },
    {
      "entropy": 6.8169921875,
      "epoch": 0.2285054818234276,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004998549504957023,
      "loss": 6.8364,
      "mean_token_accuracy": 0.14460800364613532,
      "num_tokens": 5013106.0,
      "step": 1980
    },
    {
      "entropy": 6.807040786743164,
      "epoch": 0.22908251586843623,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004998534652646506,
      "loss": 6.8062,
      "mean_token_accuracy": 0.13764368891716003,
      "num_tokens": 5026145.0,
      "step": 1985
    },
    {
      "entropy": 6.798084878921509,
      "epoch": 0.2296595499134449,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004998519724707642,
      "loss": 6.8203,
      "mean_token_accuracy": 0.14574942290782927,
      "num_tokens": 5039216.0,
      "step": 1990
    },
    {
      "entropy": 6.829916763305664,
      "epoch": 0.23023658395845356,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004998504721140934,
      "loss": 6.7989,
      "mean_token_accuracy": 0.14435165673494338,
      "num_tokens": 5051956.0,
      "step": 1995
    },
    {
      "entropy": 6.8158831119537355,
      "epoch": 0.2308136180034622,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004998489641946887,
      "loss": 6.8865,
      "mean_token_accuracy": 0.14350505471229552,
      "num_tokens": 5065325.0,
      "step": 2000
    },
    {
      "entropy": 6.828835344314575,
      "epoch": 0.23139065204847087,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004998474487126009,
      "loss": 6.8822,
      "mean_token_accuracy": 0.14764170944690705,
      "num_tokens": 5077776.0,
      "step": 2005
    },
    {
      "entropy": 6.852951622009277,
      "epoch": 0.2319676860934795,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004998459256678806,
      "loss": 6.8817,
      "mean_token_accuracy": 0.1433338388800621,
      "num_tokens": 5091167.0,
      "step": 2010
    },
    {
      "entropy": 6.881577014923096,
      "epoch": 0.23254472013848818,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004998443950605796,
      "loss": 6.8069,
      "mean_token_accuracy": 0.14394941478967666,
      "num_tokens": 5103321.0,
      "step": 2015
    },
    {
      "entropy": 6.797431755065918,
      "epoch": 0.23312175418349682,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004998428568907488,
      "loss": 6.8253,
      "mean_token_accuracy": 0.14137339890003203,
      "num_tokens": 5116999.0,
      "step": 2020
    },
    {
      "entropy": 6.807014989852905,
      "epoch": 0.23369878822850548,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998413111584403,
      "loss": 6.8226,
      "mean_token_accuracy": 0.14501554518938065,
      "num_tokens": 5130429.0,
      "step": 2025
    },
    {
      "entropy": 6.7783522605896,
      "epoch": 0.23427582227351415,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998397578637059,
      "loss": 6.762,
      "mean_token_accuracy": 0.14727519750595092,
      "num_tokens": 5142728.0,
      "step": 2030
    },
    {
      "entropy": 6.840787601470947,
      "epoch": 0.2348528563185228,
      "grad_norm": 0.9375,
      "learning_rate": 0.000499838197006598,
      "loss": 6.873,
      "mean_token_accuracy": 0.14228396862745285,
      "num_tokens": 5155105.0,
      "step": 2035
    },
    {
      "entropy": 6.886496639251709,
      "epoch": 0.23542989036353146,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499836628587169,
      "loss": 6.8909,
      "mean_token_accuracy": 0.13143049255013467,
      "num_tokens": 5169206.0,
      "step": 2040
    },
    {
      "entropy": 6.709402132034302,
      "epoch": 0.2360069244085401,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004998350526054716,
      "loss": 6.7657,
      "mean_token_accuracy": 0.14668394550681113,
      "num_tokens": 5182007.0,
      "step": 2045
    },
    {
      "entropy": 6.844460344314575,
      "epoch": 0.23658395845354876,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998334690615591,
      "loss": 6.797,
      "mean_token_accuracy": 0.1384980671107769,
      "num_tokens": 5195137.0,
      "step": 2050
    },
    {
      "entropy": 6.796845769882202,
      "epoch": 0.2371609924985574,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004998318779554844,
      "loss": 6.8063,
      "mean_token_accuracy": 0.1452823057770729,
      "num_tokens": 5207713.0,
      "step": 2055
    },
    {
      "entropy": 6.778965282440185,
      "epoch": 0.23773802654356607,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998302792873012,
      "loss": 6.7537,
      "mean_token_accuracy": 0.1435099482536316,
      "num_tokens": 5220125.0,
      "step": 2060
    },
    {
      "entropy": 6.790195322036743,
      "epoch": 0.23831506058857474,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004998286730570631,
      "loss": 6.8133,
      "mean_token_accuracy": 0.14231673255562782,
      "num_tokens": 5232142.0,
      "step": 2065
    },
    {
      "entropy": 6.807083463668823,
      "epoch": 0.23889209463358338,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998270592648245,
      "loss": 6.8058,
      "mean_token_accuracy": 0.14608865603804588,
      "num_tokens": 5245517.0,
      "step": 2070
    },
    {
      "entropy": 6.799246835708618,
      "epoch": 0.23946912867859205,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004998254379106394,
      "loss": 6.7838,
      "mean_token_accuracy": 0.14370430558919906,
      "num_tokens": 5257862.0,
      "step": 2075
    },
    {
      "entropy": 6.738052463531494,
      "epoch": 0.2400461627236007,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998238089945622,
      "loss": 6.7372,
      "mean_token_accuracy": 0.1498562604188919,
      "num_tokens": 5270141.0,
      "step": 2080
    },
    {
      "entropy": 6.822539758682251,
      "epoch": 0.24062319676860935,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004998221725166479,
      "loss": 6.7559,
      "mean_token_accuracy": 0.15351167917251587,
      "num_tokens": 5283143.0,
      "step": 2085
    },
    {
      "entropy": 6.763333368301391,
      "epoch": 0.241200230813618,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004998205284769516,
      "loss": 6.8284,
      "mean_token_accuracy": 0.14074681848287582,
      "num_tokens": 5295340.0,
      "step": 2090
    },
    {
      "entropy": 6.8454938411712645,
      "epoch": 0.24177726485862666,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998188768755285,
      "loss": 6.7681,
      "mean_token_accuracy": 0.14496009051799774,
      "num_tokens": 5306821.0,
      "step": 2095
    },
    {
      "entropy": 6.715377140045166,
      "epoch": 0.2423542989036353,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004998172177124341,
      "loss": 6.7732,
      "mean_token_accuracy": 0.1452890232205391,
      "num_tokens": 5319428.0,
      "step": 2100
    },
    {
      "entropy": 6.903585720062256,
      "epoch": 0.24293133294864397,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004998155509877242,
      "loss": 6.8206,
      "mean_token_accuracy": 0.1386203348636627,
      "num_tokens": 5331777.0,
      "step": 2105
    },
    {
      "entropy": 6.615080881118774,
      "epoch": 0.24350836699365264,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000499813876701455,
      "loss": 6.7726,
      "mean_token_accuracy": 0.14410800784826278,
      "num_tokens": 5344533.0,
      "step": 2110
    },
    {
      "entropy": 6.7770771980285645,
      "epoch": 0.24408540103866128,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998121948536828,
      "loss": 6.718,
      "mean_token_accuracy": 0.1547897882759571,
      "num_tokens": 5357123.0,
      "step": 2115
    },
    {
      "entropy": 6.803248691558838,
      "epoch": 0.24466243508366994,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004998105054444639,
      "loss": 6.8454,
      "mean_token_accuracy": 0.14135158509016038,
      "num_tokens": 5370123.0,
      "step": 2120
    },
    {
      "entropy": 6.810746097564698,
      "epoch": 0.24523946912867858,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004998088084738555,
      "loss": 6.7767,
      "mean_token_accuracy": 0.15171893686056137,
      "num_tokens": 5382870.0,
      "step": 2125
    },
    {
      "entropy": 6.694212627410889,
      "epoch": 0.24581650317368725,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998071039419144,
      "loss": 6.761,
      "mean_token_accuracy": 0.1418367989361286,
      "num_tokens": 5395052.0,
      "step": 2130
    },
    {
      "entropy": 6.8836406707763675,
      "epoch": 0.2463935372186959,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000499805391848698,
      "loss": 6.7966,
      "mean_token_accuracy": 0.1489194765686989,
      "num_tokens": 5406720.0,
      "step": 2135
    },
    {
      "entropy": 6.736827850341797,
      "epoch": 0.24697057126370456,
      "grad_norm": 0.890625,
      "learning_rate": 0.000499803672194264,
      "loss": 6.7875,
      "mean_token_accuracy": 0.14699700772762297,
      "num_tokens": 5419465.0,
      "step": 2140
    },
    {
      "entropy": 6.800150632858276,
      "epoch": 0.24754760530871323,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004998019449786701,
      "loss": 6.7748,
      "mean_token_accuracy": 0.13936677277088166,
      "num_tokens": 5432434.0,
      "step": 2145
    },
    {
      "entropy": 6.758311748504639,
      "epoch": 0.24812463935372187,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004998002102019744,
      "loss": 6.7772,
      "mean_token_accuracy": 0.1484916940331459,
      "num_tokens": 5445249.0,
      "step": 2150
    },
    {
      "entropy": 6.888931179046631,
      "epoch": 0.24870167339873053,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004997984678642354,
      "loss": 6.7955,
      "mean_token_accuracy": 0.14284422919154166,
      "num_tokens": 5458996.0,
      "step": 2155
    },
    {
      "entropy": 6.7189394474029545,
      "epoch": 0.24927870744373917,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004997967179655115,
      "loss": 6.7892,
      "mean_token_accuracy": 0.13899215385317804,
      "num_tokens": 5472174.0,
      "step": 2160
    },
    {
      "entropy": 6.748557662963867,
      "epoch": 0.24985574148874784,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004997949605058617,
      "loss": 6.7129,
      "mean_token_accuracy": 0.1528759181499481,
      "num_tokens": 5484464.0,
      "step": 2165
    },
    {
      "entropy": 6.79465274810791,
      "epoch": 0.2504327755337565,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997931954853451,
      "loss": 6.8143,
      "mean_token_accuracy": 0.14372445791959762,
      "num_tokens": 5497278.0,
      "step": 2170
    },
    {
      "entropy": 6.7927593231201175,
      "epoch": 0.2510098095787651,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499791422904021,
      "loss": 6.8328,
      "mean_token_accuracy": 0.1454320654273033,
      "num_tokens": 5510847.0,
      "step": 2175
    },
    {
      "entropy": 6.7520914554595945,
      "epoch": 0.2515868436237738,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004997896427619491,
      "loss": 6.7613,
      "mean_token_accuracy": 0.14879355430603028,
      "num_tokens": 5522543.0,
      "step": 2180
    },
    {
      "entropy": 6.717813110351562,
      "epoch": 0.25216387766878245,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004997878550591892,
      "loss": 6.7648,
      "mean_token_accuracy": 0.1428240269422531,
      "num_tokens": 5534856.0,
      "step": 2185
    },
    {
      "entropy": 6.851426029205323,
      "epoch": 0.2527409117137911,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004997860597958013,
      "loss": 6.6801,
      "mean_token_accuracy": 0.15747882872819902,
      "num_tokens": 5546881.0,
      "step": 2190
    },
    {
      "entropy": 6.717418527603149,
      "epoch": 0.2533179457587998,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004997842569718461,
      "loss": 6.8,
      "mean_token_accuracy": 0.14309746026992798,
      "num_tokens": 5559711.0,
      "step": 2195
    },
    {
      "entropy": 6.8028624057769775,
      "epoch": 0.25389497980380843,
      "grad_norm": 0.9375,
      "learning_rate": 0.000499782446587384,
      "loss": 6.783,
      "mean_token_accuracy": 0.1497884564101696,
      "num_tokens": 5573553.0,
      "step": 2200
    },
    {
      "entropy": 6.793797731399536,
      "epoch": 0.25447201384881707,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000499780628642476,
      "loss": 6.7494,
      "mean_token_accuracy": 0.1393908143043518,
      "num_tokens": 5586046.0,
      "step": 2205
    },
    {
      "entropy": 6.702590990066528,
      "epoch": 0.2550490478938257,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997788031371834,
      "loss": 6.7686,
      "mean_token_accuracy": 0.14913811907172203,
      "num_tokens": 5597876.0,
      "step": 2210
    },
    {
      "entropy": 6.8008434772491455,
      "epoch": 0.2556260819388344,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997769700715672,
      "loss": 6.6819,
      "mean_token_accuracy": 0.1523703783750534,
      "num_tokens": 5611264.0,
      "step": 2215
    },
    {
      "entropy": 6.685852336883545,
      "epoch": 0.25620311598384304,
      "grad_norm": 0.875,
      "learning_rate": 0.0004997751294456892,
      "loss": 6.7773,
      "mean_token_accuracy": 0.14500191956758499,
      "num_tokens": 5625093.0,
      "step": 2220
    },
    {
      "entropy": 6.798491287231445,
      "epoch": 0.2567801500288517,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004997732812596114,
      "loss": 6.7895,
      "mean_token_accuracy": 0.14235708564519883,
      "num_tokens": 5637206.0,
      "step": 2225
    },
    {
      "entropy": 6.776926708221436,
      "epoch": 0.2573571840738604,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004997714255133961,
      "loss": 6.6586,
      "mean_token_accuracy": 0.15341545939445494,
      "num_tokens": 5649136.0,
      "step": 2230
    },
    {
      "entropy": 6.628551292419433,
      "epoch": 0.257934218118869,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997695622071054,
      "loss": 6.7041,
      "mean_token_accuracy": 0.14951958358287812,
      "num_tokens": 5660714.0,
      "step": 2235
    },
    {
      "entropy": 6.831679010391236,
      "epoch": 0.25851125216387766,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997676913408021,
      "loss": 6.7764,
      "mean_token_accuracy": 0.14415197670459748,
      "num_tokens": 5673295.0,
      "step": 2240
    },
    {
      "entropy": 6.752670621871948,
      "epoch": 0.2590882862088863,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004997658129145493,
      "loss": 6.7818,
      "mean_token_accuracy": 0.1467595562338829,
      "num_tokens": 5686822.0,
      "step": 2245
    },
    {
      "entropy": 6.774938583374023,
      "epoch": 0.259665320253895,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997639269284099,
      "loss": 6.8382,
      "mean_token_accuracy": 0.14073604345321655,
      "num_tokens": 5700621.0,
      "step": 2250
    },
    {
      "entropy": 6.839014863967895,
      "epoch": 0.26024235429890363,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997620333824476,
      "loss": 6.7463,
      "mean_token_accuracy": 0.1523364171385765,
      "num_tokens": 5713565.0,
      "step": 2255
    },
    {
      "entropy": 6.72583498954773,
      "epoch": 0.2608193883439123,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004997601322767258,
      "loss": 6.7916,
      "mean_token_accuracy": 0.14135704040527344,
      "num_tokens": 5726487.0,
      "step": 2260
    },
    {
      "entropy": 6.8859046459197994,
      "epoch": 0.26139642238892097,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004997582236113087,
      "loss": 6.8826,
      "mean_token_accuracy": 0.13812644258141518,
      "num_tokens": 5739934.0,
      "step": 2265
    },
    {
      "entropy": 6.76940541267395,
      "epoch": 0.2619734564339296,
      "grad_norm": 0.875,
      "learning_rate": 0.0004997563073862603,
      "loss": 6.8123,
      "mean_token_accuracy": 0.13768922835588454,
      "num_tokens": 5752187.0,
      "step": 2270
    },
    {
      "entropy": 6.70945725440979,
      "epoch": 0.26255049047893825,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004997543836016453,
      "loss": 6.7426,
      "mean_token_accuracy": 0.14485511556267738,
      "num_tokens": 5764850.0,
      "step": 2275
    },
    {
      "entropy": 6.867229843139649,
      "epoch": 0.2631275245239469,
      "grad_norm": 0.76953125,
      "learning_rate": 0.0004997524522575281,
      "loss": 6.8221,
      "mean_token_accuracy": 0.14024255499243737,
      "num_tokens": 5778577.0,
      "step": 2280
    },
    {
      "entropy": 6.7879596710205075,
      "epoch": 0.2637045585689556,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004997505133539738,
      "loss": 6.7548,
      "mean_token_accuracy": 0.14954013526439666,
      "num_tokens": 5791699.0,
      "step": 2285
    },
    {
      "entropy": 6.717269659042358,
      "epoch": 0.2642815926139642,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004997485668910476,
      "loss": 6.7576,
      "mean_token_accuracy": 0.14114084988832473,
      "num_tokens": 5804553.0,
      "step": 2290
    },
    {
      "entropy": 6.840640878677368,
      "epoch": 0.26485862665897286,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997466128688151,
      "loss": 6.7691,
      "mean_token_accuracy": 0.1443391412496567,
      "num_tokens": 5816758.0,
      "step": 2295
    },
    {
      "entropy": 6.754163122177124,
      "epoch": 0.26543566070398156,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000499744651287342,
      "loss": 6.7999,
      "mean_token_accuracy": 0.13716064393520355,
      "num_tokens": 5831048.0,
      "step": 2300
    },
    {
      "entropy": 6.835683441162109,
      "epoch": 0.2660126947489902,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000499742682146694,
      "loss": 6.7946,
      "mean_token_accuracy": 0.13617580756545067,
      "num_tokens": 5843159.0,
      "step": 2305
    },
    {
      "entropy": 6.710077857971191,
      "epoch": 0.26658972879399884,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004997407054469377,
      "loss": 6.6863,
      "mean_token_accuracy": 0.15132941156625748,
      "num_tokens": 5856078.0,
      "step": 2310
    },
    {
      "entropy": 6.773683309555054,
      "epoch": 0.2671667628390075,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004997387211881392,
      "loss": 6.7288,
      "mean_token_accuracy": 0.145392032712698,
      "num_tokens": 5867835.0,
      "step": 2315
    },
    {
      "entropy": 6.747898149490356,
      "epoch": 0.26774379688401617,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004997367293703656,
      "loss": 6.7683,
      "mean_token_accuracy": 0.14005454927682875,
      "num_tokens": 5881179.0,
      "step": 2320
    },
    {
      "entropy": 6.79307336807251,
      "epoch": 0.2683208309290248,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004997347299936837,
      "loss": 6.7228,
      "mean_token_accuracy": 0.14137931019067765,
      "num_tokens": 5892774.0,
      "step": 2325
    },
    {
      "entropy": 6.781385231018066,
      "epoch": 0.26889786497403345,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004997327230581608,
      "loss": 6.7838,
      "mean_token_accuracy": 0.14386948049068451,
      "num_tokens": 5905857.0,
      "step": 2330
    },
    {
      "entropy": 6.713219213485718,
      "epoch": 0.26947489901904215,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004997307085638644,
      "loss": 6.7246,
      "mean_token_accuracy": 0.14458930790424346,
      "num_tokens": 5918128.0,
      "step": 2335
    },
    {
      "entropy": 6.815898609161377,
      "epoch": 0.2700519330640508,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004997286865108621,
      "loss": 6.8516,
      "mean_token_accuracy": 0.14051285684108733,
      "num_tokens": 5930606.0,
      "step": 2340
    },
    {
      "entropy": 6.749667167663574,
      "epoch": 0.2706289671090594,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004997266568992222,
      "loss": 6.7854,
      "mean_token_accuracy": 0.13582591190934182,
      "num_tokens": 5944139.0,
      "step": 2345
    },
    {
      "entropy": 6.768195295333863,
      "epoch": 0.27120600115406807,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004997246197290128,
      "loss": 6.7608,
      "mean_token_accuracy": 0.14503443688154222,
      "num_tokens": 5956779.0,
      "step": 2350
    },
    {
      "entropy": 6.704603576660157,
      "epoch": 0.27178303519907676,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997225750003024,
      "loss": 6.7297,
      "mean_token_accuracy": 0.15043272450566292,
      "num_tokens": 5971510.0,
      "step": 2355
    },
    {
      "entropy": 6.63649697303772,
      "epoch": 0.2723600692440854,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004997205227131599,
      "loss": 6.7573,
      "mean_token_accuracy": 0.14194964990019798,
      "num_tokens": 5983877.0,
      "step": 2360
    },
    {
      "entropy": 6.898113250732422,
      "epoch": 0.27293710328909404,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004997184628676542,
      "loss": 6.7795,
      "mean_token_accuracy": 0.14526699259877204,
      "num_tokens": 5996739.0,
      "step": 2365
    },
    {
      "entropy": 6.708590412139893,
      "epoch": 0.27351413733410274,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004997163954638546,
      "loss": 6.7437,
      "mean_token_accuracy": 0.1482561096549034,
      "num_tokens": 6009289.0,
      "step": 2370
    },
    {
      "entropy": 6.8309389591217045,
      "epoch": 0.2740911713791114,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997143205018306,
      "loss": 6.811,
      "mean_token_accuracy": 0.13893268257379532,
      "num_tokens": 6021375.0,
      "step": 2375
    },
    {
      "entropy": 6.804889583587647,
      "epoch": 0.27466820542412,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004997122379816523,
      "loss": 6.7297,
      "mean_token_accuracy": 0.14610403627157212,
      "num_tokens": 6034182.0,
      "step": 2380
    },
    {
      "entropy": 6.675373125076294,
      "epoch": 0.27524523946912866,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004997101479033893,
      "loss": 6.7187,
      "mean_token_accuracy": 0.14495936110615731,
      "num_tokens": 6046486.0,
      "step": 2385
    },
    {
      "entropy": 6.791097116470337,
      "epoch": 0.27582227351413735,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997080502671122,
      "loss": 6.7782,
      "mean_token_accuracy": 0.14465533643960954,
      "num_tokens": 6060867.0,
      "step": 2390
    },
    {
      "entropy": 6.686882209777832,
      "epoch": 0.276399307559146,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004997059450728913,
      "loss": 6.6263,
      "mean_token_accuracy": 0.15024522840976715,
      "num_tokens": 6073000.0,
      "step": 2395
    },
    {
      "entropy": 6.704486465454101,
      "epoch": 0.27697634160415463,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004997038323207977,
      "loss": 6.6973,
      "mean_token_accuracy": 0.14645934849977493,
      "num_tokens": 6085884.0,
      "step": 2400
    },
    {
      "entropy": 6.675950574874878,
      "epoch": 0.2775533756491633,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004997017120109022,
      "loss": 6.6314,
      "mean_token_accuracy": 0.14963351488113402,
      "num_tokens": 6098230.0,
      "step": 2405
    },
    {
      "entropy": 6.721731472015381,
      "epoch": 0.27813040969417197,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004996995841432762,
      "loss": 6.6896,
      "mean_token_accuracy": 0.14837536364793777,
      "num_tokens": 6111568.0,
      "step": 2410
    },
    {
      "entropy": 6.646577262878418,
      "epoch": 0.2787074437391806,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996974487179915,
      "loss": 6.6649,
      "mean_token_accuracy": 0.14184629917144775,
      "num_tokens": 6123992.0,
      "step": 2415
    },
    {
      "entropy": 6.754623317718506,
      "epoch": 0.27928447778418924,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004996953057351195,
      "loss": 6.6896,
      "mean_token_accuracy": 0.1484291136264801,
      "num_tokens": 6135452.0,
      "step": 2420
    },
    {
      "entropy": 6.6486248016357425,
      "epoch": 0.27986151182919794,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996931551947327,
      "loss": 6.6125,
      "mean_token_accuracy": 0.156364855915308,
      "num_tokens": 6150144.0,
      "step": 2425
    },
    {
      "entropy": 6.7660589694976805,
      "epoch": 0.2804385458742066,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004996909970969031,
      "loss": 6.7401,
      "mean_token_accuracy": 0.14714512154459952,
      "num_tokens": 6163397.0,
      "step": 2430
    },
    {
      "entropy": 6.687129402160645,
      "epoch": 0.2810155799192152,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004996888314417034,
      "loss": 6.731,
      "mean_token_accuracy": 0.1419209510087967,
      "num_tokens": 6175579.0,
      "step": 2435
    },
    {
      "entropy": 6.683587169647216,
      "epoch": 0.2815926139642239,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004996866582292067,
      "loss": 6.7079,
      "mean_token_accuracy": 0.14961353838443756,
      "num_tokens": 6189229.0,
      "step": 2440
    },
    {
      "entropy": 6.73968095779419,
      "epoch": 0.28216964800923255,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996844774594856,
      "loss": 6.7092,
      "mean_token_accuracy": 0.14912996739149093,
      "num_tokens": 6203817.0,
      "step": 2445
    },
    {
      "entropy": 6.717224407196045,
      "epoch": 0.2827466820542412,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004996822891326138,
      "loss": 6.7278,
      "mean_token_accuracy": 0.14414327293634416,
      "num_tokens": 6216098.0,
      "step": 2450
    },
    {
      "entropy": 6.753392219543457,
      "epoch": 0.28332371609924983,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004996800932486648,
      "loss": 6.7286,
      "mean_token_accuracy": 0.1483657404780388,
      "num_tokens": 6228987.0,
      "step": 2455
    },
    {
      "entropy": 6.664376258850098,
      "epoch": 0.28390075014425853,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996778898077126,
      "loss": 6.675,
      "mean_token_accuracy": 0.1498671442270279,
      "num_tokens": 6241676.0,
      "step": 2460
    },
    {
      "entropy": 6.794411563873291,
      "epoch": 0.28447778418926717,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004996756788098309,
      "loss": 6.7131,
      "mean_token_accuracy": 0.14685458093881607,
      "num_tokens": 6254744.0,
      "step": 2465
    },
    {
      "entropy": 6.630081748962402,
      "epoch": 0.2850548182342758,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004996734602550945,
      "loss": 6.6722,
      "mean_token_accuracy": 0.14627297297120095,
      "num_tokens": 6268242.0,
      "step": 2470
    },
    {
      "entropy": 6.659547185897827,
      "epoch": 0.2856318522792845,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004996712341435779,
      "loss": 6.6072,
      "mean_token_accuracy": 0.15240462720394135,
      "num_tokens": 6279904.0,
      "step": 2475
    },
    {
      "entropy": 6.643069171905518,
      "epoch": 0.28620888632429314,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004996690004753559,
      "loss": 6.7204,
      "mean_token_accuracy": 0.1448551282286644,
      "num_tokens": 6292885.0,
      "step": 2480
    },
    {
      "entropy": 6.64668550491333,
      "epoch": 0.2867859203693018,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004996667592505037,
      "loss": 6.7049,
      "mean_token_accuracy": 0.1487411305308342,
      "num_tokens": 6306278.0,
      "step": 2485
    },
    {
      "entropy": 6.800214529037476,
      "epoch": 0.2873629544143104,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004996645104690967,
      "loss": 6.7361,
      "mean_token_accuracy": 0.13756746500730516,
      "num_tokens": 6320360.0,
      "step": 2490
    },
    {
      "entropy": 6.657628870010376,
      "epoch": 0.2879399884593191,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004996622541312103,
      "loss": 6.7216,
      "mean_token_accuracy": 0.145097079873085,
      "num_tokens": 6332778.0,
      "step": 2495
    },
    {
      "entropy": 6.7686714172363285,
      "epoch": 0.28851702250432776,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004996599902369207,
      "loss": 6.7012,
      "mean_token_accuracy": 0.1465804323554039,
      "num_tokens": 6345028.0,
      "step": 2500
    },
    {
      "entropy": 6.653385210037231,
      "epoch": 0.2890940565493364,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996577187863039,
      "loss": 6.6454,
      "mean_token_accuracy": 0.15272147804498673,
      "num_tokens": 6356931.0,
      "step": 2505
    },
    {
      "entropy": 6.664360809326172,
      "epoch": 0.2896710905943451,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004996554397794364,
      "loss": 6.5927,
      "mean_token_accuracy": 0.157369065284729,
      "num_tokens": 6370247.0,
      "step": 2510
    },
    {
      "entropy": 6.653738355636596,
      "epoch": 0.29024812463935373,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996531532163947,
      "loss": 6.6727,
      "mean_token_accuracy": 0.14644310176372527,
      "num_tokens": 6382352.0,
      "step": 2515
    },
    {
      "entropy": 6.664462614059448,
      "epoch": 0.2908251586843624,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996508590972558,
      "loss": 6.6628,
      "mean_token_accuracy": 0.1447147697210312,
      "num_tokens": 6395949.0,
      "step": 2520
    },
    {
      "entropy": 6.685240316390991,
      "epoch": 0.291402192729371,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004996485574220969,
      "loss": 6.6785,
      "mean_token_accuracy": 0.1504662573337555,
      "num_tokens": 6408469.0,
      "step": 2525
    },
    {
      "entropy": 6.677483606338501,
      "epoch": 0.2919792267743797,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004996462481909953,
      "loss": 6.675,
      "mean_token_accuracy": 0.1545150399208069,
      "num_tokens": 6419953.0,
      "step": 2530
    },
    {
      "entropy": 6.695938968658448,
      "epoch": 0.29255626081938835,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004996439314040287,
      "loss": 6.7126,
      "mean_token_accuracy": 0.1428425744175911,
      "num_tokens": 6432567.0,
      "step": 2535
    },
    {
      "entropy": 6.797707271575928,
      "epoch": 0.293133294864397,
      "grad_norm": 1.25,
      "learning_rate": 0.000499641607061275,
      "loss": 6.6713,
      "mean_token_accuracy": 0.14336236119270324,
      "num_tokens": 6445951.0,
      "step": 2540
    },
    {
      "entropy": 6.6457275390625,
      "epoch": 0.2937103289094057,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004996392751628126,
      "loss": 6.7311,
      "mean_token_accuracy": 0.15187496691942215,
      "num_tokens": 6458799.0,
      "step": 2545
    },
    {
      "entropy": 6.727105379104614,
      "epoch": 0.2942873629544143,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996369357087196,
      "loss": 6.7279,
      "mean_token_accuracy": 0.14346445798873902,
      "num_tokens": 6471423.0,
      "step": 2550
    },
    {
      "entropy": 6.745425605773926,
      "epoch": 0.29486439699942296,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499634588699075,
      "loss": 6.7509,
      "mean_token_accuracy": 0.14119785130023957,
      "num_tokens": 6484248.0,
      "step": 2555
    },
    {
      "entropy": 6.744411516189575,
      "epoch": 0.2954414310444316,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004996322341339575,
      "loss": 6.6769,
      "mean_token_accuracy": 0.1400275394320488,
      "num_tokens": 6497048.0,
      "step": 2560
    },
    {
      "entropy": 6.662609720230103,
      "epoch": 0.2960184650894403,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004996298720134465,
      "loss": 6.6918,
      "mean_token_accuracy": 0.1442575104534626,
      "num_tokens": 6508840.0,
      "step": 2565
    },
    {
      "entropy": 6.695674324035645,
      "epoch": 0.29659549913444894,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004996275023376213,
      "loss": 6.6791,
      "mean_token_accuracy": 0.14943985342979432,
      "num_tokens": 6520976.0,
      "step": 2570
    },
    {
      "entropy": 6.68662805557251,
      "epoch": 0.2971725331794576,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004996251251065615,
      "loss": 6.6945,
      "mean_token_accuracy": 0.14707806557416916,
      "num_tokens": 6533009.0,
      "step": 2575
    },
    {
      "entropy": 6.712952613830566,
      "epoch": 0.2977495672244662,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004996227403203473,
      "loss": 6.6792,
      "mean_token_accuracy": 0.14307745099067687,
      "num_tokens": 6545196.0,
      "step": 2580
    },
    {
      "entropy": 6.744017934799194,
      "epoch": 0.2983266012694749,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004996203479790589,
      "loss": 6.7119,
      "mean_token_accuracy": 0.1456460326910019,
      "num_tokens": 6557479.0,
      "step": 2585
    },
    {
      "entropy": 6.709696102142334,
      "epoch": 0.29890363531448355,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996179480827764,
      "loss": 6.733,
      "mean_token_accuracy": 0.14646259546279908,
      "num_tokens": 6569832.0,
      "step": 2590
    },
    {
      "entropy": 6.684864521026611,
      "epoch": 0.2994806693594922,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000499615540631581,
      "loss": 6.6935,
      "mean_token_accuracy": 0.14482910484075545,
      "num_tokens": 6581651.0,
      "step": 2595
    },
    {
      "entropy": 6.737748479843139,
      "epoch": 0.3000577034045009,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004996131256255534,
      "loss": 6.7498,
      "mean_token_accuracy": 0.14754320159554482,
      "num_tokens": 6596331.0,
      "step": 2600
    },
    {
      "entropy": 6.737265586853027,
      "epoch": 0.3006347374495095,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004996107030647749,
      "loss": 6.6931,
      "mean_token_accuracy": 0.14396552741527557,
      "num_tokens": 6609712.0,
      "step": 2605
    },
    {
      "entropy": 6.733648252487183,
      "epoch": 0.30121177149451817,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996082729493269,
      "loss": 6.71,
      "mean_token_accuracy": 0.14992823898792268,
      "num_tokens": 6622780.0,
      "step": 2610
    },
    {
      "entropy": 6.6975846767425535,
      "epoch": 0.3017888055395268,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996058352792913,
      "loss": 6.7664,
      "mean_token_accuracy": 0.14214256927371025,
      "num_tokens": 6636288.0,
      "step": 2615
    },
    {
      "entropy": 6.7836583137512205,
      "epoch": 0.3023658395845355,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00049960339005475,
      "loss": 6.7468,
      "mean_token_accuracy": 0.1360473372042179,
      "num_tokens": 6649883.0,
      "step": 2620
    },
    {
      "entropy": 6.70995774269104,
      "epoch": 0.30294287362954414,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004996009372757852,
      "loss": 6.6904,
      "mean_token_accuracy": 0.14994267225265503,
      "num_tokens": 6662241.0,
      "step": 2625
    },
    {
      "entropy": 6.6174835681915285,
      "epoch": 0.3035199076745528,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004995984769424795,
      "loss": 6.6436,
      "mean_token_accuracy": 0.148514224588871,
      "num_tokens": 6674790.0,
      "step": 2630
    },
    {
      "entropy": 6.731441926956177,
      "epoch": 0.3040969417195615,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004995960090549155,
      "loss": 6.67,
      "mean_token_accuracy": 0.14512893110513686,
      "num_tokens": 6687713.0,
      "step": 2635
    },
    {
      "entropy": 6.580911350250244,
      "epoch": 0.3046739757645701,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995935336131764,
      "loss": 6.6733,
      "mean_token_accuracy": 0.1467847667634487,
      "num_tokens": 6701123.0,
      "step": 2640
    },
    {
      "entropy": 6.748194313049316,
      "epoch": 0.30525100980957875,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004995910506173452,
      "loss": 6.6264,
      "mean_token_accuracy": 0.152266626060009,
      "num_tokens": 6712674.0,
      "step": 2645
    },
    {
      "entropy": 6.585151863098145,
      "epoch": 0.3058280438545874,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004995885600675059,
      "loss": 6.6317,
      "mean_token_accuracy": 0.15141176879405976,
      "num_tokens": 6724840.0,
      "step": 2650
    },
    {
      "entropy": 6.713297176361084,
      "epoch": 0.3064050778995961,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004995860619637415,
      "loss": 6.7572,
      "mean_token_accuracy": 0.1347401686012745,
      "num_tokens": 6738746.0,
      "step": 2655
    },
    {
      "entropy": 6.730117273330689,
      "epoch": 0.30698211194460473,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004995835563061367,
      "loss": 6.6619,
      "mean_token_accuracy": 0.14460855424404145,
      "num_tokens": 6751620.0,
      "step": 2660
    },
    {
      "entropy": 6.723760747909546,
      "epoch": 0.30755914598961337,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004995810430947756,
      "loss": 6.6421,
      "mean_token_accuracy": 0.14792002290487288,
      "num_tokens": 6763957.0,
      "step": 2665
    },
    {
      "entropy": 6.4763280868530275,
      "epoch": 0.30813618003462206,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004995785223297426,
      "loss": 6.551,
      "mean_token_accuracy": 0.1542297527194023,
      "num_tokens": 6776211.0,
      "step": 2670
    },
    {
      "entropy": 6.775838565826416,
      "epoch": 0.3087132140796307,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004995759940111225,
      "loss": 6.8122,
      "mean_token_accuracy": 0.13833045810461045,
      "num_tokens": 6789745.0,
      "step": 2675
    },
    {
      "entropy": 6.757865047454834,
      "epoch": 0.30929024812463934,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004995734581390005,
      "loss": 6.7322,
      "mean_token_accuracy": 0.14661592841148377,
      "num_tokens": 6801771.0,
      "step": 2680
    },
    {
      "entropy": 6.666882467269898,
      "epoch": 0.309867282169648,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004995709147134617,
      "loss": 6.67,
      "mean_token_accuracy": 0.14816711097955704,
      "num_tokens": 6813993.0,
      "step": 2685
    },
    {
      "entropy": 6.75975227355957,
      "epoch": 0.3104443162146567,
      "grad_norm": 0.875,
      "learning_rate": 0.0004995683637345919,
      "loss": 6.728,
      "mean_token_accuracy": 0.14804089665412903,
      "num_tokens": 6826833.0,
      "step": 2690
    },
    {
      "entropy": 6.66631441116333,
      "epoch": 0.3110213502596653,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004995658052024765,
      "loss": 6.6693,
      "mean_token_accuracy": 0.1456070676445961,
      "num_tokens": 6838947.0,
      "step": 2695
    },
    {
      "entropy": 6.758131980895996,
      "epoch": 0.31159838430467396,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004995632391172019,
      "loss": 6.7287,
      "mean_token_accuracy": 0.14437391310930253,
      "num_tokens": 6851781.0,
      "step": 2700
    },
    {
      "entropy": 6.707075405120849,
      "epoch": 0.31217541834968265,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004995606654788543,
      "loss": 6.6256,
      "mean_token_accuracy": 0.15065254122018815,
      "num_tokens": 6863479.0,
      "step": 2705
    },
    {
      "entropy": 6.62445650100708,
      "epoch": 0.3127524523946913,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004995580842875202,
      "loss": 6.6145,
      "mean_token_accuracy": 0.15116894394159316,
      "num_tokens": 6877000.0,
      "step": 2710
    },
    {
      "entropy": 6.720065450668335,
      "epoch": 0.31332948643969993,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995554955432866,
      "loss": 6.7174,
      "mean_token_accuracy": 0.14094419330358504,
      "num_tokens": 6888878.0,
      "step": 2715
    },
    {
      "entropy": 6.668194341659546,
      "epoch": 0.3139065204847086,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004995528992462404,
      "loss": 6.6328,
      "mean_token_accuracy": 0.14614609777927398,
      "num_tokens": 6902147.0,
      "step": 2720
    },
    {
      "entropy": 6.681904458999634,
      "epoch": 0.31448355452971727,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000499550295396469,
      "loss": 6.6788,
      "mean_token_accuracy": 0.1509232111275196,
      "num_tokens": 6914927.0,
      "step": 2725
    },
    {
      "entropy": 6.710317325592041,
      "epoch": 0.3150605885747259,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004995476839940598,
      "loss": 6.6955,
      "mean_token_accuracy": 0.1472512513399124,
      "num_tokens": 6927757.0,
      "step": 2730
    },
    {
      "entropy": 6.674422359466552,
      "epoch": 0.31563762261973455,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000499545065039101,
      "loss": 6.6768,
      "mean_token_accuracy": 0.14419350624084473,
      "num_tokens": 6941039.0,
      "step": 2735
    },
    {
      "entropy": 6.694507312774658,
      "epoch": 0.31621465666474324,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004995424385316803,
      "loss": 6.7503,
      "mean_token_accuracy": 0.14313241690397263,
      "num_tokens": 6954876.0,
      "step": 2740
    },
    {
      "entropy": 6.665238857269287,
      "epoch": 0.3167916907097519,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004995398044718863,
      "loss": 6.596,
      "mean_token_accuracy": 0.15183332860469817,
      "num_tokens": 6966889.0,
      "step": 2745
    },
    {
      "entropy": 6.698918628692627,
      "epoch": 0.3173687247547605,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004995371628598074,
      "loss": 6.6259,
      "mean_token_accuracy": 0.1496454581618309,
      "num_tokens": 6979894.0,
      "step": 2750
    },
    {
      "entropy": 6.705928373336792,
      "epoch": 0.31794575879976916,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004995345136955328,
      "loss": 6.6708,
      "mean_token_accuracy": 0.143413520604372,
      "num_tokens": 6992734.0,
      "step": 2755
    },
    {
      "entropy": 6.664940214157104,
      "epoch": 0.31852279284477786,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004995318569791512,
      "loss": 6.6779,
      "mean_token_accuracy": 0.14752028435468673,
      "num_tokens": 7006504.0,
      "step": 2760
    },
    {
      "entropy": 6.694177198410034,
      "epoch": 0.3190998268897865,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004995291927107522,
      "loss": 6.6083,
      "mean_token_accuracy": 0.15231579244136811,
      "num_tokens": 7018978.0,
      "step": 2765
    },
    {
      "entropy": 6.5929046154022215,
      "epoch": 0.31967686093479514,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004995265208904252,
      "loss": 6.6973,
      "mean_token_accuracy": 0.14533187747001647,
      "num_tokens": 7031294.0,
      "step": 2770
    },
    {
      "entropy": 6.704652452468872,
      "epoch": 0.32025389497980383,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004995238415182603,
      "loss": 6.6554,
      "mean_token_accuracy": 0.1433829165995121,
      "num_tokens": 7044139.0,
      "step": 2775
    },
    {
      "entropy": 6.583083963394165,
      "epoch": 0.3208309290248125,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004995211545943476,
      "loss": 6.5525,
      "mean_token_accuracy": 0.15347641259431838,
      "num_tokens": 7056232.0,
      "step": 2780
    },
    {
      "entropy": 6.695966291427612,
      "epoch": 0.3214079630698211,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004995184601187772,
      "loss": 6.7123,
      "mean_token_accuracy": 0.14398207515478134,
      "num_tokens": 7070125.0,
      "step": 2785
    },
    {
      "entropy": 6.615304565429687,
      "epoch": 0.32198499711482975,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00049951575809164,
      "loss": 6.5848,
      "mean_token_accuracy": 0.15029205679893493,
      "num_tokens": 7083535.0,
      "step": 2790
    },
    {
      "entropy": 6.624817371368408,
      "epoch": 0.32256203115983845,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004995130485130269,
      "loss": 6.5539,
      "mean_token_accuracy": 0.1532074198126793,
      "num_tokens": 7095081.0,
      "step": 2795
    },
    {
      "entropy": 6.699851179122925,
      "epoch": 0.3231390652048471,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004995103313830289,
      "loss": 6.683,
      "mean_token_accuracy": 0.14111172929406166,
      "num_tokens": 7107764.0,
      "step": 2800
    },
    {
      "entropy": 6.6601025581359865,
      "epoch": 0.3237160992498557,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004995076067017373,
      "loss": 6.6111,
      "mean_token_accuracy": 0.15362927615642546,
      "num_tokens": 7119754.0,
      "step": 2805
    },
    {
      "entropy": 6.564910221099853,
      "epoch": 0.3242931332948644,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004995048744692439,
      "loss": 6.6472,
      "mean_token_accuracy": 0.1543910101056099,
      "num_tokens": 7132087.0,
      "step": 2810
    },
    {
      "entropy": 6.642766809463501,
      "epoch": 0.32487016733987306,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004995021346856405,
      "loss": 6.6196,
      "mean_token_accuracy": 0.15014703422784806,
      "num_tokens": 7143325.0,
      "step": 2815
    },
    {
      "entropy": 6.652205085754394,
      "epoch": 0.3254472013848817,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004994993873510196,
      "loss": 6.5793,
      "mean_token_accuracy": 0.15422357767820358,
      "num_tokens": 7154771.0,
      "step": 2820
    },
    {
      "entropy": 6.642023944854737,
      "epoch": 0.32602423542989034,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004994966324654731,
      "loss": 6.693,
      "mean_token_accuracy": 0.1453319787979126,
      "num_tokens": 7168352.0,
      "step": 2825
    },
    {
      "entropy": 6.497804021835327,
      "epoch": 0.32660126947489904,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000499493870029094,
      "loss": 6.4934,
      "mean_token_accuracy": 0.16205645650625228,
      "num_tokens": 7180618.0,
      "step": 2830
    },
    {
      "entropy": 6.666966104507447,
      "epoch": 0.3271783035199077,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004994911000419749,
      "loss": 6.6267,
      "mean_token_accuracy": 0.14406555965542794,
      "num_tokens": 7192634.0,
      "step": 2835
    },
    {
      "entropy": 6.592815494537353,
      "epoch": 0.3277553375649163,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004994883225042093,
      "loss": 6.6478,
      "mean_token_accuracy": 0.14309148415923117,
      "num_tokens": 7205839.0,
      "step": 2840
    },
    {
      "entropy": 6.6875081062316895,
      "epoch": 0.328332371609925,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004994855374158905,
      "loss": 6.6635,
      "mean_token_accuracy": 0.14771109595894813,
      "num_tokens": 7218371.0,
      "step": 2845
    },
    {
      "entropy": 6.6291303634643555,
      "epoch": 0.32890940565493365,
      "grad_norm": 1.8515625,
      "learning_rate": 0.000499482744777112,
      "loss": 6.4748,
      "mean_token_accuracy": 0.16835684180259705,
      "num_tokens": 7231506.0,
      "step": 2850
    },
    {
      "entropy": 6.602149820327758,
      "epoch": 0.3294864396999423,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000499479944587968,
      "loss": 6.6033,
      "mean_token_accuracy": 0.15234598368406296,
      "num_tokens": 7243088.0,
      "step": 2855
    },
    {
      "entropy": 6.582425212860107,
      "epoch": 0.33006347374495093,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004994771368485526,
      "loss": 6.5306,
      "mean_token_accuracy": 0.1502147063612938,
      "num_tokens": 7255383.0,
      "step": 2860
    },
    {
      "entropy": 6.583856105804443,
      "epoch": 0.3306405077899596,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004994743215589602,
      "loss": 6.6303,
      "mean_token_accuracy": 0.15170817896723748,
      "num_tokens": 7268270.0,
      "step": 2865
    },
    {
      "entropy": 6.590276622772217,
      "epoch": 0.33121754183496827,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004994714987192854,
      "loss": 6.543,
      "mean_token_accuracy": 0.15047916322946547,
      "num_tokens": 7281108.0,
      "step": 2870
    },
    {
      "entropy": 6.596184539794922,
      "epoch": 0.3317945758799769,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004994686683296234,
      "loss": 6.6151,
      "mean_token_accuracy": 0.14714079722762108,
      "num_tokens": 7293961.0,
      "step": 2875
    },
    {
      "entropy": 6.654780054092408,
      "epoch": 0.3323716099249856,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004994658303900691,
      "loss": 6.6303,
      "mean_token_accuracy": 0.14734134078025818,
      "num_tokens": 7306574.0,
      "step": 2880
    },
    {
      "entropy": 6.68011884689331,
      "epoch": 0.33294864396999424,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004994629849007182,
      "loss": 6.667,
      "mean_token_accuracy": 0.14702263921499253,
      "num_tokens": 7319046.0,
      "step": 2885
    },
    {
      "entropy": 6.692756223678589,
      "epoch": 0.3335256780150029,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994601318616663,
      "loss": 6.659,
      "mean_token_accuracy": 0.14255366772413253,
      "num_tokens": 7332033.0,
      "step": 2890
    },
    {
      "entropy": 6.608484172821045,
      "epoch": 0.3341027120600115,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004994572712730092,
      "loss": 6.6756,
      "mean_token_accuracy": 0.13977887481451035,
      "num_tokens": 7344658.0,
      "step": 2895
    },
    {
      "entropy": 6.74255256652832,
      "epoch": 0.3346797461050202,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004994544031348434,
      "loss": 6.5531,
      "mean_token_accuracy": 0.1509678602218628,
      "num_tokens": 7357252.0,
      "step": 2900
    },
    {
      "entropy": 6.453712368011475,
      "epoch": 0.33525678015002885,
      "grad_norm": 0.875,
      "learning_rate": 0.0004994515274472653,
      "loss": 6.5003,
      "mean_token_accuracy": 0.16169244945049285,
      "num_tokens": 7369280.0,
      "step": 2905
    },
    {
      "entropy": 6.535965251922607,
      "epoch": 0.3358338141950375,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004994486442103715,
      "loss": 6.5023,
      "mean_token_accuracy": 0.15722685903310776,
      "num_tokens": 7383017.0,
      "step": 2910
    },
    {
      "entropy": 6.617752981185913,
      "epoch": 0.3364108482400462,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499445753424259,
      "loss": 6.6474,
      "mean_token_accuracy": 0.13711773306131364,
      "num_tokens": 7395634.0,
      "step": 2915
    },
    {
      "entropy": 6.697910356521606,
      "epoch": 0.33698788228505483,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004994428550890251,
      "loss": 6.62,
      "mean_token_accuracy": 0.1489083245396614,
      "num_tokens": 7408590.0,
      "step": 2920
    },
    {
      "entropy": 6.6144743919372555,
      "epoch": 0.33756491633006347,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004994399492047672,
      "loss": 6.6102,
      "mean_token_accuracy": 0.14934324771165847,
      "num_tokens": 7420202.0,
      "step": 2925
    },
    {
      "entropy": 6.6217461109161375,
      "epoch": 0.3381419503750721,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994370357715832,
      "loss": 6.5721,
      "mean_token_accuracy": 0.15716550201177598,
      "num_tokens": 7432664.0,
      "step": 2930
    },
    {
      "entropy": 6.596539640426636,
      "epoch": 0.3387189844200808,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000499434114789571,
      "loss": 6.7155,
      "mean_token_accuracy": 0.1429716408252716,
      "num_tokens": 7446084.0,
      "step": 2935
    },
    {
      "entropy": 6.627397680282593,
      "epoch": 0.33929601846508944,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004994311862588286,
      "loss": 6.5659,
      "mean_token_accuracy": 0.15205595940351485,
      "num_tokens": 7458314.0,
      "step": 2940
    },
    {
      "entropy": 6.650746536254883,
      "epoch": 0.3398730525100981,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004994282501794551,
      "loss": 6.6073,
      "mean_token_accuracy": 0.15533693358302117,
      "num_tokens": 7471995.0,
      "step": 2945
    },
    {
      "entropy": 6.6287164211273195,
      "epoch": 0.3404500865551068,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004994253065515486,
      "loss": 6.6467,
      "mean_token_accuracy": 0.14468009397387505,
      "num_tokens": 7485676.0,
      "step": 2950
    },
    {
      "entropy": 6.698169898986817,
      "epoch": 0.3410271206001154,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004994223553752084,
      "loss": 6.5347,
      "mean_token_accuracy": 0.15355831310153006,
      "num_tokens": 7497057.0,
      "step": 2955
    },
    {
      "entropy": 6.529919147491455,
      "epoch": 0.34160415464512406,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004994193966505339,
      "loss": 6.5581,
      "mean_token_accuracy": 0.152175223082304,
      "num_tokens": 7510625.0,
      "step": 2960
    },
    {
      "entropy": 6.542059373855591,
      "epoch": 0.3421811886901327,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004994164303776244,
      "loss": 6.5599,
      "mean_token_accuracy": 0.15199756473302842,
      "num_tokens": 7522677.0,
      "step": 2965
    },
    {
      "entropy": 6.536736965179443,
      "epoch": 0.3427582227351414,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004994134565565797,
      "loss": 6.4882,
      "mean_token_accuracy": 0.15326575115323066,
      "num_tokens": 7535406.0,
      "step": 2970
    },
    {
      "entropy": 6.621460723876953,
      "epoch": 0.34333525678015003,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004994104751875,
      "loss": 6.6714,
      "mean_token_accuracy": 0.14512530714273453,
      "num_tokens": 7547577.0,
      "step": 2975
    },
    {
      "entropy": 6.65846061706543,
      "epoch": 0.3439122908251587,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004994074862704854,
      "loss": 6.6405,
      "mean_token_accuracy": 0.14858071729540825,
      "num_tokens": 7561734.0,
      "step": 2980
    },
    {
      "entropy": 6.637954330444336,
      "epoch": 0.3444893248701673,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004994044898056363,
      "loss": 6.7036,
      "mean_token_accuracy": 0.1463548108935356,
      "num_tokens": 7574256.0,
      "step": 2985
    },
    {
      "entropy": 6.763088846206665,
      "epoch": 0.345066358915176,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004994014857930538,
      "loss": 6.6147,
      "mean_token_accuracy": 0.15213310569524766,
      "num_tokens": 7587149.0,
      "step": 2990
    },
    {
      "entropy": 6.554739332199096,
      "epoch": 0.34564339296018465,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004993984742328387,
      "loss": 6.5758,
      "mean_token_accuracy": 0.14600939601659774,
      "num_tokens": 7599124.0,
      "step": 2995
    },
    {
      "entropy": 6.577654266357422,
      "epoch": 0.3462204270051933,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004993954551250925,
      "loss": 6.5727,
      "mean_token_accuracy": 0.15178048834204674,
      "num_tokens": 7612070.0,
      "step": 3000
    },
    {
      "epoch": 0.3462204270051933,
      "eval_entropy": 6.630148336634502,
      "eval_loss": 6.675414085388184,
      "eval_mean_token_accuracy": 0.15012675653883586,
      "eval_num_tokens": 7612070.0,
      "eval_runtime": 23.0236,
      "eval_samples_per_second": 1222.048,
      "eval_steps_per_second": 152.756,
      "step": 3000
    },
    {
      "entropy": 6.499375486373902,
      "epoch": 0.346797461050202,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004993924284699165,
      "loss": 6.4838,
      "mean_token_accuracy": 0.15679868310689926,
      "num_tokens": 7626096.0,
      "step": 3005
    },
    {
      "entropy": 6.6755146980285645,
      "epoch": 0.3473744950952106,
      "grad_norm": 1.0,
      "learning_rate": 0.0004993893942674127,
      "loss": 6.6132,
      "mean_token_accuracy": 0.15309149473905564,
      "num_tokens": 7639666.0,
      "step": 3010
    },
    {
      "entropy": 6.593566608428955,
      "epoch": 0.34795152914021926,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499386352517683,
      "loss": 6.5669,
      "mean_token_accuracy": 0.14867550283670425,
      "num_tokens": 7652904.0,
      "step": 3015
    },
    {
      "entropy": 6.566877222061157,
      "epoch": 0.3485285631852279,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00049938330322083,
      "loss": 6.5923,
      "mean_token_accuracy": 0.1505222924053669,
      "num_tokens": 7665523.0,
      "step": 3020
    },
    {
      "entropy": 6.687088298797607,
      "epoch": 0.3491055972302366,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004993802463769558,
      "loss": 6.6416,
      "mean_token_accuracy": 0.1465634897351265,
      "num_tokens": 7677063.0,
      "step": 3025
    },
    {
      "entropy": 6.5251555919647215,
      "epoch": 0.34968263127524524,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004993771819861636,
      "loss": 6.5582,
      "mean_token_accuracy": 0.1502845138311386,
      "num_tokens": 7688980.0,
      "step": 3030
    },
    {
      "entropy": 6.7595607280731205,
      "epoch": 0.3502596653202539,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993741100485564,
      "loss": 6.7128,
      "mean_token_accuracy": 0.14851892068982125,
      "num_tokens": 7702604.0,
      "step": 3035
    },
    {
      "entropy": 6.622026205062866,
      "epoch": 0.35083669936526257,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004993710305642374,
      "loss": 6.634,
      "mean_token_accuracy": 0.14205431789159775,
      "num_tokens": 7715324.0,
      "step": 3040
    },
    {
      "entropy": 6.7205146789550785,
      "epoch": 0.3514137334102712,
      "grad_norm": 1.125,
      "learning_rate": 0.0004993679435333102,
      "loss": 6.6669,
      "mean_token_accuracy": 0.13979112580418587,
      "num_tokens": 7729098.0,
      "step": 3045
    },
    {
      "entropy": 6.582347726821899,
      "epoch": 0.35199076745527985,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004993648489558788,
      "loss": 6.5937,
      "mean_token_accuracy": 0.14882144778966905,
      "num_tokens": 7740717.0,
      "step": 3050
    },
    {
      "entropy": 6.50939793586731,
      "epoch": 0.3525678015002885,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993617468320471,
      "loss": 6.5285,
      "mean_token_accuracy": 0.14708343297243118,
      "num_tokens": 7753928.0,
      "step": 3055
    },
    {
      "entropy": 6.690690517425537,
      "epoch": 0.3531448355452972,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004993586371619193,
      "loss": 6.6243,
      "mean_token_accuracy": 0.1489599272608757,
      "num_tokens": 7765728.0,
      "step": 3060
    },
    {
      "entropy": 6.559810018539428,
      "epoch": 0.3537218695903058,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004993555199456004,
      "loss": 6.5543,
      "mean_token_accuracy": 0.15108276754617692,
      "num_tokens": 7778199.0,
      "step": 3065
    },
    {
      "entropy": 6.546053552627564,
      "epoch": 0.35429890363531447,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004993523951831949,
      "loss": 6.5403,
      "mean_token_accuracy": 0.15936604738235474,
      "num_tokens": 7791026.0,
      "step": 3070
    },
    {
      "entropy": 6.651627731323242,
      "epoch": 0.35487593768032316,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004993492628748081,
      "loss": 6.6208,
      "mean_token_accuracy": 0.15678820461034776,
      "num_tokens": 7804333.0,
      "step": 3075
    },
    {
      "entropy": 6.577079200744629,
      "epoch": 0.3554529717253318,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004993461230205453,
      "loss": 6.6218,
      "mean_token_accuracy": 0.14729673117399217,
      "num_tokens": 7818339.0,
      "step": 3080
    },
    {
      "entropy": 6.591600799560547,
      "epoch": 0.35603000577034044,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004993429756205121,
      "loss": 6.5942,
      "mean_token_accuracy": 0.15135954916477204,
      "num_tokens": 7831537.0,
      "step": 3085
    },
    {
      "entropy": 6.64121823310852,
      "epoch": 0.3566070398153491,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004993398206748142,
      "loss": 6.6082,
      "mean_token_accuracy": 0.14833982437849044,
      "num_tokens": 7843705.0,
      "step": 3090
    },
    {
      "entropy": 6.594843626022339,
      "epoch": 0.3571840738603578,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004993366581835581,
      "loss": 6.587,
      "mean_token_accuracy": 0.150689447671175,
      "num_tokens": 7856481.0,
      "step": 3095
    },
    {
      "entropy": 6.592101001739502,
      "epoch": 0.3577611079053664,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004993334881468498,
      "loss": 6.5919,
      "mean_token_accuracy": 0.14484333842992783,
      "num_tokens": 7868732.0,
      "step": 3100
    },
    {
      "entropy": 6.591314506530762,
      "epoch": 0.35833814195037506,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004993303105647961,
      "loss": 6.5555,
      "mean_token_accuracy": 0.15204939842224122,
      "num_tokens": 7881470.0,
      "step": 3105
    },
    {
      "entropy": 6.66913537979126,
      "epoch": 0.35891517599538375,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004993271254375038,
      "loss": 6.6376,
      "mean_token_accuracy": 0.14551813378930092,
      "num_tokens": 7895074.0,
      "step": 3110
    },
    {
      "entropy": 6.6260560035705565,
      "epoch": 0.3594922100403924,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00049932393276508,
      "loss": 6.5999,
      "mean_token_accuracy": 0.15333970114588738,
      "num_tokens": 7909140.0,
      "step": 3115
    },
    {
      "entropy": 6.595978403091431,
      "epoch": 0.36006924408540103,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004993207325476323,
      "loss": 6.5674,
      "mean_token_accuracy": 0.15037311241030693,
      "num_tokens": 7922689.0,
      "step": 3120
    },
    {
      "entropy": 6.5862952709198,
      "epoch": 0.36064627813040967,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004993175247852681,
      "loss": 6.5672,
      "mean_token_accuracy": 0.15013089254498482,
      "num_tokens": 7935207.0,
      "step": 3125
    },
    {
      "entropy": 6.562540864944458,
      "epoch": 0.36122331217541837,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004993143094780954,
      "loss": 6.5094,
      "mean_token_accuracy": 0.15271905511617662,
      "num_tokens": 7947491.0,
      "step": 3130
    },
    {
      "entropy": 6.589786529541016,
      "epoch": 0.361800346220427,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004993110866262224,
      "loss": 6.5321,
      "mean_token_accuracy": 0.15213637202978134,
      "num_tokens": 7959329.0,
      "step": 3135
    },
    {
      "entropy": 6.571454858779907,
      "epoch": 0.36237738026543564,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004993078562297573,
      "loss": 6.5428,
      "mean_token_accuracy": 0.15525427907705308,
      "num_tokens": 7971794.0,
      "step": 3140
    },
    {
      "entropy": 6.546091890335083,
      "epoch": 0.36295441431044434,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004993046182888089,
      "loss": 6.7002,
      "mean_token_accuracy": 0.1529540851712227,
      "num_tokens": 7983942.0,
      "step": 3145
    },
    {
      "entropy": 6.666069650650025,
      "epoch": 0.363531448355453,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004993013728034861,
      "loss": 6.5557,
      "mean_token_accuracy": 0.15181467831134796,
      "num_tokens": 7996266.0,
      "step": 3150
    },
    {
      "entropy": 6.587541770935059,
      "epoch": 0.3641084824004616,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000499298119773898,
      "loss": 6.6097,
      "mean_token_accuracy": 0.1477380082011223,
      "num_tokens": 8009318.0,
      "step": 3155
    },
    {
      "entropy": 6.662760877609253,
      "epoch": 0.36468551644547026,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004992948592001541,
      "loss": 6.63,
      "mean_token_accuracy": 0.1521727740764618,
      "num_tokens": 8021213.0,
      "step": 3160
    },
    {
      "entropy": 6.592753887176514,
      "epoch": 0.36526255049047895,
      "grad_norm": 0.953125,
      "learning_rate": 0.000499291591082364,
      "loss": 6.5918,
      "mean_token_accuracy": 0.14336525797843933,
      "num_tokens": 8033927.0,
      "step": 3165
    },
    {
      "entropy": 6.521289873123169,
      "epoch": 0.3658395845354876,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004992883154206377,
      "loss": 6.4714,
      "mean_token_accuracy": 0.15562977492809296,
      "num_tokens": 8046709.0,
      "step": 3170
    },
    {
      "entropy": 6.656177043914795,
      "epoch": 0.36641661858049623,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004992850322150853,
      "loss": 6.6152,
      "mean_token_accuracy": 0.1452882058918476,
      "num_tokens": 8060127.0,
      "step": 3175
    },
    {
      "entropy": 6.565167570114136,
      "epoch": 0.36699365262550493,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004992817414658172,
      "loss": 6.5498,
      "mean_token_accuracy": 0.15524625331163405,
      "num_tokens": 8072927.0,
      "step": 3180
    },
    {
      "entropy": 6.472782707214355,
      "epoch": 0.36757068667051357,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004992784431729442,
      "loss": 6.4801,
      "mean_token_accuracy": 0.15559360086917878,
      "num_tokens": 8083973.0,
      "step": 3185
    },
    {
      "entropy": 6.656773710250855,
      "epoch": 0.3681477207155222,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004992751373365771,
      "loss": 6.5455,
      "mean_token_accuracy": 0.1504301205277443,
      "num_tokens": 8096801.0,
      "step": 3190
    },
    {
      "entropy": 6.640792560577393,
      "epoch": 0.36872475476053085,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004992718239568273,
      "loss": 6.5577,
      "mean_token_accuracy": 0.1523575022816658,
      "num_tokens": 8109099.0,
      "step": 3195
    },
    {
      "entropy": 6.5834003448486325,
      "epoch": 0.36930178880553954,
      "grad_norm": 0.90625,
      "learning_rate": 0.000499268503033806,
      "loss": 6.5035,
      "mean_token_accuracy": 0.16071823239326477,
      "num_tokens": 8120712.0,
      "step": 3200
    },
    {
      "entropy": 6.548770141601563,
      "epoch": 0.3698788228505482,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004992651745676249,
      "loss": 6.5362,
      "mean_token_accuracy": 0.15628171116113662,
      "num_tokens": 8132228.0,
      "step": 3205
    },
    {
      "entropy": 6.615912437438965,
      "epoch": 0.3704558568955568,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004992618385583962,
      "loss": 6.5684,
      "mean_token_accuracy": 0.15014847964048386,
      "num_tokens": 8145015.0,
      "step": 3210
    },
    {
      "entropy": 6.5245506286621096,
      "epoch": 0.3710328909405655,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004992584950062319,
      "loss": 6.5041,
      "mean_token_accuracy": 0.15304392874240874,
      "num_tokens": 8156985.0,
      "step": 3215
    },
    {
      "entropy": 6.656549978256225,
      "epoch": 0.37160992498557416,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004992551439112446,
      "loss": 6.684,
      "mean_token_accuracy": 0.14261055141687393,
      "num_tokens": 8170633.0,
      "step": 3220
    },
    {
      "entropy": 6.651282215118409,
      "epoch": 0.3721869590305828,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004992517852735469,
      "loss": 6.5632,
      "mean_token_accuracy": 0.1513322427868843,
      "num_tokens": 8182096.0,
      "step": 3225
    },
    {
      "entropy": 6.5443480014801025,
      "epoch": 0.37276399307559144,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004992484190932517,
      "loss": 6.6038,
      "mean_token_accuracy": 0.14960480108857155,
      "num_tokens": 8193768.0,
      "step": 3230
    },
    {
      "entropy": 6.508570384979248,
      "epoch": 0.37334102712060013,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004992450453704723,
      "loss": 6.4491,
      "mean_token_accuracy": 0.1603356882929802,
      "num_tokens": 8206237.0,
      "step": 3235
    },
    {
      "entropy": 6.608931207656861,
      "epoch": 0.3739180611656088,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004992416641053223,
      "loss": 6.5241,
      "mean_token_accuracy": 0.15502073764801025,
      "num_tokens": 8218066.0,
      "step": 3240
    },
    {
      "entropy": 6.552175140380859,
      "epoch": 0.3744950952106174,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004992382752979153,
      "loss": 6.5236,
      "mean_token_accuracy": 0.15697094351053237,
      "num_tokens": 8230976.0,
      "step": 3245
    },
    {
      "entropy": 6.536435127258301,
      "epoch": 0.3750721292556261,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004992348789483651,
      "loss": 6.5137,
      "mean_token_accuracy": 0.15762929171323775,
      "num_tokens": 8242713.0,
      "step": 3250
    },
    {
      "entropy": 6.5722216129302975,
      "epoch": 0.37564916330063475,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004992314750567864,
      "loss": 6.4686,
      "mean_token_accuracy": 0.16449535489082337,
      "num_tokens": 8255698.0,
      "step": 3255
    },
    {
      "entropy": 6.551311063766479,
      "epoch": 0.3762261973456434,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004992280636232933,
      "loss": 6.5327,
      "mean_token_accuracy": 0.1546404018998146,
      "num_tokens": 8267138.0,
      "step": 3260
    },
    {
      "entropy": 6.594396638870239,
      "epoch": 0.376803231390652,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004992246446480007,
      "loss": 6.5459,
      "mean_token_accuracy": 0.15878639966249466,
      "num_tokens": 8280902.0,
      "step": 3265
    },
    {
      "entropy": 6.600580310821533,
      "epoch": 0.3773802654356607,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004992212181310235,
      "loss": 6.5744,
      "mean_token_accuracy": 0.15056225955486296,
      "num_tokens": 8293027.0,
      "step": 3270
    },
    {
      "entropy": 6.65857458114624,
      "epoch": 0.37795729948066936,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499217784072477,
      "loss": 6.5968,
      "mean_token_accuracy": 0.15336607545614242,
      "num_tokens": 8306676.0,
      "step": 3275
    },
    {
      "entropy": 6.475799512863159,
      "epoch": 0.378534333525678,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004992143424724768,
      "loss": 6.5126,
      "mean_token_accuracy": 0.16293894946575166,
      "num_tokens": 8319134.0,
      "step": 3280
    },
    {
      "entropy": 6.651707696914673,
      "epoch": 0.3791113675706867,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004992108933311385,
      "loss": 6.5057,
      "mean_token_accuracy": 0.145852892100811,
      "num_tokens": 8333010.0,
      "step": 3285
    },
    {
      "entropy": 6.6024363994598385,
      "epoch": 0.37968840161569534,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004992074366485782,
      "loss": 6.592,
      "mean_token_accuracy": 0.15076866745948792,
      "num_tokens": 8345576.0,
      "step": 3290
    },
    {
      "entropy": 6.502589273452759,
      "epoch": 0.380265435660704,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499203972424912,
      "loss": 6.5741,
      "mean_token_accuracy": 0.1434454709291458,
      "num_tokens": 8359278.0,
      "step": 3295
    },
    {
      "entropy": 6.6685864448547365,
      "epoch": 0.3808424697057126,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004992005006602567,
      "loss": 6.5626,
      "mean_token_accuracy": 0.1479098156094551,
      "num_tokens": 8372159.0,
      "step": 3300
    },
    {
      "entropy": 6.532323741912842,
      "epoch": 0.3814195037507213,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004991970213547289,
      "loss": 6.6116,
      "mean_token_accuracy": 0.14874994680285453,
      "num_tokens": 8385735.0,
      "step": 3305
    },
    {
      "entropy": 6.695924139022827,
      "epoch": 0.38199653779572995,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004991935345084457,
      "loss": 6.5832,
      "mean_token_accuracy": 0.1477282911539078,
      "num_tokens": 8397714.0,
      "step": 3310
    },
    {
      "entropy": 6.620160865783691,
      "epoch": 0.3825735718407386,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004991900401215243,
      "loss": 6.6035,
      "mean_token_accuracy": 0.14539917260408403,
      "num_tokens": 8410812.0,
      "step": 3315
    },
    {
      "entropy": 6.620796823501587,
      "epoch": 0.3831506058857473,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004991865381940822,
      "loss": 6.5807,
      "mean_token_accuracy": 0.15064923465251923,
      "num_tokens": 8422400.0,
      "step": 3320
    },
    {
      "entropy": 6.624812889099121,
      "epoch": 0.3837276399307559,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004991830287262374,
      "loss": 6.5833,
      "mean_token_accuracy": 0.15170842707157134,
      "num_tokens": 8435080.0,
      "step": 3325
    },
    {
      "entropy": 6.557571887969971,
      "epoch": 0.38430467397576457,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004991795117181077,
      "loss": 6.5339,
      "mean_token_accuracy": 0.15872213691473008,
      "num_tokens": 8447866.0,
      "step": 3330
    },
    {
      "entropy": 6.483943891525269,
      "epoch": 0.3848817080207732,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004991759871698113,
      "loss": 6.485,
      "mean_token_accuracy": 0.16049433052539824,
      "num_tokens": 8461447.0,
      "step": 3335
    },
    {
      "entropy": 6.662230443954468,
      "epoch": 0.3854587420657819,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004991724550814671,
      "loss": 6.6543,
      "mean_token_accuracy": 0.14789753630757332,
      "num_tokens": 8475389.0,
      "step": 3340
    },
    {
      "entropy": 6.635013771057129,
      "epoch": 0.38603577611079054,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004991689154531937,
      "loss": 6.552,
      "mean_token_accuracy": 0.14836361557245253,
      "num_tokens": 8487271.0,
      "step": 3345
    },
    {
      "entropy": 6.52984619140625,
      "epoch": 0.3866128101557992,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004991653682851102,
      "loss": 6.5535,
      "mean_token_accuracy": 0.1502477928996086,
      "num_tokens": 8498914.0,
      "step": 3350
    },
    {
      "entropy": 6.552432537078857,
      "epoch": 0.3871898442008078,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499161813577336,
      "loss": 6.541,
      "mean_token_accuracy": 0.15296554267406465,
      "num_tokens": 8510254.0,
      "step": 3355
    },
    {
      "entropy": 6.634548664093018,
      "epoch": 0.3877668782458165,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004991582513299903,
      "loss": 6.5702,
      "mean_token_accuracy": 0.15373844653367996,
      "num_tokens": 8522612.0,
      "step": 3360
    },
    {
      "entropy": 6.518693923950195,
      "epoch": 0.38834391229082516,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004991546815431932,
      "loss": 6.4915,
      "mean_token_accuracy": 0.15533529072999955,
      "num_tokens": 8534952.0,
      "step": 3365
    },
    {
      "entropy": 6.569821548461914,
      "epoch": 0.3889209463358338,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499151104217065,
      "loss": 6.6465,
      "mean_token_accuracy": 0.15146905109286307,
      "num_tokens": 8548388.0,
      "step": 3370
    },
    {
      "entropy": 6.572770690917968,
      "epoch": 0.3894979803808425,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004991475193517254,
      "loss": 6.4428,
      "mean_token_accuracy": 0.16474489867687225,
      "num_tokens": 8561576.0,
      "step": 3375
    },
    {
      "entropy": 6.5724358558654785,
      "epoch": 0.39007501442585113,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004991439269472956,
      "loss": 6.5859,
      "mean_token_accuracy": 0.15104893893003463,
      "num_tokens": 8574253.0,
      "step": 3380
    },
    {
      "entropy": 6.5863007545471195,
      "epoch": 0.39065204847085977,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004991403270038961,
      "loss": 6.4971,
      "mean_token_accuracy": 0.15619878172874452,
      "num_tokens": 8585708.0,
      "step": 3385
    },
    {
      "entropy": 6.504155683517456,
      "epoch": 0.3912290825158684,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000499136719521648,
      "loss": 6.5983,
      "mean_token_accuracy": 0.14018301516771317,
      "num_tokens": 8598327.0,
      "step": 3390
    },
    {
      "entropy": 6.713611268997193,
      "epoch": 0.3918061165608771,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991331045006726,
      "loss": 6.6013,
      "mean_token_accuracy": 0.14437992721796036,
      "num_tokens": 8610606.0,
      "step": 3395
    },
    {
      "entropy": 6.58232536315918,
      "epoch": 0.39238315060588574,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004991294819410916,
      "loss": 6.6182,
      "mean_token_accuracy": 0.1427798442542553,
      "num_tokens": 8625201.0,
      "step": 3400
    },
    {
      "entropy": 6.528783655166626,
      "epoch": 0.3929601846508944,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004991258518430268,
      "loss": 6.5796,
      "mean_token_accuracy": 0.1532195284962654,
      "num_tokens": 8638288.0,
      "step": 3405
    },
    {
      "entropy": 6.613002729415894,
      "epoch": 0.3935372186959031,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991222142066003,
      "loss": 6.5002,
      "mean_token_accuracy": 0.1590781554579735,
      "num_tokens": 8649550.0,
      "step": 3410
    },
    {
      "entropy": 6.498236513137817,
      "epoch": 0.3941142527409117,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991185690319345,
      "loss": 6.4992,
      "mean_token_accuracy": 0.15196617618203162,
      "num_tokens": 8660956.0,
      "step": 3415
    },
    {
      "entropy": 6.6136932373046875,
      "epoch": 0.39469128678592036,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499114916319152,
      "loss": 6.5988,
      "mean_token_accuracy": 0.14406019672751427,
      "num_tokens": 8675130.0,
      "step": 3420
    },
    {
      "entropy": 6.532426023483277,
      "epoch": 0.395268320830929,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004991112560683755,
      "loss": 6.4868,
      "mean_token_accuracy": 0.15448959022760392,
      "num_tokens": 8687482.0,
      "step": 3425
    },
    {
      "entropy": 6.506831121444702,
      "epoch": 0.3958453548759377,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004991075882797283,
      "loss": 6.5039,
      "mean_token_accuracy": 0.16184502243995666,
      "num_tokens": 8700202.0,
      "step": 3430
    },
    {
      "entropy": 6.613342761993408,
      "epoch": 0.39642238892094633,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004991039129533337,
      "loss": 6.6411,
      "mean_token_accuracy": 0.14476640224456788,
      "num_tokens": 8713602.0,
      "step": 3435
    },
    {
      "entropy": 6.5777974128723145,
      "epoch": 0.396999422965955,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004991002300893152,
      "loss": 6.6043,
      "mean_token_accuracy": 0.14983822107315065,
      "num_tokens": 8727318.0,
      "step": 3440
    },
    {
      "entropy": 6.565083408355713,
      "epoch": 0.39757645701096367,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004990965396877969,
      "loss": 6.4406,
      "mean_token_accuracy": 0.1597253128886223,
      "num_tokens": 8739941.0,
      "step": 3445
    },
    {
      "entropy": 6.549089574813843,
      "epoch": 0.3981534910559723,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004990928417489027,
      "loss": 6.5998,
      "mean_token_accuracy": 0.14621611833572387,
      "num_tokens": 8753012.0,
      "step": 3450
    },
    {
      "entropy": 6.584795570373535,
      "epoch": 0.39873052510098095,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004990891362727572,
      "loss": 6.5569,
      "mean_token_accuracy": 0.15298389941453933,
      "num_tokens": 8766998.0,
      "step": 3455
    },
    {
      "entropy": 6.552631950378418,
      "epoch": 0.3993075591459896,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004990854232594848,
      "loss": 6.5093,
      "mean_token_accuracy": 0.15525896847248077,
      "num_tokens": 8778993.0,
      "step": 3460
    },
    {
      "entropy": 6.482600927352905,
      "epoch": 0.3998845931909983,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004990817027092106,
      "loss": 6.484,
      "mean_token_accuracy": 0.1566115528345108,
      "num_tokens": 8790612.0,
      "step": 3465
    },
    {
      "entropy": 6.637128734588623,
      "epoch": 0.4004616272360069,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004990779746220595,
      "loss": 6.6259,
      "mean_token_accuracy": 0.1484322130680084,
      "num_tokens": 8803040.0,
      "step": 3470
    },
    {
      "entropy": 6.6322792053222654,
      "epoch": 0.40103866128101556,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004990742389981572,
      "loss": 6.5851,
      "mean_token_accuracy": 0.14987609088420867,
      "num_tokens": 8816179.0,
      "step": 3475
    },
    {
      "entropy": 6.61499514579773,
      "epoch": 0.40161569532602426,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499070495837629,
      "loss": 6.538,
      "mean_token_accuracy": 0.14199631735682489,
      "num_tokens": 8828779.0,
      "step": 3480
    },
    {
      "entropy": 6.608755779266358,
      "epoch": 0.4021927293710329,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004990667451406012,
      "loss": 6.5638,
      "mean_token_accuracy": 0.15815705060958862,
      "num_tokens": 8842354.0,
      "step": 3485
    },
    {
      "entropy": 6.539080858230591,
      "epoch": 0.40276976341604154,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004990629869071995,
      "loss": 6.5731,
      "mean_token_accuracy": 0.14226773679256438,
      "num_tokens": 8855462.0,
      "step": 3490
    },
    {
      "entropy": 6.637126016616821,
      "epoch": 0.4033467974610502,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004990592211375506,
      "loss": 6.56,
      "mean_token_accuracy": 0.14523446410894394,
      "num_tokens": 8867882.0,
      "step": 3495
    },
    {
      "entropy": 6.5148759365081785,
      "epoch": 0.4039238315060589,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990554478317811,
      "loss": 6.4861,
      "mean_token_accuracy": 0.15368588864803315,
      "num_tokens": 8880906.0,
      "step": 3500
    },
    {
      "entropy": 6.416658067703247,
      "epoch": 0.4045008655510675,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004990516669900179,
      "loss": 6.3869,
      "mean_token_accuracy": 0.1601685345172882,
      "num_tokens": 8893005.0,
      "step": 3505
    },
    {
      "entropy": 6.549545240402222,
      "epoch": 0.40507789959607615,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004990478786123882,
      "loss": 6.5644,
      "mean_token_accuracy": 0.14719341844320297,
      "num_tokens": 8905611.0,
      "step": 3510
    },
    {
      "entropy": 6.5877350807189945,
      "epoch": 0.40565493364108485,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004990440826990193,
      "loss": 6.5046,
      "mean_token_accuracy": 0.1505645200610161,
      "num_tokens": 8918454.0,
      "step": 3515
    },
    {
      "entropy": 6.525338506698608,
      "epoch": 0.4062319676860935,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004990402792500392,
      "loss": 6.5028,
      "mean_token_accuracy": 0.15014477223157882,
      "num_tokens": 8931871.0,
      "step": 3520
    },
    {
      "entropy": 6.546277666091919,
      "epoch": 0.4068090017311021,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004990364682655754,
      "loss": 6.5449,
      "mean_token_accuracy": 0.15401435047388076,
      "num_tokens": 8944813.0,
      "step": 3525
    },
    {
      "entropy": 6.594185495376587,
      "epoch": 0.40738603577611077,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004990326497457564,
      "loss": 6.5631,
      "mean_token_accuracy": 0.15211462527513503,
      "num_tokens": 8958902.0,
      "step": 3530
    },
    {
      "entropy": 6.552911853790283,
      "epoch": 0.40796306982111946,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004990288236907104,
      "loss": 6.5726,
      "mean_token_accuracy": 0.1494864970445633,
      "num_tokens": 8971390.0,
      "step": 3535
    },
    {
      "entropy": 6.557071876525879,
      "epoch": 0.4085401038661281,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004990249901005661,
      "loss": 6.5325,
      "mean_token_accuracy": 0.1544061541557312,
      "num_tokens": 8983510.0,
      "step": 3540
    },
    {
      "entropy": 6.551651954650879,
      "epoch": 0.40911713791113674,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004990211489754527,
      "loss": 6.5207,
      "mean_token_accuracy": 0.14719257056713103,
      "num_tokens": 8997039.0,
      "step": 3545
    },
    {
      "entropy": 6.619212245941162,
      "epoch": 0.40969417195614544,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004990173003154993,
      "loss": 6.63,
      "mean_token_accuracy": 0.14687275215983392,
      "num_tokens": 9009672.0,
      "step": 3550
    },
    {
      "entropy": 6.583910369873047,
      "epoch": 0.4102712060011541,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004990134441208351,
      "loss": 6.5561,
      "mean_token_accuracy": 0.15046931207180023,
      "num_tokens": 9023477.0,
      "step": 3555
    },
    {
      "entropy": 6.508993244171142,
      "epoch": 0.4108482400461627,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004990095803915901,
      "loss": 6.4403,
      "mean_token_accuracy": 0.1626611977815628,
      "num_tokens": 9035914.0,
      "step": 3560
    },
    {
      "entropy": 6.534296274185181,
      "epoch": 0.41142527409117136,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004990057091278942,
      "loss": 6.5265,
      "mean_token_accuracy": 0.14852816760540008,
      "num_tokens": 9048205.0,
      "step": 3565
    },
    {
      "entropy": 6.614375734329224,
      "epoch": 0.41200230813618005,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004990018303298773,
      "loss": 6.5939,
      "mean_token_accuracy": 0.14754611253738403,
      "num_tokens": 9061467.0,
      "step": 3570
    },
    {
      "entropy": 6.639231014251709,
      "epoch": 0.4125793421811887,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004989979439976703,
      "loss": 6.5418,
      "mean_token_accuracy": 0.14865971505641937,
      "num_tokens": 9074434.0,
      "step": 3575
    },
    {
      "entropy": 6.439689636230469,
      "epoch": 0.41315637622619733,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004989940501314037,
      "loss": 6.4373,
      "mean_token_accuracy": 0.1580505609512329,
      "num_tokens": 9087745.0,
      "step": 3580
    },
    {
      "entropy": 6.537207984924317,
      "epoch": 0.413733410271206,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004989901487312084,
      "loss": 6.5044,
      "mean_token_accuracy": 0.1544773370027542,
      "num_tokens": 9099424.0,
      "step": 3585
    },
    {
      "entropy": 6.555319881439209,
      "epoch": 0.41431044431621467,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004989862397972157,
      "loss": 6.4647,
      "mean_token_accuracy": 0.15824805945158005,
      "num_tokens": 9111056.0,
      "step": 3590
    },
    {
      "entropy": 6.523460960388183,
      "epoch": 0.4148874783612233,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004989823233295572,
      "loss": 6.616,
      "mean_token_accuracy": 0.15002714097499847,
      "num_tokens": 9124539.0,
      "step": 3595
    },
    {
      "entropy": 6.591261959075927,
      "epoch": 0.41546451240623195,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004989783993283644,
      "loss": 6.5006,
      "mean_token_accuracy": 0.15340062379837036,
      "num_tokens": 9136580.0,
      "step": 3600
    },
    {
      "entropy": 6.5560140132904055,
      "epoch": 0.41604154645124064,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004989744677937694,
      "loss": 6.558,
      "mean_token_accuracy": 0.1486823797225952,
      "num_tokens": 9148411.0,
      "step": 3605
    },
    {
      "entropy": 6.590163278579712,
      "epoch": 0.4166185804962493,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004989705287259046,
      "loss": 6.4926,
      "mean_token_accuracy": 0.15736011117696763,
      "num_tokens": 9160778.0,
      "step": 3610
    },
    {
      "entropy": 6.539517545700074,
      "epoch": 0.4171956145412579,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004989665821249021,
      "loss": 6.5469,
      "mean_token_accuracy": 0.15103208273649216,
      "num_tokens": 9174174.0,
      "step": 3615
    },
    {
      "entropy": 6.554589176177979,
      "epoch": 0.4177726485862666,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000498962627990895,
      "loss": 6.498,
      "mean_token_accuracy": 0.15684661269187927,
      "num_tokens": 9186465.0,
      "step": 3620
    },
    {
      "entropy": 6.476647901535034,
      "epoch": 0.41834968263127525,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004989586663240161,
      "loss": 6.4983,
      "mean_token_accuracy": 0.1546318955719471,
      "num_tokens": 9198900.0,
      "step": 3625
    },
    {
      "entropy": 6.642714405059815,
      "epoch": 0.4189267166762839,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004989546971243988,
      "loss": 6.5135,
      "mean_token_accuracy": 0.15572796911001205,
      "num_tokens": 9211148.0,
      "step": 3630
    },
    {
      "entropy": 6.575963926315308,
      "epoch": 0.41950375072129253,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004989507203921763,
      "loss": 6.6052,
      "mean_token_accuracy": 0.14590780809521675,
      "num_tokens": 9226216.0,
      "step": 3635
    },
    {
      "entropy": 6.622480678558349,
      "epoch": 0.42008078476630123,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004989467361274828,
      "loss": 6.5449,
      "mean_token_accuracy": 0.15304823964834213,
      "num_tokens": 9238361.0,
      "step": 3640
    },
    {
      "entropy": 6.509000062942505,
      "epoch": 0.42065781881130987,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004989427443304519,
      "loss": 6.4601,
      "mean_token_accuracy": 0.15948846489191054,
      "num_tokens": 9251828.0,
      "step": 3645
    },
    {
      "entropy": 6.559967422485352,
      "epoch": 0.4212348528563185,
      "grad_norm": 1.0,
      "learning_rate": 0.000498938745001218,
      "loss": 6.5181,
      "mean_token_accuracy": 0.15092962980270386,
      "num_tokens": 9264433.0,
      "step": 3650
    },
    {
      "entropy": 6.6109813213348385,
      "epoch": 0.4218118869013272,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004989347381399158,
      "loss": 6.5411,
      "mean_token_accuracy": 0.15280013978481294,
      "num_tokens": 9277859.0,
      "step": 3655
    },
    {
      "entropy": 6.520666599273682,
      "epoch": 0.42238892094633584,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004989307237466799,
      "loss": 6.5641,
      "mean_token_accuracy": 0.14598092585802078,
      "num_tokens": 9290585.0,
      "step": 3660
    },
    {
      "entropy": 6.613101196289063,
      "epoch": 0.4229659549913445,
      "grad_norm": 1.0,
      "learning_rate": 0.0004989267018216453,
      "loss": 6.5681,
      "mean_token_accuracy": 0.1445908688008785,
      "num_tokens": 9302633.0,
      "step": 3665
    },
    {
      "entropy": 6.561917877197265,
      "epoch": 0.4235429890363531,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004989226723649473,
      "loss": 6.5976,
      "mean_token_accuracy": 0.14954569786787034,
      "num_tokens": 9314577.0,
      "step": 3670
    },
    {
      "entropy": 6.520503234863281,
      "epoch": 0.4241200230813618,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004989186353767214,
      "loss": 6.4554,
      "mean_token_accuracy": 0.15625739395618438,
      "num_tokens": 9327260.0,
      "step": 3675
    },
    {
      "entropy": 6.538907718658447,
      "epoch": 0.42469705712637046,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004989145908571035,
      "loss": 6.5312,
      "mean_token_accuracy": 0.15637308359146118,
      "num_tokens": 9339684.0,
      "step": 3680
    },
    {
      "entropy": 6.546295690536499,
      "epoch": 0.4252740911713791,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004989105388062295,
      "loss": 6.5426,
      "mean_token_accuracy": 0.15097250938415527,
      "num_tokens": 9352532.0,
      "step": 3685
    },
    {
      "entropy": 6.64055027961731,
      "epoch": 0.4258511252163878,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004989064792242358,
      "loss": 6.5626,
      "mean_token_accuracy": 0.1484258934855461,
      "num_tokens": 9364680.0,
      "step": 3690
    },
    {
      "entropy": 6.6101151466369625,
      "epoch": 0.42642815926139643,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004989024121112589,
      "loss": 6.5283,
      "mean_token_accuracy": 0.15172728449106215,
      "num_tokens": 9377472.0,
      "step": 3695
    },
    {
      "entropy": 6.551726818084717,
      "epoch": 0.4270051933064051,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004988983374674356,
      "loss": 6.5669,
      "mean_token_accuracy": 0.1464052118360996,
      "num_tokens": 9391646.0,
      "step": 3700
    },
    {
      "entropy": 6.570635223388672,
      "epoch": 0.4275822273514137,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004988942552929029,
      "loss": 6.5009,
      "mean_token_accuracy": 0.14611846059560776,
      "num_tokens": 9402864.0,
      "step": 3705
    },
    {
      "entropy": 6.598347139358521,
      "epoch": 0.4281592613964224,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004988901655877981,
      "loss": 6.5095,
      "mean_token_accuracy": 0.15050406157970428,
      "num_tokens": 9414498.0,
      "step": 3710
    },
    {
      "entropy": 6.510675382614136,
      "epoch": 0.42873629544143105,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004988860683522589,
      "loss": 6.5526,
      "mean_token_accuracy": 0.15163294076919556,
      "num_tokens": 9427903.0,
      "step": 3715
    },
    {
      "entropy": 6.571003532409668,
      "epoch": 0.4293133294864397,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000498881963586423,
      "loss": 6.5115,
      "mean_token_accuracy": 0.1570156693458557,
      "num_tokens": 9441566.0,
      "step": 3720
    },
    {
      "entropy": 6.459043407440186,
      "epoch": 0.4298903635314484,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004988778512904282,
      "loss": 6.3985,
      "mean_token_accuracy": 0.1653307482600212,
      "num_tokens": 9453614.0,
      "step": 3725
    },
    {
      "entropy": 6.614306831359864,
      "epoch": 0.430467397576457,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004988737314644134,
      "loss": 6.5521,
      "mean_token_accuracy": 0.14477383494377136,
      "num_tokens": 9468046.0,
      "step": 3730
    },
    {
      "entropy": 6.440754842758179,
      "epoch": 0.43104443162146566,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004988696041085168,
      "loss": 6.4568,
      "mean_token_accuracy": 0.1593452662229538,
      "num_tokens": 9481469.0,
      "step": 3735
    },
    {
      "entropy": 6.5900390625,
      "epoch": 0.4316214656664743,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004988654692228772,
      "loss": 6.5369,
      "mean_token_accuracy": 0.15275566279888153,
      "num_tokens": 9492833.0,
      "step": 3740
    },
    {
      "entropy": 6.569700098037719,
      "epoch": 0.432198499711483,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004988613268076335,
      "loss": 6.4576,
      "mean_token_accuracy": 0.14876370429992675,
      "num_tokens": 9504955.0,
      "step": 3745
    },
    {
      "entropy": 6.529074096679688,
      "epoch": 0.43277553375649164,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004988571768629257,
      "loss": 6.4753,
      "mean_token_accuracy": 0.15705748796463012,
      "num_tokens": 9516830.0,
      "step": 3750
    },
    {
      "entropy": 6.5411705493927,
      "epoch": 0.4333525678015003,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004988530193888927,
      "loss": 6.5,
      "mean_token_accuracy": 0.15189379155635835,
      "num_tokens": 9528965.0,
      "step": 3755
    },
    {
      "entropy": 6.62169885635376,
      "epoch": 0.4339296018465089,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004988488543856747,
      "loss": 6.5829,
      "mean_token_accuracy": 0.14869293197989464,
      "num_tokens": 9541838.0,
      "step": 3760
    },
    {
      "entropy": 6.534994745254517,
      "epoch": 0.4345066358915176,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004988446818534115,
      "loss": 6.5117,
      "mean_token_accuracy": 0.15386671274900438,
      "num_tokens": 9554502.0,
      "step": 3765
    },
    {
      "entropy": 6.514486980438233,
      "epoch": 0.43508366993652625,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004988405017922438,
      "loss": 6.3677,
      "mean_token_accuracy": 0.1565380185842514,
      "num_tokens": 9567028.0,
      "step": 3770
    },
    {
      "entropy": 6.524361276626587,
      "epoch": 0.4356607039815349,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000498836314202312,
      "loss": 6.5308,
      "mean_token_accuracy": 0.14339614510536194,
      "num_tokens": 9580336.0,
      "step": 3775
    },
    {
      "entropy": 6.575292921066284,
      "epoch": 0.4362377380265436,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004988321190837568,
      "loss": 6.4711,
      "mean_token_accuracy": 0.15123758614063262,
      "num_tokens": 9592410.0,
      "step": 3780
    },
    {
      "entropy": 6.532257223129273,
      "epoch": 0.4368147720715522,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004988279164367196,
      "loss": 6.4888,
      "mean_token_accuracy": 0.14704009145498276,
      "num_tokens": 9605693.0,
      "step": 3785
    },
    {
      "entropy": 6.507564067840576,
      "epoch": 0.43739180611656087,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004988237062613415,
      "loss": 6.5066,
      "mean_token_accuracy": 0.1507734850049019,
      "num_tokens": 9618555.0,
      "step": 3790
    },
    {
      "entropy": 6.520890188217163,
      "epoch": 0.4379688401615695,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004988194885577644,
      "loss": 6.5205,
      "mean_token_accuracy": 0.14715251475572586,
      "num_tokens": 9630330.0,
      "step": 3795
    },
    {
      "entropy": 6.5729286670684814,
      "epoch": 0.4385458742065782,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004988152633261299,
      "loss": 6.5129,
      "mean_token_accuracy": 0.15779052078723907,
      "num_tokens": 9642516.0,
      "step": 3800
    },
    {
      "entropy": 6.564908695220947,
      "epoch": 0.43912290825158684,
      "grad_norm": 1.0625,
      "learning_rate": 0.00049881103056658,
      "loss": 6.5034,
      "mean_token_accuracy": 0.1489268183708191,
      "num_tokens": 9654886.0,
      "step": 3805
    },
    {
      "entropy": 6.578825187683106,
      "epoch": 0.4396999422965955,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004988067902792575,
      "loss": 6.5055,
      "mean_token_accuracy": 0.1582085445523262,
      "num_tokens": 9666346.0,
      "step": 3810
    },
    {
      "entropy": 6.4737834453582765,
      "epoch": 0.4402769763416042,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004988025424643047,
      "loss": 6.496,
      "mean_token_accuracy": 0.14756360203027724,
      "num_tokens": 9679971.0,
      "step": 3815
    },
    {
      "entropy": 6.584052419662475,
      "epoch": 0.4408540103866128,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004987982871218645,
      "loss": 6.5718,
      "mean_token_accuracy": 0.14819382280111312,
      "num_tokens": 9693032.0,
      "step": 3820
    },
    {
      "entropy": 6.4983185768127445,
      "epoch": 0.44143104443162146,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004987940242520802,
      "loss": 6.4763,
      "mean_token_accuracy": 0.15692102909088135,
      "num_tokens": 9705706.0,
      "step": 3825
    },
    {
      "entropy": 6.431740427017212,
      "epoch": 0.4420080784766301,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000498789753855095,
      "loss": 6.4184,
      "mean_token_accuracy": 0.15480095595121385,
      "num_tokens": 9718235.0,
      "step": 3830
    },
    {
      "entropy": 6.588514423370361,
      "epoch": 0.4425851125216388,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004987854759310526,
      "loss": 6.5646,
      "mean_token_accuracy": 0.1443893849849701,
      "num_tokens": 9730758.0,
      "step": 3835
    },
    {
      "entropy": 6.4969950199127195,
      "epoch": 0.44316214656664743,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004987811904800968,
      "loss": 6.4085,
      "mean_token_accuracy": 0.16231226474046706,
      "num_tokens": 9743131.0,
      "step": 3840
    },
    {
      "entropy": 6.4805018424987795,
      "epoch": 0.44373918061165607,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004987768975023719,
      "loss": 6.4915,
      "mean_token_accuracy": 0.1469715990126133,
      "num_tokens": 9755965.0,
      "step": 3845
    },
    {
      "entropy": 6.591073417663575,
      "epoch": 0.44431621465666477,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004987725969980222,
      "loss": 6.3983,
      "mean_token_accuracy": 0.16017842739820481,
      "num_tokens": 9768460.0,
      "step": 3850
    },
    {
      "entropy": 6.490756368637085,
      "epoch": 0.4448932487016734,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004987682889671923,
      "loss": 6.5747,
      "mean_token_accuracy": 0.14486720710992812,
      "num_tokens": 9781653.0,
      "step": 3855
    },
    {
      "entropy": 6.54525899887085,
      "epoch": 0.44547028274668204,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004987639734100272,
      "loss": 6.3933,
      "mean_token_accuracy": 0.16008301600813865,
      "num_tokens": 9793814.0,
      "step": 3860
    },
    {
      "entropy": 6.527516651153564,
      "epoch": 0.4460473167916907,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004987596503266721,
      "loss": 6.4588,
      "mean_token_accuracy": 0.15815410763025284,
      "num_tokens": 9806635.0,
      "step": 3865
    },
    {
      "entropy": 6.438649272918701,
      "epoch": 0.4466243508366994,
      "grad_norm": 1.125,
      "learning_rate": 0.0004987553197172722,
      "loss": 6.4573,
      "mean_token_accuracy": 0.1540544807910919,
      "num_tokens": 9819213.0,
      "step": 3870
    },
    {
      "entropy": 6.522204399108887,
      "epoch": 0.447201384881708,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004987509815819732,
      "loss": 6.434,
      "mean_token_accuracy": 0.15137404426932335,
      "num_tokens": 9831823.0,
      "step": 3875
    },
    {
      "entropy": 6.457322072982788,
      "epoch": 0.44777841892671666,
      "grad_norm": 1.25,
      "learning_rate": 0.0004987466359209213,
      "loss": 6.4255,
      "mean_token_accuracy": 0.15407421290874482,
      "num_tokens": 9844993.0,
      "step": 3880
    },
    {
      "entropy": 6.389491128921509,
      "epoch": 0.44835545297172535,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004987422827342622,
      "loss": 6.267,
      "mean_token_accuracy": 0.16830738484859467,
      "num_tokens": 9857853.0,
      "step": 3885
    },
    {
      "entropy": 6.509896183013916,
      "epoch": 0.448932487016734,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004987379220221426,
      "loss": 6.5172,
      "mean_token_accuracy": 0.15210521966218948,
      "num_tokens": 9871465.0,
      "step": 3890
    },
    {
      "entropy": 6.567883157730103,
      "epoch": 0.44950952106174263,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004987335537847092,
      "loss": 6.4992,
      "mean_token_accuracy": 0.15913779884576798,
      "num_tokens": 9884973.0,
      "step": 3895
    },
    {
      "entropy": 6.604511165618897,
      "epoch": 0.4500865551067513,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004987291780221088,
      "loss": 6.63,
      "mean_token_accuracy": 0.14119090288877487,
      "num_tokens": 9898143.0,
      "step": 3900
    },
    {
      "entropy": 6.6406971454620365,
      "epoch": 0.45066358915175997,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004987247947344887,
      "loss": 6.5488,
      "mean_token_accuracy": 0.15167086869478225,
      "num_tokens": 9910584.0,
      "step": 3905
    },
    {
      "entropy": 6.498317432403565,
      "epoch": 0.4512406231967686,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987204039219962,
      "loss": 6.5933,
      "mean_token_accuracy": 0.1443414121866226,
      "num_tokens": 9922901.0,
      "step": 3910
    },
    {
      "entropy": 6.630822372436524,
      "epoch": 0.45181765724177725,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004987160055847791,
      "loss": 6.5785,
      "mean_token_accuracy": 0.1416196659207344,
      "num_tokens": 9936131.0,
      "step": 3915
    },
    {
      "entropy": 6.5027612209320065,
      "epoch": 0.45239469128678594,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004987115997229852,
      "loss": 6.3256,
      "mean_token_accuracy": 0.16678364723920822,
      "num_tokens": 9950292.0,
      "step": 3920
    },
    {
      "entropy": 6.439765405654907,
      "epoch": 0.4529717253317946,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004987071863367629,
      "loss": 6.5071,
      "mean_token_accuracy": 0.1504189372062683,
      "num_tokens": 9963765.0,
      "step": 3925
    },
    {
      "entropy": 6.531398010253906,
      "epoch": 0.4535487593768032,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004987027654262604,
      "loss": 6.4726,
      "mean_token_accuracy": 0.15666739046573638,
      "num_tokens": 9975981.0,
      "step": 3930
    },
    {
      "entropy": 6.490634346008301,
      "epoch": 0.45412579342181186,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004986983369916264,
      "loss": 6.4564,
      "mean_token_accuracy": 0.15376958549022673,
      "num_tokens": 9988763.0,
      "step": 3935
    },
    {
      "entropy": 6.440743064880371,
      "epoch": 0.45470282746682056,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004986939010330102,
      "loss": 6.4546,
      "mean_token_accuracy": 0.15868337750434874,
      "num_tokens": 10000663.0,
      "step": 3940
    },
    {
      "entropy": 6.533361148834229,
      "epoch": 0.4552798615118292,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004986894575505606,
      "loss": 6.4557,
      "mean_token_accuracy": 0.15553340166807175,
      "num_tokens": 10012673.0,
      "step": 3945
    },
    {
      "entropy": 6.474095296859741,
      "epoch": 0.45585689555683784,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004986850065444272,
      "loss": 6.498,
      "mean_token_accuracy": 0.14931581914424896,
      "num_tokens": 10024410.0,
      "step": 3950
    },
    {
      "entropy": 6.494760465621948,
      "epoch": 0.45643392960184653,
      "grad_norm": 1.125,
      "learning_rate": 0.0004986805480147598,
      "loss": 6.3974,
      "mean_token_accuracy": 0.1665262684226036,
      "num_tokens": 10036782.0,
      "step": 3955
    },
    {
      "entropy": 6.505864000320434,
      "epoch": 0.4570109636468552,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004986760819617082,
      "loss": 6.4771,
      "mean_token_accuracy": 0.15344146639108658,
      "num_tokens": 10049234.0,
      "step": 3960
    },
    {
      "entropy": 6.5312902450561525,
      "epoch": 0.4575879976918638,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004986716083854228,
      "loss": 6.5078,
      "mean_token_accuracy": 0.15330180525779724,
      "num_tokens": 10061853.0,
      "step": 3965
    },
    {
      "entropy": 6.644676733016968,
      "epoch": 0.45816503173687245,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004986671272860538,
      "loss": 6.5368,
      "mean_token_accuracy": 0.14532949328422545,
      "num_tokens": 10074890.0,
      "step": 3970
    },
    {
      "entropy": 6.538569355010987,
      "epoch": 0.45874206578188115,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004986626386637521,
      "loss": 6.5778,
      "mean_token_accuracy": 0.146486609429121,
      "num_tokens": 10090226.0,
      "step": 3975
    },
    {
      "entropy": 6.519270896911621,
      "epoch": 0.4593190998268898,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004986581425186688,
      "loss": 6.523,
      "mean_token_accuracy": 0.1502547562122345,
      "num_tokens": 10103716.0,
      "step": 3980
    },
    {
      "entropy": 6.607999801635742,
      "epoch": 0.4598961338718984,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004986536388509548,
      "loss": 6.6513,
      "mean_token_accuracy": 0.14018961787223816,
      "num_tokens": 10117553.0,
      "step": 3985
    },
    {
      "entropy": 6.557234477996826,
      "epoch": 0.4604731679169071,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004986491276607618,
      "loss": 6.4643,
      "mean_token_accuracy": 0.1531487300992012,
      "num_tokens": 10129897.0,
      "step": 3990
    },
    {
      "entropy": 6.659291791915893,
      "epoch": 0.46105020196191576,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004986446089482416,
      "loss": 6.6046,
      "mean_token_accuracy": 0.1465468868613243,
      "num_tokens": 10141974.0,
      "step": 3995
    },
    {
      "entropy": 6.536488389968872,
      "epoch": 0.4616272360069244,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004986400827135459,
      "loss": 6.5097,
      "mean_token_accuracy": 0.14302397817373275,
      "num_tokens": 10155471.0,
      "step": 4000
    },
    {
      "entropy": 6.5658046245574955,
      "epoch": 0.46220427005193304,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004986355489568272,
      "loss": 6.4691,
      "mean_token_accuracy": 0.15055305808782576,
      "num_tokens": 10167362.0,
      "step": 4005
    },
    {
      "entropy": 6.537802076339721,
      "epoch": 0.46278130409694174,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004986310076782379,
      "loss": 6.4977,
      "mean_token_accuracy": 0.15193159133195877,
      "num_tokens": 10180706.0,
      "step": 4010
    },
    {
      "entropy": 6.592318201065064,
      "epoch": 0.4633583381419504,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004986264588779307,
      "loss": 6.5293,
      "mean_token_accuracy": 0.14840541929006576,
      "num_tokens": 10193642.0,
      "step": 4015
    },
    {
      "entropy": 6.5473737716674805,
      "epoch": 0.463935372186959,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004986219025560586,
      "loss": 6.5426,
      "mean_token_accuracy": 0.14978598654270173,
      "num_tokens": 10206552.0,
      "step": 4020
    },
    {
      "entropy": 6.49361743927002,
      "epoch": 0.4645124062319677,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000498617338712775,
      "loss": 6.5333,
      "mean_token_accuracy": 0.1523551255464554,
      "num_tokens": 10219576.0,
      "step": 4025
    },
    {
      "entropy": 6.594914531707763,
      "epoch": 0.46508944027697635,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004986127673482332,
      "loss": 6.45,
      "mean_token_accuracy": 0.14977422058582307,
      "num_tokens": 10231816.0,
      "step": 4030
    },
    {
      "entropy": 6.419863367080689,
      "epoch": 0.465666474321985,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004986081884625871,
      "loss": 6.3446,
      "mean_token_accuracy": 0.1564165748655796,
      "num_tokens": 10244329.0,
      "step": 4035
    },
    {
      "entropy": 6.477906513214111,
      "epoch": 0.46624350836699363,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004986036020559906,
      "loss": 6.4702,
      "mean_token_accuracy": 0.14742427617311477,
      "num_tokens": 10256521.0,
      "step": 4040
    },
    {
      "entropy": 6.522656869888306,
      "epoch": 0.4668205424120023,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000498599008128598,
      "loss": 6.4637,
      "mean_token_accuracy": 0.15066753178834916,
      "num_tokens": 10269023.0,
      "step": 4045
    },
    {
      "entropy": 6.550335311889649,
      "epoch": 0.46739757645701097,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004985944066805639,
      "loss": 6.5303,
      "mean_token_accuracy": 0.145648692548275,
      "num_tokens": 10281476.0,
      "step": 4050
    },
    {
      "entropy": 6.533018398284912,
      "epoch": 0.4679746105020196,
      "grad_norm": 1.046875,
      "learning_rate": 0.000498589797712043,
      "loss": 6.4314,
      "mean_token_accuracy": 0.1552262082695961,
      "num_tokens": 10293749.0,
      "step": 4055
    },
    {
      "entropy": 6.530173254013062,
      "epoch": 0.4685516445470283,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004985851812231903,
      "loss": 6.4491,
      "mean_token_accuracy": 0.1614367574453354,
      "num_tokens": 10306554.0,
      "step": 4060
    },
    {
      "entropy": 6.535722732543945,
      "epoch": 0.46912867859203694,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004985805572141611,
      "loss": 6.4918,
      "mean_token_accuracy": 0.15370505601167678,
      "num_tokens": 10319374.0,
      "step": 4065
    },
    {
      "entropy": 6.504807853698731,
      "epoch": 0.4697057126370456,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004985759256851108,
      "loss": 6.5402,
      "mean_token_accuracy": 0.148026804625988,
      "num_tokens": 10333358.0,
      "step": 4070
    },
    {
      "entropy": 6.581374835968018,
      "epoch": 0.4702827466820542,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004985712866361954,
      "loss": 6.4693,
      "mean_token_accuracy": 0.1560882419347763,
      "num_tokens": 10344830.0,
      "step": 4075
    },
    {
      "entropy": 6.525490379333496,
      "epoch": 0.4708597807270629,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004985666400675709,
      "loss": 6.5393,
      "mean_token_accuracy": 0.14634379893541336,
      "num_tokens": 10357813.0,
      "step": 4080
    },
    {
      "entropy": 6.626258659362793,
      "epoch": 0.47143681477207156,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004985619859793934,
      "loss": 6.5771,
      "mean_token_accuracy": 0.14652895107865332,
      "num_tokens": 10370925.0,
      "step": 4085
    },
    {
      "entropy": 6.626109457015991,
      "epoch": 0.4720138488170802,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004985573243718195,
      "loss": 6.5273,
      "mean_token_accuracy": 0.15081269145011902,
      "num_tokens": 10384172.0,
      "step": 4090
    },
    {
      "entropy": 6.53129997253418,
      "epoch": 0.4725908828620889,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004985526552450061,
      "loss": 6.4578,
      "mean_token_accuracy": 0.15571485012769698,
      "num_tokens": 10396797.0,
      "step": 4095
    },
    {
      "entropy": 6.435747146606445,
      "epoch": 0.47316791690709753,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004985479785991103,
      "loss": 6.5141,
      "mean_token_accuracy": 0.15079636201262475,
      "num_tokens": 10409139.0,
      "step": 4100
    },
    {
      "entropy": 6.662794017791748,
      "epoch": 0.47374495095210617,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004985432944342892,
      "loss": 6.4464,
      "mean_token_accuracy": 0.15483423620462416,
      "num_tokens": 10421442.0,
      "step": 4105
    },
    {
      "entropy": 6.48177399635315,
      "epoch": 0.4743219849971148,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004985386027507003,
      "loss": 6.451,
      "mean_token_accuracy": 0.15766309797763825,
      "num_tokens": 10433761.0,
      "step": 4110
    },
    {
      "entropy": 6.481580543518066,
      "epoch": 0.4748990190421235,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004985339035485018,
      "loss": 6.5378,
      "mean_token_accuracy": 0.15052165538072587,
      "num_tokens": 10446508.0,
      "step": 4115
    },
    {
      "entropy": 6.603940057754516,
      "epoch": 0.47547605308713214,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004985291968278513,
      "loss": 6.4444,
      "mean_token_accuracy": 0.16052599251270294,
      "num_tokens": 10459352.0,
      "step": 4120
    },
    {
      "entropy": 6.517511796951294,
      "epoch": 0.4760530871321408,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004985244825889074,
      "loss": 6.5074,
      "mean_token_accuracy": 0.14690134227275847,
      "num_tokens": 10470324.0,
      "step": 4125
    },
    {
      "entropy": 6.570990753173828,
      "epoch": 0.4766301211771495,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004985197608318284,
      "loss": 6.4878,
      "mean_token_accuracy": 0.14887885600328446,
      "num_tokens": 10482715.0,
      "step": 4130
    },
    {
      "entropy": 6.609026861190796,
      "epoch": 0.4772071552221581,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004985150315567735,
      "loss": 6.487,
      "mean_token_accuracy": 0.1574724867939949,
      "num_tokens": 10493207.0,
      "step": 4135
    },
    {
      "entropy": 6.378232765197754,
      "epoch": 0.47778418926716676,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004985102947639013,
      "loss": 6.4334,
      "mean_token_accuracy": 0.15071919709444045,
      "num_tokens": 10505187.0,
      "step": 4140
    },
    {
      "entropy": 6.598970556259156,
      "epoch": 0.4783612233121754,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004985055504533715,
      "loss": 6.4992,
      "mean_token_accuracy": 0.15731345862150192,
      "num_tokens": 10518301.0,
      "step": 4145
    },
    {
      "entropy": 6.497900485992432,
      "epoch": 0.4789382573571841,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004985007986253435,
      "loss": 6.4883,
      "mean_token_accuracy": 0.1517148196697235,
      "num_tokens": 10531288.0,
      "step": 4150
    },
    {
      "entropy": 6.472272443771362,
      "epoch": 0.47951529140219273,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004984960392799772,
      "loss": 6.3708,
      "mean_token_accuracy": 0.16418855488300324,
      "num_tokens": 10544120.0,
      "step": 4155
    },
    {
      "entropy": 6.441378259658814,
      "epoch": 0.4800923254472014,
      "grad_norm": 1.0,
      "learning_rate": 0.0004984912724174326,
      "loss": 6.4619,
      "mean_token_accuracy": 0.16539474278688432,
      "num_tokens": 10556453.0,
      "step": 4160
    },
    {
      "entropy": 6.62937822341919,
      "epoch": 0.48066935949221,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004984864980378701,
      "loss": 6.562,
      "mean_token_accuracy": 0.14174575582146645,
      "num_tokens": 10569404.0,
      "step": 4165
    },
    {
      "entropy": 6.56431303024292,
      "epoch": 0.4812463935372187,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004984817161414502,
      "loss": 6.4701,
      "mean_token_accuracy": 0.14950313568115234,
      "num_tokens": 10582312.0,
      "step": 4170
    },
    {
      "entropy": 6.535763168334961,
      "epoch": 0.48182342758222735,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004984769267283338,
      "loss": 6.5436,
      "mean_token_accuracy": 0.15009379237890244,
      "num_tokens": 10594391.0,
      "step": 4175
    },
    {
      "entropy": 6.580943632125854,
      "epoch": 0.482400461627236,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004984721297986819,
      "loss": 6.4553,
      "mean_token_accuracy": 0.15187350213527678,
      "num_tokens": 10606354.0,
      "step": 4180
    },
    {
      "entropy": 6.471270561218262,
      "epoch": 0.4829774956722447,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004984673253526561,
      "loss": 6.4609,
      "mean_token_accuracy": 0.1508080005645752,
      "num_tokens": 10618051.0,
      "step": 4185
    },
    {
      "entropy": 6.506736993789673,
      "epoch": 0.4835545297172533,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004984625133904176,
      "loss": 6.3627,
      "mean_token_accuracy": 0.16271423697471618,
      "num_tokens": 10629688.0,
      "step": 4190
    },
    {
      "entropy": 6.526623010635376,
      "epoch": 0.48413156376226196,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004984576939121286,
      "loss": 6.5112,
      "mean_token_accuracy": 0.14801880121231079,
      "num_tokens": 10642249.0,
      "step": 4195
    },
    {
      "entropy": 6.464288949966431,
      "epoch": 0.4847085978072706,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004984528669179511,
      "loss": 6.4527,
      "mean_token_accuracy": 0.15741877555847167,
      "num_tokens": 10655231.0,
      "step": 4200
    },
    {
      "entropy": 6.433733367919922,
      "epoch": 0.4852856318522793,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004984480324080472,
      "loss": 6.433,
      "mean_token_accuracy": 0.15953529626131058,
      "num_tokens": 10668443.0,
      "step": 4205
    },
    {
      "entropy": 6.567685604095459,
      "epoch": 0.48586266589728794,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00049844319038258,
      "loss": 6.4526,
      "mean_token_accuracy": 0.14846644103527068,
      "num_tokens": 10680523.0,
      "step": 4210
    },
    {
      "entropy": 6.609936475753784,
      "epoch": 0.4864396999422966,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004984383408417119,
      "loss": 6.4982,
      "mean_token_accuracy": 0.15419950783252717,
      "num_tokens": 10693223.0,
      "step": 4215
    },
    {
      "entropy": 6.461764812469482,
      "epoch": 0.4870167339873053,
      "grad_norm": 1.125,
      "learning_rate": 0.0004984334837856064,
      "loss": 6.4814,
      "mean_token_accuracy": 0.15989888161420823,
      "num_tokens": 10705276.0,
      "step": 4220
    },
    {
      "entropy": 6.499186706542969,
      "epoch": 0.4875937680323139,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004984286192144263,
      "loss": 6.4839,
      "mean_token_accuracy": 0.15633203238248825,
      "num_tokens": 10718495.0,
      "step": 4225
    },
    {
      "entropy": 6.529287338256836,
      "epoch": 0.48817080207732255,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004984237471283359,
      "loss": 6.4068,
      "mean_token_accuracy": 0.16196464896202087,
      "num_tokens": 10730885.0,
      "step": 4230
    },
    {
      "entropy": 6.472171211242676,
      "epoch": 0.4887478361223312,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004984188675274986,
      "loss": 6.461,
      "mean_token_accuracy": 0.1529358334839344,
      "num_tokens": 10742906.0,
      "step": 4235
    },
    {
      "entropy": 6.449957227706909,
      "epoch": 0.4893248701673399,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004984139804120787,
      "loss": 6.4132,
      "mean_token_accuracy": 0.15651057213544844,
      "num_tokens": 10757050.0,
      "step": 4240
    },
    {
      "entropy": 6.5376416683197025,
      "epoch": 0.4899019042123485,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004984090857822406,
      "loss": 6.4716,
      "mean_token_accuracy": 0.15311045944690704,
      "num_tokens": 10768667.0,
      "step": 4245
    },
    {
      "entropy": 6.574718618392945,
      "epoch": 0.49047893825735717,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004984041836381488,
      "loss": 6.4875,
      "mean_token_accuracy": 0.14979530423879622,
      "num_tokens": 10781687.0,
      "step": 4250
    },
    {
      "entropy": 6.4205322265625,
      "epoch": 0.49105597230236586,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004983992739799682,
      "loss": 6.3705,
      "mean_token_accuracy": 0.15820584297180176,
      "num_tokens": 10794966.0,
      "step": 4255
    },
    {
      "entropy": 6.563854122161866,
      "epoch": 0.4916330063473745,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004983943568078642,
      "loss": 6.485,
      "mean_token_accuracy": 0.14713219404220582,
      "num_tokens": 10807946.0,
      "step": 4260
    },
    {
      "entropy": 6.4821882247924805,
      "epoch": 0.49221004039238314,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004983894321220017,
      "loss": 6.4434,
      "mean_token_accuracy": 0.15347274988889695,
      "num_tokens": 10819258.0,
      "step": 4265
    },
    {
      "entropy": 6.48727593421936,
      "epoch": 0.4927870744373918,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004983844999225468,
      "loss": 6.4617,
      "mean_token_accuracy": 0.15150419622659683,
      "num_tokens": 10831439.0,
      "step": 4270
    },
    {
      "entropy": 6.4945567607879635,
      "epoch": 0.4933641084824005,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004983795602096651,
      "loss": 6.4142,
      "mean_token_accuracy": 0.15454161539673805,
      "num_tokens": 10844214.0,
      "step": 4275
    },
    {
      "entropy": 6.479346752166748,
      "epoch": 0.4939411425274091,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000498374612983523,
      "loss": 6.5263,
      "mean_token_accuracy": 0.14760390818119049,
      "num_tokens": 10856675.0,
      "step": 4280
    },
    {
      "entropy": 6.4818150997161865,
      "epoch": 0.49451817657241776,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004983696582442866,
      "loss": 6.4437,
      "mean_token_accuracy": 0.15344759076833725,
      "num_tokens": 10869296.0,
      "step": 4285
    },
    {
      "entropy": 6.481857109069824,
      "epoch": 0.49509521061742645,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004983646959921227,
      "loss": 6.5071,
      "mean_token_accuracy": 0.15597839653491974,
      "num_tokens": 10881484.0,
      "step": 4290
    },
    {
      "entropy": 6.552584600448609,
      "epoch": 0.4956722446624351,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004983597262271984,
      "loss": 6.475,
      "mean_token_accuracy": 0.1549718886613846,
      "num_tokens": 10894322.0,
      "step": 4295
    },
    {
      "entropy": 6.431555366516113,
      "epoch": 0.49624927870744373,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004983547489496804,
      "loss": 6.4135,
      "mean_token_accuracy": 0.1639399468898773,
      "num_tokens": 10907568.0,
      "step": 4300
    },
    {
      "entropy": 6.562343311309815,
      "epoch": 0.49682631275245237,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004983497641597365,
      "loss": 6.594,
      "mean_token_accuracy": 0.14406462758779526,
      "num_tokens": 10921371.0,
      "step": 4305
    },
    {
      "entropy": 6.653347396850586,
      "epoch": 0.49740334679746107,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004983447718575342,
      "loss": 6.5363,
      "mean_token_accuracy": 0.14576052129268646,
      "num_tokens": 10934725.0,
      "step": 4310
    },
    {
      "entropy": 6.522889471054077,
      "epoch": 0.4979803808424697,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004983397720432414,
      "loss": 6.4682,
      "mean_token_accuracy": 0.14770981222391127,
      "num_tokens": 10947985.0,
      "step": 4315
    },
    {
      "entropy": 6.352623462677002,
      "epoch": 0.49855741488747835,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004983347647170264,
      "loss": 6.3137,
      "mean_token_accuracy": 0.1705171674489975,
      "num_tokens": 10961112.0,
      "step": 4320
    },
    {
      "entropy": 6.449280834197998,
      "epoch": 0.49913444893248704,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004983297498790574,
      "loss": 6.4221,
      "mean_token_accuracy": 0.15522706508636475,
      "num_tokens": 10973414.0,
      "step": 4325
    },
    {
      "entropy": 6.548866939544678,
      "epoch": 0.4997114829774957,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004983247275295034,
      "loss": 6.3865,
      "mean_token_accuracy": 0.15996749848127365,
      "num_tokens": 10985609.0,
      "step": 4330
    },
    {
      "entropy": 6.459391069412232,
      "epoch": 0.5002885170225043,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004983196976685329,
      "loss": 6.4758,
      "mean_token_accuracy": 0.1579616606235504,
      "num_tokens": 10998928.0,
      "step": 4335
    },
    {
      "entropy": 6.43979172706604,
      "epoch": 0.500865551067513,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004983146602963155,
      "loss": 6.3929,
      "mean_token_accuracy": 0.15220501720905305,
      "num_tokens": 11012254.0,
      "step": 4340
    },
    {
      "entropy": 6.537631368637085,
      "epoch": 0.5014425851125216,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004983096154130203,
      "loss": 6.4513,
      "mean_token_accuracy": 0.14437098503112794,
      "num_tokens": 11024040.0,
      "step": 4345
    },
    {
      "entropy": 6.602092456817627,
      "epoch": 0.5020196191575302,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004983045630188171,
      "loss": 6.6151,
      "mean_token_accuracy": 0.14231864586472512,
      "num_tokens": 11037207.0,
      "step": 4350
    },
    {
      "entropy": 6.534118032455444,
      "epoch": 0.502596653202539,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004982995031138759,
      "loss": 6.4824,
      "mean_token_accuracy": 0.1532538965344429,
      "num_tokens": 11049990.0,
      "step": 4355
    },
    {
      "entropy": 6.572572803497314,
      "epoch": 0.5031736872475476,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004982944356983666,
      "loss": 6.5372,
      "mean_token_accuracy": 0.1520415410399437,
      "num_tokens": 11062365.0,
      "step": 4360
    },
    {
      "entropy": 6.488394451141358,
      "epoch": 0.5037507212925563,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004982893607724602,
      "loss": 6.4183,
      "mean_token_accuracy": 0.15387073308229446,
      "num_tokens": 11073721.0,
      "step": 4365
    },
    {
      "entropy": 6.456384468078613,
      "epoch": 0.5043277553375649,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000498284278336327,
      "loss": 6.4569,
      "mean_token_accuracy": 0.15044577419757843,
      "num_tokens": 11085834.0,
      "step": 4370
    },
    {
      "entropy": 6.54858431816101,
      "epoch": 0.5049047893825735,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004982791883901379,
      "loss": 6.5009,
      "mean_token_accuracy": 0.14734342098236083,
      "num_tokens": 11099432.0,
      "step": 4375
    },
    {
      "entropy": 6.414014101028442,
      "epoch": 0.5054818234275822,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004982740909340643,
      "loss": 6.2897,
      "mean_token_accuracy": 0.16746317744255065,
      "num_tokens": 11111169.0,
      "step": 4380
    },
    {
      "entropy": 6.447120141983032,
      "epoch": 0.5060588574725908,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004982689859682775,
      "loss": 6.4313,
      "mean_token_accuracy": 0.15230001211166383,
      "num_tokens": 11124131.0,
      "step": 4385
    },
    {
      "entropy": 6.560237264633178,
      "epoch": 0.5066358915175996,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004982638734929492,
      "loss": 6.4459,
      "mean_token_accuracy": 0.15483202338218688,
      "num_tokens": 11135810.0,
      "step": 4390
    },
    {
      "entropy": 6.4838165760040285,
      "epoch": 0.5072129255626082,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004982587535082515,
      "loss": 6.4969,
      "mean_token_accuracy": 0.1554883137345314,
      "num_tokens": 11148081.0,
      "step": 4395
    },
    {
      "entropy": 6.5296706199646,
      "epoch": 0.5077899596076169,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004982536260143565,
      "loss": 6.406,
      "mean_token_accuracy": 0.14896203130483626,
      "num_tokens": 11161427.0,
      "step": 4400
    },
    {
      "entropy": 6.422544956207275,
      "epoch": 0.5083669936526255,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004982484910114367,
      "loss": 6.3091,
      "mean_token_accuracy": 0.16211914420127868,
      "num_tokens": 11174183.0,
      "step": 4405
    },
    {
      "entropy": 6.440419244766235,
      "epoch": 0.5089440276976341,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004982433484996648,
      "loss": 6.4885,
      "mean_token_accuracy": 0.15194780081510545,
      "num_tokens": 11187151.0,
      "step": 4410
    },
    {
      "entropy": 6.542991018295288,
      "epoch": 0.5095210617426428,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004982381984792138,
      "loss": 6.3658,
      "mean_token_accuracy": 0.15823031216859818,
      "num_tokens": 11199603.0,
      "step": 4415
    },
    {
      "entropy": 6.415654516220092,
      "epoch": 0.5100980957876514,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004982330409502568,
      "loss": 6.4373,
      "mean_token_accuracy": 0.15554686188697814,
      "num_tokens": 11211709.0,
      "step": 4420
    },
    {
      "entropy": 6.46930947303772,
      "epoch": 0.5106751298326602,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004982278759129674,
      "loss": 6.4284,
      "mean_token_accuracy": 0.15755887180566788,
      "num_tokens": 11224192.0,
      "step": 4425
    },
    {
      "entropy": 6.489309501647949,
      "epoch": 0.5112521638776688,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004982227033675194,
      "loss": 6.4188,
      "mean_token_accuracy": 0.15911545157432555,
      "num_tokens": 11237026.0,
      "step": 4430
    },
    {
      "entropy": 6.381870603561401,
      "epoch": 0.5118291979226774,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004982175233140865,
      "loss": 6.4202,
      "mean_token_accuracy": 0.1576029047369957,
      "num_tokens": 11250646.0,
      "step": 4435
    },
    {
      "entropy": 6.651771831512451,
      "epoch": 0.5124062319676861,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004982123357528431,
      "loss": 6.5272,
      "mean_token_accuracy": 0.14984582662582396,
      "num_tokens": 11262991.0,
      "step": 4440
    },
    {
      "entropy": 6.521687078475952,
      "epoch": 0.5129832660126947,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004982071406839636,
      "loss": 6.436,
      "mean_token_accuracy": 0.14819750487804412,
      "num_tokens": 11275535.0,
      "step": 4445
    },
    {
      "entropy": 6.54852614402771,
      "epoch": 0.5135603000577034,
      "grad_norm": 1.0,
      "learning_rate": 0.0004982019381076229,
      "loss": 6.5718,
      "mean_token_accuracy": 0.14646391719579696,
      "num_tokens": 11288473.0,
      "step": 4450
    },
    {
      "entropy": 6.577858972549438,
      "epoch": 0.514137334102712,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004981967280239958,
      "loss": 6.5739,
      "mean_token_accuracy": 0.1495085082948208,
      "num_tokens": 11300755.0,
      "step": 4455
    },
    {
      "entropy": 6.524977397918701,
      "epoch": 0.5147143681477208,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004981915104332578,
      "loss": 6.4232,
      "mean_token_accuracy": 0.15421567112207413,
      "num_tokens": 11312912.0,
      "step": 4460
    },
    {
      "entropy": 6.407872343063355,
      "epoch": 0.5152914021927294,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000498186285335584,
      "loss": 6.4623,
      "mean_token_accuracy": 0.1514579698443413,
      "num_tokens": 11325714.0,
      "step": 4465
    },
    {
      "entropy": 6.602529144287109,
      "epoch": 0.515868436237738,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004981810527311505,
      "loss": 6.4021,
      "mean_token_accuracy": 0.16395699083805085,
      "num_tokens": 11337820.0,
      "step": 4470
    },
    {
      "entropy": 6.4563802719116214,
      "epoch": 0.5164454702827467,
      "grad_norm": 1.109375,
      "learning_rate": 0.000498175812620133,
      "loss": 6.452,
      "mean_token_accuracy": 0.1533729150891304,
      "num_tokens": 11350005.0,
      "step": 4475
    },
    {
      "entropy": 6.54690728187561,
      "epoch": 0.5170225043277553,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004981705650027081,
      "loss": 6.5551,
      "mean_token_accuracy": 0.15120244324207305,
      "num_tokens": 11362918.0,
      "step": 4480
    },
    {
      "entropy": 6.602835369110108,
      "epoch": 0.517599538372764,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000498165309879052,
      "loss": 6.4368,
      "mean_token_accuracy": 0.14729430228471757,
      "num_tokens": 11375186.0,
      "step": 4485
    },
    {
      "entropy": 6.541594409942627,
      "epoch": 0.5181765724177726,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004981600472493415,
      "loss": 6.4797,
      "mean_token_accuracy": 0.1510907381772995,
      "num_tokens": 11387607.0,
      "step": 4490
    },
    {
      "entropy": 6.500184345245361,
      "epoch": 0.5187536064627813,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004981547771137537,
      "loss": 6.504,
      "mean_token_accuracy": 0.15213208347558976,
      "num_tokens": 11400090.0,
      "step": 4495
    },
    {
      "entropy": 6.582069826126099,
      "epoch": 0.51933064050779,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004981494994724658,
      "loss": 6.5404,
      "mean_token_accuracy": 0.1519355371594429,
      "num_tokens": 11413220.0,
      "step": 4500
    },
    {
      "entropy": 6.5244566917419435,
      "epoch": 0.5199076745527986,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004981442143256554,
      "loss": 6.433,
      "mean_token_accuracy": 0.154581780731678,
      "num_tokens": 11425043.0,
      "step": 4505
    },
    {
      "entropy": 6.458435344696045,
      "epoch": 0.5204847085978073,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004981389216735001,
      "loss": 6.4291,
      "mean_token_accuracy": 0.15687761902809144,
      "num_tokens": 11438097.0,
      "step": 4510
    },
    {
      "entropy": 6.51494402885437,
      "epoch": 0.5210617426428159,
      "grad_norm": 0.96875,
      "learning_rate": 0.000498133621516178,
      "loss": 6.4219,
      "mean_token_accuracy": 0.1568854421377182,
      "num_tokens": 11450903.0,
      "step": 4515
    },
    {
      "entropy": 6.497988605499268,
      "epoch": 0.5216387766878245,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004981283138538675,
      "loss": 6.5001,
      "mean_token_accuracy": 0.15478087440133095,
      "num_tokens": 11464566.0,
      "step": 4520
    },
    {
      "entropy": 6.530912065505982,
      "epoch": 0.5222158107328332,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000498122998686747,
      "loss": 6.4482,
      "mean_token_accuracy": 0.15248721837997437,
      "num_tokens": 11476664.0,
      "step": 4525
    },
    {
      "entropy": 6.53693323135376,
      "epoch": 0.5227928447778419,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004981176760149951,
      "loss": 6.4836,
      "mean_token_accuracy": 0.14837454855442048,
      "num_tokens": 11489052.0,
      "step": 4530
    },
    {
      "entropy": 6.507598543167115,
      "epoch": 0.5233698788228506,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004981123458387911,
      "loss": 6.4297,
      "mean_token_accuracy": 0.15648974925279618,
      "num_tokens": 11501774.0,
      "step": 4535
    },
    {
      "entropy": 6.530131435394287,
      "epoch": 0.5239469128678592,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004981070081583142,
      "loss": 6.5613,
      "mean_token_accuracy": 0.15191105455160142,
      "num_tokens": 11514397.0,
      "step": 4540
    },
    {
      "entropy": 6.54612283706665,
      "epoch": 0.5245239469128679,
      "grad_norm": 1.046875,
      "learning_rate": 0.000498101662973744,
      "loss": 6.4437,
      "mean_token_accuracy": 0.14765722304582596,
      "num_tokens": 11527413.0,
      "step": 4545
    },
    {
      "entropy": 6.548727941513062,
      "epoch": 0.5251009809578765,
      "grad_norm": 1.0,
      "learning_rate": 0.00049809631028526,
      "loss": 6.4986,
      "mean_token_accuracy": 0.14623937308788298,
      "num_tokens": 11540423.0,
      "step": 4550
    },
    {
      "entropy": 6.514033603668213,
      "epoch": 0.5256780150028851,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004980909500930424,
      "loss": 6.3898,
      "mean_token_accuracy": 0.1627660170197487,
      "num_tokens": 11552825.0,
      "step": 4555
    },
    {
      "entropy": 6.429790449142456,
      "epoch": 0.5262550490478938,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004980855823972717,
      "loss": 6.3647,
      "mean_token_accuracy": 0.1591410979628563,
      "num_tokens": 11565556.0,
      "step": 4560
    },
    {
      "entropy": 6.487172555923462,
      "epoch": 0.5268320830929025,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000498080207198128,
      "loss": 6.4592,
      "mean_token_accuracy": 0.15544339492917061,
      "num_tokens": 11580056.0,
      "step": 4565
    },
    {
      "entropy": 6.513148021697998,
      "epoch": 0.5274091171379112,
      "grad_norm": 0.875,
      "learning_rate": 0.0004980748244957925,
      "loss": 6.4885,
      "mean_token_accuracy": 0.1587923526763916,
      "num_tokens": 11592104.0,
      "step": 4570
    },
    {
      "entropy": 6.455549335479736,
      "epoch": 0.5279861511829198,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004980694342904461,
      "loss": 6.3804,
      "mean_token_accuracy": 0.1524585172533989,
      "num_tokens": 11605349.0,
      "step": 4575
    },
    {
      "entropy": 6.416559314727783,
      "epoch": 0.5285631852279284,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004980640365822701,
      "loss": 6.3554,
      "mean_token_accuracy": 0.16644309014081954,
      "num_tokens": 11617787.0,
      "step": 4580
    },
    {
      "entropy": 6.38735728263855,
      "epoch": 0.5291402192729371,
      "grad_norm": 1.4375,
      "learning_rate": 0.000498058631371446,
      "loss": 6.3752,
      "mean_token_accuracy": 0.16232477128505707,
      "num_tokens": 11630279.0,
      "step": 4585
    },
    {
      "entropy": 6.551797866821289,
      "epoch": 0.5297172533179457,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004980532186581555,
      "loss": 6.4775,
      "mean_token_accuracy": 0.1578046441078186,
      "num_tokens": 11643378.0,
      "step": 4590
    },
    {
      "entropy": 6.405869960784912,
      "epoch": 0.5302942873629544,
      "grad_norm": 1.109375,
      "learning_rate": 0.000498047798442581,
      "loss": 6.4341,
      "mean_token_accuracy": 0.16507940590381623,
      "num_tokens": 11656472.0,
      "step": 4595
    },
    {
      "entropy": 6.498985528945923,
      "epoch": 0.5308713214079631,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004980423707249044,
      "loss": 6.4035,
      "mean_token_accuracy": 0.15753294676542282,
      "num_tokens": 11669265.0,
      "step": 4600
    },
    {
      "entropy": 6.4637227058410645,
      "epoch": 0.5314483554529718,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004980369355053086,
      "loss": 6.4368,
      "mean_token_accuracy": 0.15306321531534195,
      "num_tokens": 11681620.0,
      "step": 4605
    },
    {
      "entropy": 6.435328674316406,
      "epoch": 0.5320253894979804,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004980314927839763,
      "loss": 6.3407,
      "mean_token_accuracy": 0.15545193627476692,
      "num_tokens": 11695720.0,
      "step": 4610
    },
    {
      "entropy": 6.442215824127198,
      "epoch": 0.532602423542989,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004980260425610903,
      "loss": 6.4048,
      "mean_token_accuracy": 0.15217029601335524,
      "num_tokens": 11709563.0,
      "step": 4615
    },
    {
      "entropy": 6.498468637466431,
      "epoch": 0.5331794575879977,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004980205848368343,
      "loss": 6.4619,
      "mean_token_accuracy": 0.15577619522809982,
      "num_tokens": 11722401.0,
      "step": 4620
    },
    {
      "entropy": 6.5028361797332765,
      "epoch": 0.5337564916330063,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004980151196113917,
      "loss": 6.4759,
      "mean_token_accuracy": 0.15297331362962724,
      "num_tokens": 11735183.0,
      "step": 4625
    },
    {
      "entropy": 6.533588933944702,
      "epoch": 0.534333525678015,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004980096468849464,
      "loss": 6.453,
      "mean_token_accuracy": 0.14690574258565903,
      "num_tokens": 11748282.0,
      "step": 4630
    },
    {
      "entropy": 6.46541657447815,
      "epoch": 0.5349105597230237,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004980041666576823,
      "loss": 6.4177,
      "mean_token_accuracy": 0.15209294706583024,
      "num_tokens": 11760208.0,
      "step": 4635
    },
    {
      "entropy": 6.4671567440032955,
      "epoch": 0.5354875937680323,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004979986789297837,
      "loss": 6.4782,
      "mean_token_accuracy": 0.14976225644350052,
      "num_tokens": 11772509.0,
      "step": 4640
    },
    {
      "entropy": 6.592843008041382,
      "epoch": 0.536064627813041,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004979931837014355,
      "loss": 6.4139,
      "mean_token_accuracy": 0.15582239180803298,
      "num_tokens": 11786101.0,
      "step": 4645
    },
    {
      "entropy": 6.481404590606689,
      "epoch": 0.5366416618580496,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004979876809728223,
      "loss": 6.4717,
      "mean_token_accuracy": 0.15234363228082656,
      "num_tokens": 11798325.0,
      "step": 4650
    },
    {
      "entropy": 6.487795019149781,
      "epoch": 0.5372186959030583,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004979821707441292,
      "loss": 6.4637,
      "mean_token_accuracy": 0.14984379410743714,
      "num_tokens": 11812905.0,
      "step": 4655
    },
    {
      "entropy": 6.444129610061646,
      "epoch": 0.5377957299480669,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004979766530155416,
      "loss": 6.3696,
      "mean_token_accuracy": 0.156754370033741,
      "num_tokens": 11826495.0,
      "step": 4660
    },
    {
      "entropy": 6.541183757781982,
      "epoch": 0.5383727639930755,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004979711277872449,
      "loss": 6.4862,
      "mean_token_accuracy": 0.14373173713684081,
      "num_tokens": 11840472.0,
      "step": 4665
    },
    {
      "entropy": 6.566144275665283,
      "epoch": 0.5389497980380843,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004979655950594252,
      "loss": 6.4322,
      "mean_token_accuracy": 0.15399955511093139,
      "num_tokens": 11852800.0,
      "step": 4670
    },
    {
      "entropy": 6.471300411224365,
      "epoch": 0.5395268320830929,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004979600548322684,
      "loss": 6.4561,
      "mean_token_accuracy": 0.15268122404813766,
      "num_tokens": 11866111.0,
      "step": 4675
    },
    {
      "entropy": 6.414181089401245,
      "epoch": 0.5401038661281016,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000497954507105961,
      "loss": 6.4397,
      "mean_token_accuracy": 0.14690505862236022,
      "num_tokens": 11878103.0,
      "step": 4680
    },
    {
      "entropy": 6.5201537132263185,
      "epoch": 0.5406809001731102,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004979489518806893,
      "loss": 6.4357,
      "mean_token_accuracy": 0.15971802920103073,
      "num_tokens": 11890851.0,
      "step": 4685
    },
    {
      "entropy": 6.504558753967285,
      "epoch": 0.5412579342181189,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004979433891566405,
      "loss": 6.4572,
      "mean_token_accuracy": 0.14689093828201294,
      "num_tokens": 11903351.0,
      "step": 4690
    },
    {
      "entropy": 6.565521907806397,
      "epoch": 0.5418349682631275,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004979378189340015,
      "loss": 6.5165,
      "mean_token_accuracy": 0.1526387646794319,
      "num_tokens": 11915239.0,
      "step": 4695
    },
    {
      "entropy": 6.508232831954956,
      "epoch": 0.5424120023081361,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004979322412129597,
      "loss": 6.3842,
      "mean_token_accuracy": 0.15869044959545137,
      "num_tokens": 11928678.0,
      "step": 4700
    },
    {
      "entropy": 6.508393621444702,
      "epoch": 0.5429890363531449,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004979266559937028,
      "loss": 6.4184,
      "mean_token_accuracy": 0.15322802662849427,
      "num_tokens": 11940547.0,
      "step": 4705
    },
    {
      "entropy": 6.473295593261719,
      "epoch": 0.5435660703981535,
      "grad_norm": 1.125,
      "learning_rate": 0.0004979210632764185,
      "loss": 6.4582,
      "mean_token_accuracy": 0.1515795946121216,
      "num_tokens": 11953069.0,
      "step": 4710
    },
    {
      "entropy": 6.433367586135864,
      "epoch": 0.5441431044431622,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004979154630612949,
      "loss": 6.4027,
      "mean_token_accuracy": 0.15667079091072084,
      "num_tokens": 11964301.0,
      "step": 4715
    },
    {
      "entropy": 6.4675774574279785,
      "epoch": 0.5447201384881708,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004979098553485205,
      "loss": 6.4311,
      "mean_token_accuracy": 0.15274412259459497,
      "num_tokens": 11976311.0,
      "step": 4720
    },
    {
      "entropy": 6.463955593109131,
      "epoch": 0.5452971725331794,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004979042401382838,
      "loss": 6.3981,
      "mean_token_accuracy": 0.1605428010225296,
      "num_tokens": 11989749.0,
      "step": 4725
    },
    {
      "entropy": 6.4383635997772215,
      "epoch": 0.5458742065781881,
      "grad_norm": 1.0,
      "learning_rate": 0.0004978986174307737,
      "loss": 6.3636,
      "mean_token_accuracy": 0.16022162437438964,
      "num_tokens": 12002384.0,
      "step": 4730
    },
    {
      "entropy": 6.575389289855957,
      "epoch": 0.5464512406231967,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004978929872261794,
      "loss": 6.4486,
      "mean_token_accuracy": 0.16075280755758287,
      "num_tokens": 12014573.0,
      "step": 4735
    },
    {
      "entropy": 6.298378801345825,
      "epoch": 0.5470282746682055,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004978873495246901,
      "loss": 6.2861,
      "mean_token_accuracy": 0.1614734098315239,
      "num_tokens": 12027340.0,
      "step": 4740
    },
    {
      "entropy": 6.513762092590332,
      "epoch": 0.5476053087132141,
      "grad_norm": 1.125,
      "learning_rate": 0.0004978817043264955,
      "loss": 6.4559,
      "mean_token_accuracy": 0.15529761761426925,
      "num_tokens": 12039498.0,
      "step": 4745
    },
    {
      "entropy": 6.5160966396331785,
      "epoch": 0.5481823427582228,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004978760516317856,
      "loss": 6.4704,
      "mean_token_accuracy": 0.15069929808378218,
      "num_tokens": 12051426.0,
      "step": 4750
    },
    {
      "entropy": 6.503102254867554,
      "epoch": 0.5487593768032314,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004978703914407503,
      "loss": 6.4365,
      "mean_token_accuracy": 0.15093417465686798,
      "num_tokens": 12063975.0,
      "step": 4755
    },
    {
      "entropy": 6.484177446365356,
      "epoch": 0.54933641084824,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004978647237535802,
      "loss": 6.3967,
      "mean_token_accuracy": 0.15591831505298615,
      "num_tokens": 12075159.0,
      "step": 4760
    },
    {
      "entropy": 6.481649875640869,
      "epoch": 0.5499134448932487,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004978590485704657,
      "loss": 6.3927,
      "mean_token_accuracy": 0.1560533568263054,
      "num_tokens": 12087052.0,
      "step": 4765
    },
    {
      "entropy": 6.428612995147705,
      "epoch": 0.5504904789382573,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004978533658915979,
      "loss": 6.4738,
      "mean_token_accuracy": 0.14724875316023828,
      "num_tokens": 12100021.0,
      "step": 4770
    },
    {
      "entropy": 6.49065089225769,
      "epoch": 0.5510675129832661,
      "grad_norm": 1.0,
      "learning_rate": 0.0004978476757171678,
      "loss": 6.4561,
      "mean_token_accuracy": 0.1556130602955818,
      "num_tokens": 12113067.0,
      "step": 4775
    },
    {
      "entropy": 6.525161457061768,
      "epoch": 0.5516445470282747,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004978419780473668,
      "loss": 6.4383,
      "mean_token_accuracy": 0.15904852747917175,
      "num_tokens": 12126108.0,
      "step": 4780
    },
    {
      "entropy": 6.392880296707153,
      "epoch": 0.5522215810732833,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004978362728823865,
      "loss": 6.3151,
      "mean_token_accuracy": 0.16119097918272018,
      "num_tokens": 12138821.0,
      "step": 4785
    },
    {
      "entropy": 6.473163700103759,
      "epoch": 0.552798615118292,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004978305602224189,
      "loss": 6.5342,
      "mean_token_accuracy": 0.14587229639291763,
      "num_tokens": 12151621.0,
      "step": 4790
    },
    {
      "entropy": 6.575317764282227,
      "epoch": 0.5533756491633006,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004978248400676562,
      "loss": 6.4485,
      "mean_token_accuracy": 0.15152132362127305,
      "num_tokens": 12163232.0,
      "step": 4795
    },
    {
      "entropy": 6.459644412994384,
      "epoch": 0.5539526832083093,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004978191124182905,
      "loss": 6.4386,
      "mean_token_accuracy": 0.15065861567854882,
      "num_tokens": 12176857.0,
      "step": 4800
    },
    {
      "entropy": 6.3848557472229,
      "epoch": 0.5545297172533179,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004978133772745149,
      "loss": 6.3397,
      "mean_token_accuracy": 0.16201334744691848,
      "num_tokens": 12190311.0,
      "step": 4805
    },
    {
      "entropy": 6.487551927566528,
      "epoch": 0.5551067512983267,
      "grad_norm": 1.125,
      "learning_rate": 0.0004978076346365218,
      "loss": 6.4279,
      "mean_token_accuracy": 0.1516387790441513,
      "num_tokens": 12202490.0,
      "step": 4810
    },
    {
      "entropy": 6.547900247573852,
      "epoch": 0.5556837853433353,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004978018845045047,
      "loss": 6.4522,
      "mean_token_accuracy": 0.15225011259317398,
      "num_tokens": 12214759.0,
      "step": 4815
    },
    {
      "entropy": 6.461217832565308,
      "epoch": 0.5562608193883439,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004977961268786568,
      "loss": 6.3829,
      "mean_token_accuracy": 0.16132269203662872,
      "num_tokens": 12227639.0,
      "step": 4820
    },
    {
      "entropy": 6.437919187545776,
      "epoch": 0.5568378534333526,
      "grad_norm": 1.0,
      "learning_rate": 0.0004977903617591719,
      "loss": 6.4836,
      "mean_token_accuracy": 0.15411624759435655,
      "num_tokens": 12241049.0,
      "step": 4825
    },
    {
      "entropy": 6.485408544540405,
      "epoch": 0.5574148874783612,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004977845891462439,
      "loss": 6.3624,
      "mean_token_accuracy": 0.15602847784757615,
      "num_tokens": 12253980.0,
      "step": 4830
    },
    {
      "entropy": 6.468500137329102,
      "epoch": 0.5579919215233698,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004977788090400668,
      "loss": 6.4423,
      "mean_token_accuracy": 0.1594425231218338,
      "num_tokens": 12265103.0,
      "step": 4835
    },
    {
      "entropy": 6.404529762268067,
      "epoch": 0.5585689555683785,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004977730214408352,
      "loss": 6.4313,
      "mean_token_accuracy": 0.1530008316040039,
      "num_tokens": 12278574.0,
      "step": 4840
    },
    {
      "entropy": 6.413032960891724,
      "epoch": 0.5591459896133872,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004977672263487435,
      "loss": 6.4108,
      "mean_token_accuracy": 0.15949763730168343,
      "num_tokens": 12291038.0,
      "step": 4845
    },
    {
      "entropy": 6.535495901107788,
      "epoch": 0.5597230236583959,
      "grad_norm": 0.984375,
      "learning_rate": 0.000497761423763987,
      "loss": 6.4648,
      "mean_token_accuracy": 0.1519089475274086,
      "num_tokens": 12304339.0,
      "step": 4850
    },
    {
      "entropy": 6.517710733413696,
      "epoch": 0.5603000577034045,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004977556136867606,
      "loss": 6.4815,
      "mean_token_accuracy": 0.1531308799982071,
      "num_tokens": 12317927.0,
      "step": 4855
    },
    {
      "entropy": 6.488817358016968,
      "epoch": 0.5608770917484132,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004977497961172598,
      "loss": 6.4125,
      "mean_token_accuracy": 0.15044625252485275,
      "num_tokens": 12332904.0,
      "step": 4860
    },
    {
      "entropy": 6.557137966156006,
      "epoch": 0.5614541257934218,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004977439710556802,
      "loss": 6.3425,
      "mean_token_accuracy": 0.15948520302772523,
      "num_tokens": 12345449.0,
      "step": 4865
    },
    {
      "entropy": 6.374728155136109,
      "epoch": 0.5620311598384304,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004977381385022179,
      "loss": 6.4144,
      "mean_token_accuracy": 0.1593753643333912,
      "num_tokens": 12357334.0,
      "step": 4870
    },
    {
      "entropy": 6.495050764083862,
      "epoch": 0.5626081938834391,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004977322984570688,
      "loss": 6.4514,
      "mean_token_accuracy": 0.15414685010910034,
      "num_tokens": 12370094.0,
      "step": 4875
    },
    {
      "entropy": 6.512135982513428,
      "epoch": 0.5631852279284478,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004977264509204296,
      "loss": 6.4571,
      "mean_token_accuracy": 0.1477771133184433,
      "num_tokens": 12382875.0,
      "step": 4880
    },
    {
      "entropy": 6.446519422531128,
      "epoch": 0.5637622619734565,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004977205958924967,
      "loss": 6.417,
      "mean_token_accuracy": 0.1503426268696785,
      "num_tokens": 12395534.0,
      "step": 4885
    },
    {
      "entropy": 6.5179808139801025,
      "epoch": 0.5643392960184651,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004977147333734673,
      "loss": 6.4349,
      "mean_token_accuracy": 0.15383182018995284,
      "num_tokens": 12407300.0,
      "step": 4890
    },
    {
      "entropy": 6.438922119140625,
      "epoch": 0.5649163300634737,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004977088633635385,
      "loss": 6.3403,
      "mean_token_accuracy": 0.15970048308372498,
      "num_tokens": 12419499.0,
      "step": 4895
    },
    {
      "entropy": 6.382441711425781,
      "epoch": 0.5654933641084824,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004977029858629076,
      "loss": 6.2988,
      "mean_token_accuracy": 0.1677648812532425,
      "num_tokens": 12431279.0,
      "step": 4900
    },
    {
      "entropy": 6.338820362091065,
      "epoch": 0.566070398153491,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004976971008717726,
      "loss": 6.239,
      "mean_token_accuracy": 0.16438972651958467,
      "num_tokens": 12443976.0,
      "step": 4905
    },
    {
      "entropy": 6.444260597229004,
      "epoch": 0.5666474321984997,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004976912083903309,
      "loss": 6.4089,
      "mean_token_accuracy": 0.15466192364692688,
      "num_tokens": 12457752.0,
      "step": 4910
    },
    {
      "entropy": 6.427901744842529,
      "epoch": 0.5672244662435084,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004976853084187814,
      "loss": 6.3956,
      "mean_token_accuracy": 0.15571021735668183,
      "num_tokens": 12469776.0,
      "step": 4915
    },
    {
      "entropy": 6.5230522632598875,
      "epoch": 0.5678015002885171,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004976794009573219,
      "loss": 6.4463,
      "mean_token_accuracy": 0.1475426286458969,
      "num_tokens": 12483300.0,
      "step": 4920
    },
    {
      "entropy": 6.424286699295044,
      "epoch": 0.5683785343335257,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004976734860061515,
      "loss": 6.3869,
      "mean_token_accuracy": 0.15617836862802506,
      "num_tokens": 12494564.0,
      "step": 4925
    },
    {
      "entropy": 6.464950609207153,
      "epoch": 0.5689555683785343,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004976675635654688,
      "loss": 6.4098,
      "mean_token_accuracy": 0.15661260336637498,
      "num_tokens": 12507761.0,
      "step": 4930
    },
    {
      "entropy": 6.511030769348144,
      "epoch": 0.569532602423543,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004976616336354733,
      "loss": 6.3961,
      "mean_token_accuracy": 0.1580558568239212,
      "num_tokens": 12520045.0,
      "step": 4935
    },
    {
      "entropy": 6.418640422821045,
      "epoch": 0.5701096364685516,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004976556962163645,
      "loss": 6.3443,
      "mean_token_accuracy": 0.16202140748500823,
      "num_tokens": 12532471.0,
      "step": 4940
    },
    {
      "entropy": 6.476809978485107,
      "epoch": 0.5706866705135603,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004976497513083419,
      "loss": 6.4439,
      "mean_token_accuracy": 0.15516655892133713,
      "num_tokens": 12544610.0,
      "step": 4945
    },
    {
      "entropy": 6.455000352859497,
      "epoch": 0.571263704558569,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004976437989116055,
      "loss": 6.4565,
      "mean_token_accuracy": 0.15856476873159409,
      "num_tokens": 12558682.0,
      "step": 4950
    },
    {
      "entropy": 6.455588340759277,
      "epoch": 0.5718407386035776,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004976378390263554,
      "loss": 6.36,
      "mean_token_accuracy": 0.15180347859859467,
      "num_tokens": 12570711.0,
      "step": 4955
    },
    {
      "entropy": 6.44568657875061,
      "epoch": 0.5724177726485863,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004976318716527924,
      "loss": 6.3709,
      "mean_token_accuracy": 0.1573265388607979,
      "num_tokens": 12582468.0,
      "step": 4960
    },
    {
      "entropy": 6.445875930786133,
      "epoch": 0.5729948066935949,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004976258967911168,
      "loss": 6.4162,
      "mean_token_accuracy": 0.15737249553203583,
      "num_tokens": 12594991.0,
      "step": 4965
    },
    {
      "entropy": 6.512746000289917,
      "epoch": 0.5735718407386036,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004976199144415298,
      "loss": 6.3888,
      "mean_token_accuracy": 0.15889376401901245,
      "num_tokens": 12606928.0,
      "step": 4970
    },
    {
      "entropy": 6.512881851196289,
      "epoch": 0.5741488747836122,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004976139246042325,
      "loss": 6.4763,
      "mean_token_accuracy": 0.15111626386642457,
      "num_tokens": 12620539.0,
      "step": 4975
    },
    {
      "entropy": 6.499468421936035,
      "epoch": 0.5747259088286208,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004976079272794265,
      "loss": 6.4125,
      "mean_token_accuracy": 0.15034622550010682,
      "num_tokens": 12632629.0,
      "step": 4980
    },
    {
      "entropy": 6.418478775024414,
      "epoch": 0.5753029428736296,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004976019224673134,
      "loss": 6.3918,
      "mean_token_accuracy": 0.1634598359465599,
      "num_tokens": 12646631.0,
      "step": 4985
    },
    {
      "entropy": 6.428795003890992,
      "epoch": 0.5758799769186382,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004975959101680953,
      "loss": 6.3772,
      "mean_token_accuracy": 0.15648327320814132,
      "num_tokens": 12659154.0,
      "step": 4990
    },
    {
      "entropy": 6.473361968994141,
      "epoch": 0.5764570109636469,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004975898903819742,
      "loss": 6.4635,
      "mean_token_accuracy": 0.1549312099814415,
      "num_tokens": 12671796.0,
      "step": 4995
    },
    {
      "entropy": 6.566014671325684,
      "epoch": 0.5770340450086555,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004975838631091527,
      "loss": 6.4306,
      "mean_token_accuracy": 0.15129499211907388,
      "num_tokens": 12684432.0,
      "step": 5000
    },
    {
      "entropy": 6.457323169708252,
      "epoch": 0.5776110790536642,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004975778283498336,
      "loss": 6.4398,
      "mean_token_accuracy": 0.15979505255818366,
      "num_tokens": 12697585.0,
      "step": 5005
    },
    {
      "entropy": 6.396718311309814,
      "epoch": 0.5781881130986728,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004975717861042198,
      "loss": 6.343,
      "mean_token_accuracy": 0.1617707148194313,
      "num_tokens": 12710654.0,
      "step": 5010
    },
    {
      "entropy": 6.428943681716919,
      "epoch": 0.5787651471436814,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004975657363725146,
      "loss": 6.4053,
      "mean_token_accuracy": 0.15304268896579742,
      "num_tokens": 12723141.0,
      "step": 5015
    },
    {
      "entropy": 6.524564743041992,
      "epoch": 0.5793421811886902,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004975596791549213,
      "loss": 6.4832,
      "mean_token_accuracy": 0.15016788989305496,
      "num_tokens": 12735921.0,
      "step": 5020
    },
    {
      "entropy": 6.449824237823487,
      "epoch": 0.5799192152336988,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004975536144516437,
      "loss": 6.4643,
      "mean_token_accuracy": 0.15014379173517228,
      "num_tokens": 12747704.0,
      "step": 5025
    },
    {
      "entropy": 6.559833002090454,
      "epoch": 0.5804962492787075,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000497547542262886,
      "loss": 6.4414,
      "mean_token_accuracy": 0.14836343228816987,
      "num_tokens": 12759368.0,
      "step": 5030
    },
    {
      "entropy": 6.449873924255371,
      "epoch": 0.5810732833237161,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004975414625888521,
      "loss": 6.4041,
      "mean_token_accuracy": 0.15791085958480836,
      "num_tokens": 12772226.0,
      "step": 5035
    },
    {
      "entropy": 6.496578931808472,
      "epoch": 0.5816503173687247,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004975353754297468,
      "loss": 6.4607,
      "mean_token_accuracy": 0.1514489993453026,
      "num_tokens": 12784817.0,
      "step": 5040
    },
    {
      "entropy": 6.494936227798462,
      "epoch": 0.5822273514137334,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004975292807857745,
      "loss": 6.3993,
      "mean_token_accuracy": 0.15533664226531982,
      "num_tokens": 12796996.0,
      "step": 5045
    },
    {
      "entropy": 6.469889163970947,
      "epoch": 0.582804385458742,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004975231786571406,
      "loss": 6.4693,
      "mean_token_accuracy": 0.15263034403324127,
      "num_tokens": 12809581.0,
      "step": 5050
    },
    {
      "entropy": 6.475830459594727,
      "epoch": 0.5833814195037508,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004975170690440499,
      "loss": 6.4249,
      "mean_token_accuracy": 0.154001185297966,
      "num_tokens": 12822060.0,
      "step": 5055
    },
    {
      "entropy": 6.510536241531372,
      "epoch": 0.5839584535487594,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004975109519467083,
      "loss": 6.4041,
      "mean_token_accuracy": 0.15710070431232454,
      "num_tokens": 12835554.0,
      "step": 5060
    },
    {
      "entropy": 6.482144832611084,
      "epoch": 0.584535487593768,
      "grad_norm": 1.0,
      "learning_rate": 0.0004975048273653212,
      "loss": 6.3234,
      "mean_token_accuracy": 0.16243607848882674,
      "num_tokens": 12848187.0,
      "step": 5065
    },
    {
      "entropy": 6.329149436950684,
      "epoch": 0.5851125216387767,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004974986953000948,
      "loss": 6.3599,
      "mean_token_accuracy": 0.16322699934244156,
      "num_tokens": 12861498.0,
      "step": 5070
    },
    {
      "entropy": 6.614460182189942,
      "epoch": 0.5856895556837853,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004974925557512354,
      "loss": 6.5254,
      "mean_token_accuracy": 0.1417018637061119,
      "num_tokens": 12874906.0,
      "step": 5075
    },
    {
      "entropy": 6.4083353042602536,
      "epoch": 0.586266589728794,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004974864087189493,
      "loss": 6.3564,
      "mean_token_accuracy": 0.15833602845668793,
      "num_tokens": 12887169.0,
      "step": 5080
    },
    {
      "entropy": 6.390086889266968,
      "epoch": 0.5868436237738026,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004974802542034434,
      "loss": 6.4169,
      "mean_token_accuracy": 0.15532376170158385,
      "num_tokens": 12899639.0,
      "step": 5085
    },
    {
      "entropy": 6.496376276016235,
      "epoch": 0.5874206578188114,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004974740922049246,
      "loss": 6.3277,
      "mean_token_accuracy": 0.1536332845687866,
      "num_tokens": 12911913.0,
      "step": 5090
    },
    {
      "entropy": 6.496623134613037,
      "epoch": 0.58799769186382,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004974679227236004,
      "loss": 6.4769,
      "mean_token_accuracy": 0.1492656260728836,
      "num_tokens": 12924200.0,
      "step": 5095
    },
    {
      "entropy": 6.421284198760986,
      "epoch": 0.5885747259088286,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004974617457596779,
      "loss": 6.459,
      "mean_token_accuracy": 0.15051645785570145,
      "num_tokens": 12936726.0,
      "step": 5100
    },
    {
      "entropy": 6.4862532138824465,
      "epoch": 0.5891517599538373,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004974555613133652,
      "loss": 6.352,
      "mean_token_accuracy": 0.15397184193134308,
      "num_tokens": 12949751.0,
      "step": 5105
    },
    {
      "entropy": 6.441688489913941,
      "epoch": 0.5897287939988459,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004974493693848702,
      "loss": 6.4455,
      "mean_token_accuracy": 0.15539218187332154,
      "num_tokens": 12962573.0,
      "step": 5110
    },
    {
      "entropy": 6.553739356994629,
      "epoch": 0.5903058280438546,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004974431699744011,
      "loss": 6.4854,
      "mean_token_accuracy": 0.1539460763335228,
      "num_tokens": 12975635.0,
      "step": 5115
    },
    {
      "entropy": 6.455906963348388,
      "epoch": 0.5908828620888632,
      "grad_norm": 1.0,
      "learning_rate": 0.0004974369630821665,
      "loss": 6.4771,
      "mean_token_accuracy": 0.14752131924033166,
      "num_tokens": 12987822.0,
      "step": 5120
    },
    {
      "entropy": 6.479493856430054,
      "epoch": 0.5914598961338718,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004974307487083752,
      "loss": 6.4246,
      "mean_token_accuracy": 0.15978273302316665,
      "num_tokens": 13000082.0,
      "step": 5125
    },
    {
      "entropy": 6.48796534538269,
      "epoch": 0.5920369301788806,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004974245268532361,
      "loss": 6.3154,
      "mean_token_accuracy": 0.16128845363855362,
      "num_tokens": 13011902.0,
      "step": 5130
    },
    {
      "entropy": 6.432144737243652,
      "epoch": 0.5926139642238892,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004974182975169585,
      "loss": 6.3639,
      "mean_token_accuracy": 0.15628034323453904,
      "num_tokens": 13024507.0,
      "step": 5135
    },
    {
      "entropy": 6.419040822982788,
      "epoch": 0.5931909982688979,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000497412060699752,
      "loss": 6.4764,
      "mean_token_accuracy": 0.159742970764637,
      "num_tokens": 13037955.0,
      "step": 5140
    },
    {
      "entropy": 6.520135450363159,
      "epoch": 0.5937680323139065,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004974058164018263,
      "loss": 6.4429,
      "mean_token_accuracy": 0.15843138396739959,
      "num_tokens": 13050631.0,
      "step": 5145
    },
    {
      "entropy": 6.419425678253174,
      "epoch": 0.5943450663589152,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004973995646233914,
      "loss": 6.3847,
      "mean_token_accuracy": 0.15587171018123627,
      "num_tokens": 13062566.0,
      "step": 5150
    },
    {
      "entropy": 6.380414962768555,
      "epoch": 0.5949221004039238,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004973933053646576,
      "loss": 6.3415,
      "mean_token_accuracy": 0.15572840720415115,
      "num_tokens": 13075454.0,
      "step": 5155
    },
    {
      "entropy": 6.5283668518066404,
      "epoch": 0.5954991344489324,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004973870386258355,
      "loss": 6.4055,
      "mean_token_accuracy": 0.153472663462162,
      "num_tokens": 13087780.0,
      "step": 5160
    },
    {
      "entropy": 6.456124019622803,
      "epoch": 0.5960761684939412,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004973807644071357,
      "loss": 6.426,
      "mean_token_accuracy": 0.15076865404844284,
      "num_tokens": 13100957.0,
      "step": 5165
    },
    {
      "entropy": 6.4691649913787845,
      "epoch": 0.5966532025389498,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004973744827087695,
      "loss": 6.3416,
      "mean_token_accuracy": 0.16071075648069383,
      "num_tokens": 13113232.0,
      "step": 5170
    },
    {
      "entropy": 6.420907115936279,
      "epoch": 0.5972302365839585,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000497368193530948,
      "loss": 6.3373,
      "mean_token_accuracy": 0.16452635377645491,
      "num_tokens": 13125950.0,
      "step": 5175
    },
    {
      "entropy": 6.361990451812744,
      "epoch": 0.5978072706289671,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004973618968738828,
      "loss": 6.3246,
      "mean_token_accuracy": 0.15938978493213654,
      "num_tokens": 13138936.0,
      "step": 5180
    },
    {
      "entropy": 6.509745073318482,
      "epoch": 0.5983843046739757,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004973555927377856,
      "loss": 6.4599,
      "mean_token_accuracy": 0.15344280153512954,
      "num_tokens": 13152787.0,
      "step": 5185
    },
    {
      "entropy": 6.455010890960693,
      "epoch": 0.5989613387189844,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004973492811228685,
      "loss": 6.3684,
      "mean_token_accuracy": 0.15773278176784516,
      "num_tokens": 13165459.0,
      "step": 5190
    },
    {
      "entropy": 6.511909866333008,
      "epoch": 0.599538372763993,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004973429620293438,
      "loss": 6.4035,
      "mean_token_accuracy": 0.14859146922826766,
      "num_tokens": 13178766.0,
      "step": 5195
    },
    {
      "entropy": 6.49504246711731,
      "epoch": 0.6001154068090018,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004973366354574239,
      "loss": 6.5153,
      "mean_token_accuracy": 0.1469987377524376,
      "num_tokens": 13193712.0,
      "step": 5200
    },
    {
      "entropy": 6.507981824874878,
      "epoch": 0.6006924408540104,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004973303014073219,
      "loss": 6.4971,
      "mean_token_accuracy": 0.15589592456817628,
      "num_tokens": 13207363.0,
      "step": 5205
    },
    {
      "entropy": 6.402018308639526,
      "epoch": 0.601269474899019,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004973239598792504,
      "loss": 6.2918,
      "mean_token_accuracy": 0.16719549894332886,
      "num_tokens": 13219131.0,
      "step": 5210
    },
    {
      "entropy": 6.441523122787475,
      "epoch": 0.6018465089440277,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004973176108734232,
      "loss": 6.4084,
      "mean_token_accuracy": 0.15474872440099716,
      "num_tokens": 13231595.0,
      "step": 5215
    },
    {
      "entropy": 6.509952116012573,
      "epoch": 0.6024235429890363,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004973112543900535,
      "loss": 6.4074,
      "mean_token_accuracy": 0.15591059774160385,
      "num_tokens": 13243727.0,
      "step": 5220
    },
    {
      "entropy": 6.477614974975586,
      "epoch": 0.603000577034045,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004973048904293551,
      "loss": 6.3875,
      "mean_token_accuracy": 0.1595266968011856,
      "num_tokens": 13255002.0,
      "step": 5225
    },
    {
      "entropy": 6.455194807052612,
      "epoch": 0.6035776110790536,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004972985189915422,
      "loss": 6.426,
      "mean_token_accuracy": 0.153925558924675,
      "num_tokens": 13268523.0,
      "step": 5230
    },
    {
      "entropy": 6.3574200630187985,
      "epoch": 0.6041546451240624,
      "grad_norm": 1.03125,
      "learning_rate": 0.000497292140076829,
      "loss": 6.3212,
      "mean_token_accuracy": 0.15963851511478425,
      "num_tokens": 13280625.0,
      "step": 5235
    },
    {
      "entropy": 6.536995458602905,
      "epoch": 0.604731679169071,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004972857536854301,
      "loss": 6.4912,
      "mean_token_accuracy": 0.14426816552877425,
      "num_tokens": 13294880.0,
      "step": 5240
    },
    {
      "entropy": 6.474337816238403,
      "epoch": 0.6053087132140796,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004972793598175603,
      "loss": 6.4199,
      "mean_token_accuracy": 0.15171322450041771,
      "num_tokens": 13307875.0,
      "step": 5245
    },
    {
      "entropy": 6.47451434135437,
      "epoch": 0.6058857472590883,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004972729584734347,
      "loss": 6.2962,
      "mean_token_accuracy": 0.16146529018878936,
      "num_tokens": 13320067.0,
      "step": 5250
    },
    {
      "entropy": 6.345598316192627,
      "epoch": 0.6064627813040969,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004972665496532685,
      "loss": 6.3764,
      "mean_token_accuracy": 0.15575868189334868,
      "num_tokens": 13332036.0,
      "step": 5255
    },
    {
      "entropy": 6.45696177482605,
      "epoch": 0.6070398153491056,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004972601333572774,
      "loss": 6.3856,
      "mean_token_accuracy": 0.15872769802808762,
      "num_tokens": 13343831.0,
      "step": 5260
    },
    {
      "entropy": 6.474030685424805,
      "epoch": 0.6076168493941142,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004972537095856769,
      "loss": 6.4343,
      "mean_token_accuracy": 0.15111970528960228,
      "num_tokens": 13356452.0,
      "step": 5265
    },
    {
      "entropy": 6.401644086837768,
      "epoch": 0.608193883439123,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004972472783386834,
      "loss": 6.3577,
      "mean_token_accuracy": 0.15509337335824966,
      "num_tokens": 13369052.0,
      "step": 5270
    },
    {
      "entropy": 6.507857131958008,
      "epoch": 0.6087709174841316,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004972408396165132,
      "loss": 6.45,
      "mean_token_accuracy": 0.152229243516922,
      "num_tokens": 13381943.0,
      "step": 5275
    },
    {
      "entropy": 6.461705207824707,
      "epoch": 0.6093479515291402,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004972343934193826,
      "loss": 6.4007,
      "mean_token_accuracy": 0.15218027532100678,
      "num_tokens": 13394069.0,
      "step": 5280
    },
    {
      "entropy": 6.466672945022583,
      "epoch": 0.6099249855741489,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004972279397475086,
      "loss": 6.4687,
      "mean_token_accuracy": 0.15327067375183107,
      "num_tokens": 13407397.0,
      "step": 5285
    },
    {
      "entropy": 6.449629306793213,
      "epoch": 0.6105020196191575,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004972214786011083,
      "loss": 6.3874,
      "mean_token_accuracy": 0.1496693566441536,
      "num_tokens": 13419716.0,
      "step": 5290
    },
    {
      "entropy": 6.421255874633789,
      "epoch": 0.6110790536641661,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000497215009980399,
      "loss": 6.3549,
      "mean_token_accuracy": 0.1546877220273018,
      "num_tokens": 13432839.0,
      "step": 5295
    },
    {
      "entropy": 6.450389766693116,
      "epoch": 0.6116560877091748,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000497208533885598,
      "loss": 6.4004,
      "mean_token_accuracy": 0.15024428814649582,
      "num_tokens": 13445318.0,
      "step": 5300
    },
    {
      "entropy": 6.41174168586731,
      "epoch": 0.6122331217541835,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004972020503169235,
      "loss": 6.3756,
      "mean_token_accuracy": 0.15530214309692383,
      "num_tokens": 13456652.0,
      "step": 5305
    },
    {
      "entropy": 6.422957992553711,
      "epoch": 0.6128101557991922,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004971955592745934,
      "loss": 6.3728,
      "mean_token_accuracy": 0.15331609174609184,
      "num_tokens": 13469133.0,
      "step": 5310
    },
    {
      "entropy": 6.466459846496582,
      "epoch": 0.6133871898442008,
      "grad_norm": 1.53125,
      "learning_rate": 0.000497189060758826,
      "loss": 6.3888,
      "mean_token_accuracy": 0.15325451642274857,
      "num_tokens": 13481741.0,
      "step": 5315
    },
    {
      "entropy": 6.476299571990967,
      "epoch": 0.6139642238892095,
      "grad_norm": 1.21875,
      "learning_rate": 0.00049718255476984,
      "loss": 6.4715,
      "mean_token_accuracy": 0.1463205024600029,
      "num_tokens": 13494257.0,
      "step": 5320
    },
    {
      "entropy": 6.401739931106567,
      "epoch": 0.6145412579342181,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004971760413078539,
      "loss": 6.3389,
      "mean_token_accuracy": 0.16656553149223327,
      "num_tokens": 13507714.0,
      "step": 5325
    },
    {
      "entropy": 6.396955442428589,
      "epoch": 0.6151182919792267,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004971695203730872,
      "loss": 6.3824,
      "mean_token_accuracy": 0.1524408280849457,
      "num_tokens": 13519499.0,
      "step": 5330
    },
    {
      "entropy": 6.469518184661865,
      "epoch": 0.6156953260242354,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000497162991965759,
      "loss": 6.3671,
      "mean_token_accuracy": 0.16050689667463303,
      "num_tokens": 13531538.0,
      "step": 5335
    },
    {
      "entropy": 6.513633584976196,
      "epoch": 0.6162723600692441,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004971564560860889,
      "loss": 6.4323,
      "mean_token_accuracy": 0.15556212216615678,
      "num_tokens": 13544576.0,
      "step": 5340
    },
    {
      "entropy": 6.409820222854615,
      "epoch": 0.6168493941142528,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004971499127342968,
      "loss": 6.3697,
      "mean_token_accuracy": 0.1503436639904976,
      "num_tokens": 13557318.0,
      "step": 5345
    },
    {
      "entropy": 6.334451961517334,
      "epoch": 0.6174264281592614,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004971433619106027,
      "loss": 6.2567,
      "mean_token_accuracy": 0.1651250034570694,
      "num_tokens": 13569334.0,
      "step": 5350
    },
    {
      "entropy": 6.422667169570923,
      "epoch": 0.61800346220427,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000497136803615227,
      "loss": 6.3524,
      "mean_token_accuracy": 0.1614537626504898,
      "num_tokens": 13581748.0,
      "step": 5355
    },
    {
      "entropy": 6.433790826797486,
      "epoch": 0.6185804962492787,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004971302378483902,
      "loss": 6.4167,
      "mean_token_accuracy": 0.15154110938310622,
      "num_tokens": 13593892.0,
      "step": 5360
    },
    {
      "entropy": 6.428574800491333,
      "epoch": 0.6191575302942873,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004971236646103132,
      "loss": 6.301,
      "mean_token_accuracy": 0.16057015508413314,
      "num_tokens": 13605593.0,
      "step": 5365
    },
    {
      "entropy": 6.442511415481567,
      "epoch": 0.619734564339296,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004971170839012171,
      "loss": 6.481,
      "mean_token_accuracy": 0.1505902349948883,
      "num_tokens": 13618356.0,
      "step": 5370
    },
    {
      "entropy": 6.45281867980957,
      "epoch": 0.6203115983843047,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004971104957213233,
      "loss": 6.4229,
      "mean_token_accuracy": 0.15445287376642228,
      "num_tokens": 13629939.0,
      "step": 5375
    },
    {
      "entropy": 6.290883255004883,
      "epoch": 0.6208886324293134,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004971039000708531,
      "loss": 6.2762,
      "mean_token_accuracy": 0.16316339373588562,
      "num_tokens": 13642123.0,
      "step": 5380
    },
    {
      "entropy": 6.455602836608887,
      "epoch": 0.621465666474322,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004970972969500286,
      "loss": 6.3683,
      "mean_token_accuracy": 0.15798481702804565,
      "num_tokens": 13654956.0,
      "step": 5385
    },
    {
      "entropy": 6.4037621974945065,
      "epoch": 0.6220427005193306,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000497090686359072,
      "loss": 6.3347,
      "mean_token_accuracy": 0.16185543388128282,
      "num_tokens": 13667549.0,
      "step": 5390
    },
    {
      "entropy": 6.440101575851441,
      "epoch": 0.6226197345643393,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004970840682982054,
      "loss": 6.4033,
      "mean_token_accuracy": 0.15634123980998993,
      "num_tokens": 13680747.0,
      "step": 5395
    },
    {
      "entropy": 6.49430022239685,
      "epoch": 0.6231967686093479,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004970774427676514,
      "loss": 6.4462,
      "mean_token_accuracy": 0.14711104929447175,
      "num_tokens": 13695500.0,
      "step": 5400
    },
    {
      "entropy": 6.469353055953979,
      "epoch": 0.6237738026543566,
      "grad_norm": 1.375,
      "learning_rate": 0.000497070809767633,
      "loss": 6.419,
      "mean_token_accuracy": 0.15500350445508956,
      "num_tokens": 13707807.0,
      "step": 5405
    },
    {
      "entropy": 6.42448844909668,
      "epoch": 0.6243508366993653,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004970641692983731,
      "loss": 6.3486,
      "mean_token_accuracy": 0.15815356224775315,
      "num_tokens": 13720401.0,
      "step": 5410
    },
    {
      "entropy": 6.446044778823852,
      "epoch": 0.624927870744374,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004970575213600954,
      "loss": 6.325,
      "mean_token_accuracy": 0.16023847758769988,
      "num_tokens": 13732611.0,
      "step": 5415
    },
    {
      "entropy": 6.357176494598389,
      "epoch": 0.6255049047893826,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004970508659530231,
      "loss": 6.3995,
      "mean_token_accuracy": 0.1527742862701416,
      "num_tokens": 13744876.0,
      "step": 5420
    },
    {
      "entropy": 6.445756578445435,
      "epoch": 0.6260819388343912,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004970442030773802,
      "loss": 6.4277,
      "mean_token_accuracy": 0.14950262010097504,
      "num_tokens": 13757265.0,
      "step": 5425
    },
    {
      "entropy": 6.550195217132568,
      "epoch": 0.6266589728793999,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004970375327333909,
      "loss": 6.3437,
      "mean_token_accuracy": 0.15652683079242707,
      "num_tokens": 13770101.0,
      "step": 5430
    },
    {
      "entropy": 6.464341163635254,
      "epoch": 0.6272360069244085,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004970308549212796,
      "loss": 6.3907,
      "mean_token_accuracy": 0.15585145503282546,
      "num_tokens": 13784104.0,
      "step": 5435
    },
    {
      "entropy": 6.4140418529510494,
      "epoch": 0.6278130409694171,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004970241696412705,
      "loss": 6.3367,
      "mean_token_accuracy": 0.16261254549026488,
      "num_tokens": 13796901.0,
      "step": 5440
    },
    {
      "entropy": 6.412482213973999,
      "epoch": 0.6283900750144259,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000497017476893589,
      "loss": 6.3957,
      "mean_token_accuracy": 0.1554963245987892,
      "num_tokens": 13809640.0,
      "step": 5445
    },
    {
      "entropy": 6.511721754074097,
      "epoch": 0.6289671090594345,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004970107766784598,
      "loss": 6.3978,
      "mean_token_accuracy": 0.15825678110122682,
      "num_tokens": 13821684.0,
      "step": 5450
    },
    {
      "entropy": 6.422944641113281,
      "epoch": 0.6295441431044432,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004970040689961084,
      "loss": 6.4122,
      "mean_token_accuracy": 0.15048492103815078,
      "num_tokens": 13833973.0,
      "step": 5455
    },
    {
      "entropy": 6.428346252441406,
      "epoch": 0.6301211771494518,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004969973538467605,
      "loss": 6.364,
      "mean_token_accuracy": 0.1592654436826706,
      "num_tokens": 13845954.0,
      "step": 5460
    },
    {
      "entropy": 6.52258038520813,
      "epoch": 0.6306982111944605,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004969906312306419,
      "loss": 6.3851,
      "mean_token_accuracy": 0.15411052033305167,
      "num_tokens": 13858261.0,
      "step": 5465
    },
    {
      "entropy": 6.4935260772705075,
      "epoch": 0.6312752452394691,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004969839011479786,
      "loss": 6.3697,
      "mean_token_accuracy": 0.15756288170814514,
      "num_tokens": 13870409.0,
      "step": 5470
    },
    {
      "entropy": 6.479258728027344,
      "epoch": 0.6318522792844777,
      "grad_norm": 1.015625,
      "learning_rate": 0.000496977163598997,
      "loss": 6.4502,
      "mean_token_accuracy": 0.1490995317697525,
      "num_tokens": 13883980.0,
      "step": 5475
    },
    {
      "entropy": 6.454884004592896,
      "epoch": 0.6324293133294865,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004969704185839239,
      "loss": 6.4757,
      "mean_token_accuracy": 0.1558075003325939,
      "num_tokens": 13897901.0,
      "step": 5480
    },
    {
      "entropy": 6.494711399078369,
      "epoch": 0.6330063473744951,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004969636661029859,
      "loss": 6.4281,
      "mean_token_accuracy": 0.15396897196769715,
      "num_tokens": 13909655.0,
      "step": 5485
    },
    {
      "entropy": 6.542307186126709,
      "epoch": 0.6335833814195038,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004969569061564103,
      "loss": 6.3866,
      "mean_token_accuracy": 0.15406434684991838,
      "num_tokens": 13923121.0,
      "step": 5490
    },
    {
      "entropy": 6.512728452682495,
      "epoch": 0.6341604154645124,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004969501387444245,
      "loss": 6.4745,
      "mean_token_accuracy": 0.14800321459770202,
      "num_tokens": 13936100.0,
      "step": 5495
    },
    {
      "entropy": 6.397259521484375,
      "epoch": 0.634737449509521,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004969433638672559,
      "loss": 6.3731,
      "mean_token_accuracy": 0.1528458282351494,
      "num_tokens": 13949799.0,
      "step": 5500
    },
    {
      "entropy": 6.449781227111816,
      "epoch": 0.6353144835545297,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004969365815251325,
      "loss": 6.3517,
      "mean_token_accuracy": 0.16037549674510956,
      "num_tokens": 13961805.0,
      "step": 5505
    },
    {
      "entropy": 6.431216382980347,
      "epoch": 0.6358915175995383,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004969297917182825,
      "loss": 6.4116,
      "mean_token_accuracy": 0.1560327798128128,
      "num_tokens": 13974264.0,
      "step": 5510
    },
    {
      "entropy": 6.442997694015503,
      "epoch": 0.6364685516445471,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004969229944469342,
      "loss": 6.3583,
      "mean_token_accuracy": 0.15966611355543137,
      "num_tokens": 13986582.0,
      "step": 5515
    },
    {
      "entropy": 6.426758909225464,
      "epoch": 0.6370455856895557,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004969161897113162,
      "loss": 6.3421,
      "mean_token_accuracy": 0.15339141488075256,
      "num_tokens": 13998675.0,
      "step": 5520
    },
    {
      "entropy": 6.422752380371094,
      "epoch": 0.6376226197345644,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004969093775116574,
      "loss": 6.4029,
      "mean_token_accuracy": 0.15763304010033607,
      "num_tokens": 14010744.0,
      "step": 5525
    },
    {
      "entropy": 6.494753170013428,
      "epoch": 0.638199653779573,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004969025578481869,
      "loss": 6.3752,
      "mean_token_accuracy": 0.1540719449520111,
      "num_tokens": 14023721.0,
      "step": 5530
    },
    {
      "entropy": 6.412088108062744,
      "epoch": 0.6387766878245816,
      "grad_norm": 1.09375,
      "learning_rate": 0.000496895730721134,
      "loss": 6.3064,
      "mean_token_accuracy": 0.1553841710090637,
      "num_tokens": 14035998.0,
      "step": 5535
    },
    {
      "entropy": 6.326477575302124,
      "epoch": 0.6393537218695903,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004968888961307286,
      "loss": 6.2854,
      "mean_token_accuracy": 0.1607911095023155,
      "num_tokens": 14048407.0,
      "step": 5540
    },
    {
      "entropy": 6.339253997802734,
      "epoch": 0.6399307559145989,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004968820540772003,
      "loss": 6.2771,
      "mean_token_accuracy": 0.16459511518478392,
      "num_tokens": 14060898.0,
      "step": 5545
    },
    {
      "entropy": 6.44248251914978,
      "epoch": 0.6405077899596077,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004968752045607794,
      "loss": 6.3693,
      "mean_token_accuracy": 0.15352287739515305,
      "num_tokens": 14073474.0,
      "step": 5550
    },
    {
      "entropy": 6.490054559707642,
      "epoch": 0.6410848240046163,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004968683475816961,
      "loss": 6.4382,
      "mean_token_accuracy": 0.15229271054267884,
      "num_tokens": 14085229.0,
      "step": 5555
    },
    {
      "entropy": 6.396782684326172,
      "epoch": 0.641661858049625,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004968614831401811,
      "loss": 6.3663,
      "mean_token_accuracy": 0.1561431884765625,
      "num_tokens": 14097943.0,
      "step": 5560
    },
    {
      "entropy": 6.5521018505096436,
      "epoch": 0.6422388920946336,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004968546112364654,
      "loss": 6.4918,
      "mean_token_accuracy": 0.14537775963544847,
      "num_tokens": 14112397.0,
      "step": 5565
    },
    {
      "entropy": 6.509760475158691,
      "epoch": 0.6428159261396422,
      "grad_norm": 1.015625,
      "learning_rate": 0.00049684773187078,
      "loss": 6.3372,
      "mean_token_accuracy": 0.1628203347325325,
      "num_tokens": 14127096.0,
      "step": 5570
    },
    {
      "entropy": 6.470750141143799,
      "epoch": 0.6433929601846509,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004968408450433565,
      "loss": 6.4032,
      "mean_token_accuracy": 0.15446191430091857,
      "num_tokens": 14139325.0,
      "step": 5575
    },
    {
      "entropy": 6.408360242843628,
      "epoch": 0.6439699942296595,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004968339507544263,
      "loss": 6.372,
      "mean_token_accuracy": 0.15325330793857575,
      "num_tokens": 14151334.0,
      "step": 5580
    },
    {
      "entropy": 6.513002586364746,
      "epoch": 0.6445470282746683,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004968270490042212,
      "loss": 6.2877,
      "mean_token_accuracy": 0.1661871552467346,
      "num_tokens": 14164195.0,
      "step": 5585
    },
    {
      "entropy": 6.350423002243042,
      "epoch": 0.6451240623196769,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004968201397929737,
      "loss": 6.3498,
      "mean_token_accuracy": 0.15434951931238175,
      "num_tokens": 14175937.0,
      "step": 5590
    },
    {
      "entropy": 6.414374256134034,
      "epoch": 0.6457010963646855,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004968132231209158,
      "loss": 6.3717,
      "mean_token_accuracy": 0.15191204100847244,
      "num_tokens": 14189288.0,
      "step": 5595
    },
    {
      "entropy": 6.445919704437256,
      "epoch": 0.6462781304096942,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004968062989882803,
      "loss": 6.3602,
      "mean_token_accuracy": 0.1612029880285263,
      "num_tokens": 14202428.0,
      "step": 5600
    },
    {
      "entropy": 6.365402173995972,
      "epoch": 0.6468551644547028,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004967993673953003,
      "loss": 6.374,
      "mean_token_accuracy": 0.1585848733782768,
      "num_tokens": 14216871.0,
      "step": 5605
    },
    {
      "entropy": 6.427926301956177,
      "epoch": 0.6474321984997115,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004967924283422087,
      "loss": 6.3446,
      "mean_token_accuracy": 0.15603581666946412,
      "num_tokens": 14228503.0,
      "step": 5610
    },
    {
      "entropy": 6.419540977478027,
      "epoch": 0.6480092325447201,
      "grad_norm": 1.078125,
      "learning_rate": 0.000496785481829239,
      "loss": 6.2569,
      "mean_token_accuracy": 0.159651517868042,
      "num_tokens": 14240239.0,
      "step": 5615
    },
    {
      "entropy": 6.431472396850586,
      "epoch": 0.6485862665897288,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004967785278566245,
      "loss": 6.3515,
      "mean_token_accuracy": 0.16212584376335143,
      "num_tokens": 14253180.0,
      "step": 5620
    },
    {
      "entropy": 6.35306830406189,
      "epoch": 0.6491633006347375,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004967715664245997,
      "loss": 6.4028,
      "mean_token_accuracy": 0.15209043100476266,
      "num_tokens": 14267275.0,
      "step": 5625
    },
    {
      "entropy": 6.490705680847168,
      "epoch": 0.6497403346797461,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004967645975333983,
      "loss": 6.4214,
      "mean_token_accuracy": 0.151943638920784,
      "num_tokens": 14279650.0,
      "step": 5630
    },
    {
      "entropy": 6.448146152496338,
      "epoch": 0.6503173687247548,
      "grad_norm": 1.125,
      "learning_rate": 0.0004967576211832548,
      "loss": 6.3107,
      "mean_token_accuracy": 0.15573958605527877,
      "num_tokens": 14291647.0,
      "step": 5635
    },
    {
      "entropy": 6.294909715652466,
      "epoch": 0.6508944027697634,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004967506373744039,
      "loss": 6.3088,
      "mean_token_accuracy": 0.15778163820505142,
      "num_tokens": 14304335.0,
      "step": 5640
    },
    {
      "entropy": 6.458283758163452,
      "epoch": 0.651471436814772,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004967436461070805,
      "loss": 6.3245,
      "mean_token_accuracy": 0.1590469002723694,
      "num_tokens": 14317226.0,
      "step": 5645
    },
    {
      "entropy": 6.445477533340454,
      "epoch": 0.6520484708597807,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004967366473815196,
      "loss": 6.3971,
      "mean_token_accuracy": 0.1622937113046646,
      "num_tokens": 14329000.0,
      "step": 5650
    },
    {
      "entropy": 6.398467636108398,
      "epoch": 0.6526255049047894,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004967296411979568,
      "loss": 6.3551,
      "mean_token_accuracy": 0.15491232872009278,
      "num_tokens": 14341714.0,
      "step": 5655
    },
    {
      "entropy": 6.430992841720581,
      "epoch": 0.6532025389497981,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004967226275566275,
      "loss": 6.343,
      "mean_token_accuracy": 0.15515244752168655,
      "num_tokens": 14353889.0,
      "step": 5660
    },
    {
      "entropy": 6.513474941253662,
      "epoch": 0.6537795729948067,
      "grad_norm": 1.25,
      "learning_rate": 0.000496715606457768,
      "loss": 6.4455,
      "mean_token_accuracy": 0.14796115681529046,
      "num_tokens": 14367093.0,
      "step": 5665
    },
    {
      "entropy": 6.285620260238647,
      "epoch": 0.6543566070398154,
      "grad_norm": 1.375,
      "learning_rate": 0.0004967085779016142,
      "loss": 6.1806,
      "mean_token_accuracy": 0.1702989459037781,
      "num_tokens": 14380064.0,
      "step": 5670
    },
    {
      "entropy": 6.407661008834839,
      "epoch": 0.654933641084824,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004967015418884022,
      "loss": 6.3764,
      "mean_token_accuracy": 0.1556804895401001,
      "num_tokens": 14392026.0,
      "step": 5675
    },
    {
      "entropy": 6.432088088989258,
      "epoch": 0.6555106751298326,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004966944984183692,
      "loss": 6.3355,
      "mean_token_accuracy": 0.15418035686016082,
      "num_tokens": 14404011.0,
      "step": 5680
    },
    {
      "entropy": 6.411783838272095,
      "epoch": 0.6560877091748413,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004966874474917518,
      "loss": 6.3084,
      "mean_token_accuracy": 0.1624397337436676,
      "num_tokens": 14416356.0,
      "step": 5685
    },
    {
      "entropy": 6.378304719924927,
      "epoch": 0.65666474321985,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004966803891087873,
      "loss": 6.3547,
      "mean_token_accuracy": 0.15426328927278518,
      "num_tokens": 14428873.0,
      "step": 5690
    },
    {
      "entropy": 6.3676087856292725,
      "epoch": 0.6572417772648587,
      "grad_norm": 1.015625,
      "learning_rate": 0.000496673323269713,
      "loss": 6.3471,
      "mean_token_accuracy": 0.1549686573445797,
      "num_tokens": 14441703.0,
      "step": 5695
    },
    {
      "entropy": 6.462108898162842,
      "epoch": 0.6578188113098673,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004966662499747666,
      "loss": 6.3378,
      "mean_token_accuracy": 0.15838538110256195,
      "num_tokens": 14453701.0,
      "step": 5700
    },
    {
      "entropy": 6.442297697067261,
      "epoch": 0.6583958453548759,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004966591692241859,
      "loss": 6.3605,
      "mean_token_accuracy": 0.159871344268322,
      "num_tokens": 14466108.0,
      "step": 5705
    },
    {
      "entropy": 6.382602977752685,
      "epoch": 0.6589728793998846,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004966520810182092,
      "loss": 6.4249,
      "mean_token_accuracy": 0.15794423073530198,
      "num_tokens": 14479130.0,
      "step": 5710
    },
    {
      "entropy": 6.496137189865112,
      "epoch": 0.6595499134448932,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004966449853570749,
      "loss": 6.3762,
      "mean_token_accuracy": 0.1580379068851471,
      "num_tokens": 14492797.0,
      "step": 5715
    },
    {
      "entropy": 6.4269836902618405,
      "epoch": 0.6601269474899019,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004966378822410215,
      "loss": 6.2976,
      "mean_token_accuracy": 0.15341187492012978,
      "num_tokens": 14505279.0,
      "step": 5720
    },
    {
      "entropy": 6.4128107070922855,
      "epoch": 0.6607039815349106,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004966307716702881,
      "loss": 6.3927,
      "mean_token_accuracy": 0.1563589259982109,
      "num_tokens": 14517867.0,
      "step": 5725
    },
    {
      "entropy": 6.408047723770141,
      "epoch": 0.6612810155799193,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004966236536451138,
      "loss": 6.3127,
      "mean_token_accuracy": 0.16040023863315583,
      "num_tokens": 14529804.0,
      "step": 5730
    },
    {
      "entropy": 6.335665273666382,
      "epoch": 0.6618580496249279,
      "grad_norm": 1.328125,
      "learning_rate": 0.000496616528165738,
      "loss": 6.1953,
      "mean_token_accuracy": 0.1733541414141655,
      "num_tokens": 14541484.0,
      "step": 5735
    },
    {
      "entropy": 6.338911056518555,
      "epoch": 0.6624350836699365,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004966093952324003,
      "loss": 6.29,
      "mean_token_accuracy": 0.16199951022863388,
      "num_tokens": 14554796.0,
      "step": 5740
    },
    {
      "entropy": 6.33594856262207,
      "epoch": 0.6630121177149452,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004966022548453406,
      "loss": 6.229,
      "mean_token_accuracy": 0.17701880484819413,
      "num_tokens": 14567090.0,
      "step": 5745
    },
    {
      "entropy": 6.365872430801391,
      "epoch": 0.6635891517599538,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004965951070047993,
      "loss": 6.3498,
      "mean_token_accuracy": 0.1567435458302498,
      "num_tokens": 14580334.0,
      "step": 5750
    },
    {
      "entropy": 6.41003999710083,
      "epoch": 0.6641661858049625,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004965879517110166,
      "loss": 6.2838,
      "mean_token_accuracy": 0.1624019905924797,
      "num_tokens": 14593670.0,
      "step": 5755
    },
    {
      "entropy": 6.389162254333496,
      "epoch": 0.6647432198499712,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004965807889642333,
      "loss": 6.2998,
      "mean_token_accuracy": 0.1612920418381691,
      "num_tokens": 14606612.0,
      "step": 5760
    },
    {
      "entropy": 6.432580184936524,
      "epoch": 0.6653202538949798,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00049657361876469,
      "loss": 6.3209,
      "mean_token_accuracy": 0.15926805287599563,
      "num_tokens": 14618323.0,
      "step": 5765
    },
    {
      "entropy": 6.317815780639648,
      "epoch": 0.6658972879399885,
      "grad_norm": 1.125,
      "learning_rate": 0.0004965664411126282,
      "loss": 6.3178,
      "mean_token_accuracy": 0.15956676304340361,
      "num_tokens": 14631580.0,
      "step": 5770
    },
    {
      "entropy": 6.396694374084473,
      "epoch": 0.6664743219849971,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004965592560082894,
      "loss": 6.2406,
      "mean_token_accuracy": 0.16112178713083267,
      "num_tokens": 14643429.0,
      "step": 5775
    },
    {
      "entropy": 6.481974506378174,
      "epoch": 0.6670513560300058,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004965520634519149,
      "loss": 6.4542,
      "mean_token_accuracy": 0.15182799398899077,
      "num_tokens": 14656890.0,
      "step": 5780
    },
    {
      "entropy": 6.357728338241577,
      "epoch": 0.6676283900750144,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004965448634437468,
      "loss": 6.3569,
      "mean_token_accuracy": 0.15490619838237762,
      "num_tokens": 14670721.0,
      "step": 5785
    },
    {
      "entropy": 6.474006032943725,
      "epoch": 0.668205424120023,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004965376559840272,
      "loss": 6.3668,
      "mean_token_accuracy": 0.1540812149643898,
      "num_tokens": 14682808.0,
      "step": 5790
    },
    {
      "entropy": 6.406538820266723,
      "epoch": 0.6687824581650318,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004965304410729986,
      "loss": 6.3472,
      "mean_token_accuracy": 0.16533400267362594,
      "num_tokens": 14694580.0,
      "step": 5795
    },
    {
      "entropy": 6.406550645828247,
      "epoch": 0.6693594922100404,
      "grad_norm": 1.25,
      "learning_rate": 0.0004965232187109037,
      "loss": 6.3702,
      "mean_token_accuracy": 0.15002246648073198,
      "num_tokens": 14707832.0,
      "step": 5800
    },
    {
      "entropy": 6.479951620101929,
      "epoch": 0.6699365262550491,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004965159888979854,
      "loss": 6.3762,
      "mean_token_accuracy": 0.1557182252407074,
      "num_tokens": 14719293.0,
      "step": 5805
    },
    {
      "entropy": 6.3719995498657225,
      "epoch": 0.6705135603000577,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004965087516344869,
      "loss": 6.2593,
      "mean_token_accuracy": 0.1609267607331276,
      "num_tokens": 14731421.0,
      "step": 5810
    },
    {
      "entropy": 6.385224628448486,
      "epoch": 0.6710905943450663,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004965015069206515,
      "loss": 6.3663,
      "mean_token_accuracy": 0.15497736930847167,
      "num_tokens": 14744160.0,
      "step": 5815
    },
    {
      "entropy": 6.485455846786499,
      "epoch": 0.671667628390075,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000496494254756723,
      "loss": 6.3743,
      "mean_token_accuracy": 0.15152983143925666,
      "num_tokens": 14756820.0,
      "step": 5820
    },
    {
      "entropy": 6.383307981491089,
      "epoch": 0.6722446624350836,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004964869951429451,
      "loss": 6.3202,
      "mean_token_accuracy": 0.1601814091205597,
      "num_tokens": 14770000.0,
      "step": 5825
    },
    {
      "entropy": 6.42850890159607,
      "epoch": 0.6728216964800924,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004964797280795622,
      "loss": 6.4055,
      "mean_token_accuracy": 0.15470815896987916,
      "num_tokens": 14782777.0,
      "step": 5830
    },
    {
      "entropy": 6.456723546981811,
      "epoch": 0.673398730525101,
      "grad_norm": 1.5,
      "learning_rate": 0.0004964724535668188,
      "loss": 6.4232,
      "mean_token_accuracy": 0.15354692935943604,
      "num_tokens": 14796561.0,
      "step": 5835
    },
    {
      "entropy": 6.482658529281617,
      "epoch": 0.6739757645701097,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004964651716049593,
      "loss": 6.3815,
      "mean_token_accuracy": 0.15633953362703323,
      "num_tokens": 14809975.0,
      "step": 5840
    },
    {
      "entropy": 6.483266162872314,
      "epoch": 0.6745527986151183,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004964578821942288,
      "loss": 6.4186,
      "mean_token_accuracy": 0.15237474739551543,
      "num_tokens": 14823578.0,
      "step": 5845
    },
    {
      "entropy": 6.511593341827393,
      "epoch": 0.6751298326601269,
      "grad_norm": 1.125,
      "learning_rate": 0.0004964505853348724,
      "loss": 6.3897,
      "mean_token_accuracy": 0.15733065456151962,
      "num_tokens": 14836520.0,
      "step": 5850
    },
    {
      "entropy": 6.478420352935791,
      "epoch": 0.6757068667051356,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004964432810271356,
      "loss": 6.4127,
      "mean_token_accuracy": 0.15258462578058243,
      "num_tokens": 14847806.0,
      "step": 5855
    },
    {
      "entropy": 6.373460102081299,
      "epoch": 0.6762839007501442,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004964359692712641,
      "loss": 6.2865,
      "mean_token_accuracy": 0.16094840466976165,
      "num_tokens": 14859751.0,
      "step": 5860
    },
    {
      "entropy": 6.4748025894165036,
      "epoch": 0.676860934795153,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004964286500675037,
      "loss": 6.3248,
      "mean_token_accuracy": 0.1585204929113388,
      "num_tokens": 14872215.0,
      "step": 5865
    },
    {
      "entropy": 6.4774822235107425,
      "epoch": 0.6774379688401616,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004964213234161007,
      "loss": 6.3927,
      "mean_token_accuracy": 0.15827755331993104,
      "num_tokens": 14884895.0,
      "step": 5870
    },
    {
      "entropy": 6.452782249450683,
      "epoch": 0.6780150028851702,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004964139893173014,
      "loss": 6.3416,
      "mean_token_accuracy": 0.16145257353782655,
      "num_tokens": 14897131.0,
      "step": 5875
    },
    {
      "entropy": 6.370659875869751,
      "epoch": 0.6785920369301789,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004964066477713525,
      "loss": 6.2874,
      "mean_token_accuracy": 0.16406427472829818,
      "num_tokens": 14908835.0,
      "step": 5880
    },
    {
      "entropy": 6.399390459060669,
      "epoch": 0.6791690709751875,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004963992987785011,
      "loss": 6.3846,
      "mean_token_accuracy": 0.15855973213911057,
      "num_tokens": 14920808.0,
      "step": 5885
    },
    {
      "entropy": 6.458285140991211,
      "epoch": 0.6797461050201962,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004963919423389942,
      "loss": 6.2375,
      "mean_token_accuracy": 0.1642283409833908,
      "num_tokens": 14932558.0,
      "step": 5890
    },
    {
      "entropy": 6.297548389434814,
      "epoch": 0.6803231390652048,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004963845784530794,
      "loss": 6.3511,
      "mean_token_accuracy": 0.15776774287223816,
      "num_tokens": 14946623.0,
      "step": 5895
    },
    {
      "entropy": 6.378008270263672,
      "epoch": 0.6809001731102136,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000496377207121004,
      "loss": 6.3678,
      "mean_token_accuracy": 0.16057287603616716,
      "num_tokens": 14959168.0,
      "step": 5900
    },
    {
      "entropy": 6.521073818206787,
      "epoch": 0.6814772071552222,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004963698283430164,
      "loss": 6.3899,
      "mean_token_accuracy": 0.1492276594042778,
      "num_tokens": 14972710.0,
      "step": 5905
    },
    {
      "entropy": 6.413679790496826,
      "epoch": 0.6820542412002308,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004963624421193646,
      "loss": 6.2531,
      "mean_token_accuracy": 0.16104798913002014,
      "num_tokens": 14984707.0,
      "step": 5910
    },
    {
      "entropy": 6.247140789031983,
      "epoch": 0.6826312752452395,
      "grad_norm": 1.0625,
      "learning_rate": 0.000496355048450297,
      "loss": 6.2789,
      "mean_token_accuracy": 0.1632912799715996,
      "num_tokens": 14996623.0,
      "step": 5915
    },
    {
      "entropy": 6.468965721130371,
      "epoch": 0.6832083092902481,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004963476473360623,
      "loss": 6.3952,
      "mean_token_accuracy": 0.15641281455755235,
      "num_tokens": 15009579.0,
      "step": 5920
    },
    {
      "entropy": 6.431449842453003,
      "epoch": 0.6837853433352568,
      "grad_norm": 1.125,
      "learning_rate": 0.0004963402387769094,
      "loss": 6.3415,
      "mean_token_accuracy": 0.1618417516350746,
      "num_tokens": 15022239.0,
      "step": 5925
    },
    {
      "entropy": 6.412566518783569,
      "epoch": 0.6843623773802654,
      "grad_norm": 1.25,
      "learning_rate": 0.0004963328227730876,
      "loss": 6.3535,
      "mean_token_accuracy": 0.15646750032901763,
      "num_tokens": 15034724.0,
      "step": 5930
    },
    {
      "entropy": 6.368452167510986,
      "epoch": 0.684939411425274,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004963253993248461,
      "loss": 6.2741,
      "mean_token_accuracy": 0.1602345511317253,
      "num_tokens": 15047534.0,
      "step": 5935
    },
    {
      "entropy": 6.4526793479919435,
      "epoch": 0.6855164454702828,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004963179684324349,
      "loss": 6.3696,
      "mean_token_accuracy": 0.1580394074320793,
      "num_tokens": 15060468.0,
      "step": 5940
    },
    {
      "entropy": 6.470442628860473,
      "epoch": 0.6860934795152914,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004963105300961036,
      "loss": 6.3424,
      "mean_token_accuracy": 0.15655080676078797,
      "num_tokens": 15073424.0,
      "step": 5945
    },
    {
      "entropy": 6.379420328140259,
      "epoch": 0.6866705135603001,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004963030843161026,
      "loss": 6.3928,
      "mean_token_accuracy": 0.1477957546710968,
      "num_tokens": 15086582.0,
      "step": 5950
    },
    {
      "entropy": 6.454089879989624,
      "epoch": 0.6872475476053087,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004962956310926823,
      "loss": 6.3865,
      "mean_token_accuracy": 0.15109603703022004,
      "num_tokens": 15099288.0,
      "step": 5955
    },
    {
      "entropy": 6.470096492767334,
      "epoch": 0.6878245816503173,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004962881704260934,
      "loss": 6.3128,
      "mean_token_accuracy": 0.16175386160612107,
      "num_tokens": 15111800.0,
      "step": 5960
    },
    {
      "entropy": 6.4026881694793705,
      "epoch": 0.688401615695326,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004962807023165868,
      "loss": 6.3462,
      "mean_token_accuracy": 0.15559173971414567,
      "num_tokens": 15123384.0,
      "step": 5965
    },
    {
      "entropy": 6.316280031204224,
      "epoch": 0.6889786497403346,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004962732267644138,
      "loss": 6.2621,
      "mean_token_accuracy": 0.15509586781263351,
      "num_tokens": 15134857.0,
      "step": 5970
    },
    {
      "entropy": 6.403274297714233,
      "epoch": 0.6895556837853434,
      "grad_norm": 1.0,
      "learning_rate": 0.0004962657437698254,
      "loss": 6.3995,
      "mean_token_accuracy": 0.15861170887947082,
      "num_tokens": 15149124.0,
      "step": 5975
    },
    {
      "entropy": 6.2963621616363525,
      "epoch": 0.690132717830352,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004962582533330739,
      "loss": 6.1791,
      "mean_token_accuracy": 0.17183619886636733,
      "num_tokens": 15163315.0,
      "step": 5980
    },
    {
      "entropy": 6.44432053565979,
      "epoch": 0.6907097518753607,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004962507554544109,
      "loss": 6.3562,
      "mean_token_accuracy": 0.1551659256219864,
      "num_tokens": 15175371.0,
      "step": 5985
    },
    {
      "entropy": 6.438646078109741,
      "epoch": 0.6912867859203693,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004962432501340886,
      "loss": 6.3286,
      "mean_token_accuracy": 0.1565194919705391,
      "num_tokens": 15187633.0,
      "step": 5990
    },
    {
      "entropy": 6.397422838211059,
      "epoch": 0.6918638199653779,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004962357373723596,
      "loss": 6.363,
      "mean_token_accuracy": 0.153975972533226,
      "num_tokens": 15199652.0,
      "step": 5995
    },
    {
      "entropy": 6.447413349151612,
      "epoch": 0.6924408540103866,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004962282171694763,
      "loss": 6.3718,
      "mean_token_accuracy": 0.15816727578639983,
      "num_tokens": 15213309.0,
      "step": 6000
    },
    {
      "epoch": 0.6924408540103866,
      "eval_entropy": 6.365170311853226,
      "eval_loss": 6.399885177612305,
      "eval_mean_token_accuracy": 0.15836162994704972,
      "eval_num_tokens": 15213309.0,
      "eval_runtime": 23.0881,
      "eval_samples_per_second": 1218.636,
      "eval_steps_per_second": 152.33,
      "step": 6000
    },
    {
      "entropy": 6.434838533401489,
      "epoch": 0.6930178880553952,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004962206895256919,
      "loss": 6.3227,
      "mean_token_accuracy": 0.15490258783102034,
      "num_tokens": 15226466.0,
      "step": 6005
    },
    {
      "entropy": 6.350151920318604,
      "epoch": 0.693594922100404,
      "grad_norm": 1.0,
      "learning_rate": 0.0004962131544412595,
      "loss": 6.2355,
      "mean_token_accuracy": 0.160215525329113,
      "num_tokens": 15239306.0,
      "step": 6010
    },
    {
      "entropy": 6.3231117725372314,
      "epoch": 0.6941719561454126,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004962056119164325,
      "loss": 6.3736,
      "mean_token_accuracy": 0.15513332039117814,
      "num_tokens": 15252365.0,
      "step": 6015
    },
    {
      "entropy": 6.539852333068848,
      "epoch": 0.6947489901904212,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004961980619514646,
      "loss": 6.3769,
      "mean_token_accuracy": 0.15713487714529037,
      "num_tokens": 15265770.0,
      "step": 6020
    },
    {
      "entropy": 6.454382371902466,
      "epoch": 0.6953260242354299,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004961905045466098,
      "loss": 6.3705,
      "mean_token_accuracy": 0.15594374164938926,
      "num_tokens": 15278766.0,
      "step": 6025
    },
    {
      "entropy": 6.44936056137085,
      "epoch": 0.6959030582804385,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004961829397021222,
      "loss": 6.3979,
      "mean_token_accuracy": 0.15106612890958787,
      "num_tokens": 15292180.0,
      "step": 6030
    },
    {
      "entropy": 6.527615118026733,
      "epoch": 0.6964800923254472,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004961753674182564,
      "loss": 6.4024,
      "mean_token_accuracy": 0.1545901194214821,
      "num_tokens": 15305313.0,
      "step": 6035
    },
    {
      "entropy": 6.460575008392334,
      "epoch": 0.6970571263704558,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004961677876952669,
      "loss": 6.4529,
      "mean_token_accuracy": 0.14837836623191833,
      "num_tokens": 15318734.0,
      "step": 6040
    },
    {
      "entropy": 6.408411502838135,
      "epoch": 0.6976341604154646,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004961602005334087,
      "loss": 6.3084,
      "mean_token_accuracy": 0.15698490887880326,
      "num_tokens": 15330915.0,
      "step": 6045
    },
    {
      "entropy": 6.48039698600769,
      "epoch": 0.6982111944604732,
      "grad_norm": 1.09375,
      "learning_rate": 0.000496152605932937,
      "loss": 6.413,
      "mean_token_accuracy": 0.1527087762951851,
      "num_tokens": 15343634.0,
      "step": 6050
    },
    {
      "entropy": 6.43453049659729,
      "epoch": 0.6987882285054818,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004961450038941074,
      "loss": 6.346,
      "mean_token_accuracy": 0.15504136309027672,
      "num_tokens": 15356049.0,
      "step": 6055
    },
    {
      "entropy": 6.374099016189575,
      "epoch": 0.6993652625504905,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004961373944171754,
      "loss": 6.2034,
      "mean_token_accuracy": 0.16652617007493972,
      "num_tokens": 15367804.0,
      "step": 6060
    },
    {
      "entropy": 6.345891332626342,
      "epoch": 0.6999422965954991,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004961297775023968,
      "loss": 6.3386,
      "mean_token_accuracy": 0.15769254565238952,
      "num_tokens": 15379921.0,
      "step": 6065
    },
    {
      "entropy": 6.381840372085572,
      "epoch": 0.7005193306405078,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000496122153150028,
      "loss": 6.305,
      "mean_token_accuracy": 0.1672724038362503,
      "num_tokens": 15393043.0,
      "step": 6070
    },
    {
      "entropy": 6.46082649230957,
      "epoch": 0.7010963646855164,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004961145213603255,
      "loss": 6.3838,
      "mean_token_accuracy": 0.15145749300718309,
      "num_tokens": 15407577.0,
      "step": 6075
    },
    {
      "entropy": 6.420774984359741,
      "epoch": 0.7016733987305251,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000496106882133546,
      "loss": 6.3874,
      "mean_token_accuracy": 0.15847085267305375,
      "num_tokens": 15420157.0,
      "step": 6080
    },
    {
      "entropy": 6.355338621139526,
      "epoch": 0.7022504327755338,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004960992354699463,
      "loss": 6.3433,
      "mean_token_accuracy": 0.15855381339788438,
      "num_tokens": 15432882.0,
      "step": 6085
    },
    {
      "entropy": 6.414382886886597,
      "epoch": 0.7028274668205424,
      "grad_norm": 1.125,
      "learning_rate": 0.0004960915813697836,
      "loss": 6.2785,
      "mean_token_accuracy": 0.1584576264023781,
      "num_tokens": 15444541.0,
      "step": 6090
    },
    {
      "entropy": 6.374680280685425,
      "epoch": 0.7034045008655511,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004960839198333154,
      "loss": 6.3886,
      "mean_token_accuracy": 0.1521532028913498,
      "num_tokens": 15457104.0,
      "step": 6095
    },
    {
      "entropy": 6.474993896484375,
      "epoch": 0.7039815349105597,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004960762508607994,
      "loss": 6.3363,
      "mean_token_accuracy": 0.15963913798332213,
      "num_tokens": 15469724.0,
      "step": 6100
    },
    {
      "entropy": 6.408040475845337,
      "epoch": 0.7045585689555683,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004960685744524934,
      "loss": 6.3615,
      "mean_token_accuracy": 0.15959097146987916,
      "num_tokens": 15481261.0,
      "step": 6105
    },
    {
      "entropy": 6.349176836013794,
      "epoch": 0.705135603000577,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004960608906086558,
      "loss": 6.3263,
      "mean_token_accuracy": 0.15958225280046462,
      "num_tokens": 15494996.0,
      "step": 6110
    },
    {
      "entropy": 6.383945178985596,
      "epoch": 0.7057126370455857,
      "grad_norm": 1.15625,
      "learning_rate": 0.000496053199329545,
      "loss": 6.2653,
      "mean_token_accuracy": 0.1622140273451805,
      "num_tokens": 15506772.0,
      "step": 6115
    },
    {
      "entropy": 6.397179031372071,
      "epoch": 0.7062896710905944,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004960455006154196,
      "loss": 6.3788,
      "mean_token_accuracy": 0.1565240055322647,
      "num_tokens": 15520336.0,
      "step": 6120
    },
    {
      "entropy": 6.496908664703369,
      "epoch": 0.706866705135603,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004960377944665386,
      "loss": 6.4599,
      "mean_token_accuracy": 0.144797170907259,
      "num_tokens": 15535139.0,
      "step": 6125
    },
    {
      "entropy": 6.467147493362427,
      "epoch": 0.7074437391806117,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004960300808831611,
      "loss": 6.3347,
      "mean_token_accuracy": 0.16120013743638992,
      "num_tokens": 15547523.0,
      "step": 6130
    },
    {
      "entropy": 6.36202392578125,
      "epoch": 0.7080207732256203,
      "grad_norm": 1.125,
      "learning_rate": 0.0004960223598655467,
      "loss": 6.3274,
      "mean_token_accuracy": 0.16222111135721207,
      "num_tokens": 15560816.0,
      "step": 6135
    },
    {
      "entropy": 6.441680145263672,
      "epoch": 0.7085978072706289,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000496014631413955,
      "loss": 6.3397,
      "mean_token_accuracy": 0.1555134743452072,
      "num_tokens": 15574364.0,
      "step": 6140
    },
    {
      "entropy": 6.285444116592407,
      "epoch": 0.7091748413156376,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004960068955286459,
      "loss": 6.1421,
      "mean_token_accuracy": 0.17765824496746063,
      "num_tokens": 15586969.0,
      "step": 6145
    },
    {
      "entropy": 6.409412527084351,
      "epoch": 0.7097518753606463,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004959991522098797,
      "loss": 6.3262,
      "mean_token_accuracy": 0.15116020292043686,
      "num_tokens": 15600459.0,
      "step": 6150
    },
    {
      "entropy": 6.448535394668579,
      "epoch": 0.710328909405655,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004959914014579168,
      "loss": 6.374,
      "mean_token_accuracy": 0.1490347906947136,
      "num_tokens": 15613484.0,
      "step": 6155
    },
    {
      "entropy": 6.388345050811767,
      "epoch": 0.7109059434506636,
      "grad_norm": 1.25,
      "learning_rate": 0.0004959836432730178,
      "loss": 6.3652,
      "mean_token_accuracy": 0.15277317613363267,
      "num_tokens": 15626036.0,
      "step": 6160
    },
    {
      "entropy": 6.245787715911865,
      "epoch": 0.7114829774956722,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004959758776554438,
      "loss": 6.3123,
      "mean_token_accuracy": 0.17007513344287872,
      "num_tokens": 15640229.0,
      "step": 6165
    },
    {
      "entropy": 6.4315046787261965,
      "epoch": 0.7120600115406809,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000495968104605456,
      "loss": 6.201,
      "mean_token_accuracy": 0.16845831722021104,
      "num_tokens": 15652655.0,
      "step": 6170
    },
    {
      "entropy": 6.439366960525513,
      "epoch": 0.7126370455856895,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004959603241233157,
      "loss": 6.3651,
      "mean_token_accuracy": 0.15772322565317154,
      "num_tokens": 15665535.0,
      "step": 6175
    },
    {
      "entropy": 6.400780725479126,
      "epoch": 0.7132140796306982,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004959525362092846,
      "loss": 6.3344,
      "mean_token_accuracy": 0.1570476308465004,
      "num_tokens": 15678631.0,
      "step": 6180
    },
    {
      "entropy": 6.44068169593811,
      "epoch": 0.7137911136757069,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004959447408636247,
      "loss": 6.4116,
      "mean_token_accuracy": 0.1541247546672821,
      "num_tokens": 15691662.0,
      "step": 6185
    },
    {
      "entropy": 6.447620677947998,
      "epoch": 0.7143681477207156,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004959369380865983,
      "loss": 6.3363,
      "mean_token_accuracy": 0.16041404604911805,
      "num_tokens": 15705017.0,
      "step": 6190
    },
    {
      "entropy": 6.4838611602783205,
      "epoch": 0.7149451817657242,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004959291278784676,
      "loss": 6.3283,
      "mean_token_accuracy": 0.157425194978714,
      "num_tokens": 15717537.0,
      "step": 6195
    },
    {
      "entropy": 6.357590961456299,
      "epoch": 0.7155222158107328,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004959213102394954,
      "loss": 6.1948,
      "mean_token_accuracy": 0.15846336930990218,
      "num_tokens": 15730075.0,
      "step": 6200
    },
    {
      "entropy": 6.400454807281494,
      "epoch": 0.7160992498557415,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004959134851699449,
      "loss": 6.287,
      "mean_token_accuracy": 0.1596755638718605,
      "num_tokens": 15743187.0,
      "step": 6205
    },
    {
      "entropy": 6.392225646972657,
      "epoch": 0.7166762839007501,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004959056526700788,
      "loss": 6.3808,
      "mean_token_accuracy": 0.14887106716632842,
      "num_tokens": 15755459.0,
      "step": 6210
    },
    {
      "entropy": 6.362481260299683,
      "epoch": 0.7172533179457588,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004958978127401609,
      "loss": 6.2067,
      "mean_token_accuracy": 0.16227734386920928,
      "num_tokens": 15767929.0,
      "step": 6215
    },
    {
      "entropy": 6.310817003250122,
      "epoch": 0.7178303519907675,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004958899653804548,
      "loss": 6.2217,
      "mean_token_accuracy": 0.1684525191783905,
      "num_tokens": 15780079.0,
      "step": 6220
    },
    {
      "entropy": 6.461270189285278,
      "epoch": 0.7184073860357761,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004958821105912246,
      "loss": 6.3086,
      "mean_token_accuracy": 0.159131196141243,
      "num_tokens": 15792417.0,
      "step": 6225
    },
    {
      "entropy": 6.395058441162109,
      "epoch": 0.7189844200807848,
      "grad_norm": 1.140625,
      "learning_rate": 0.000495874248372734,
      "loss": 6.3089,
      "mean_token_accuracy": 0.16227512508630754,
      "num_tokens": 15804993.0,
      "step": 6230
    },
    {
      "entropy": 6.347893571853637,
      "epoch": 0.7195614541257934,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000495866378725248,
      "loss": 6.3674,
      "mean_token_accuracy": 0.15955740809440613,
      "num_tokens": 15819765.0,
      "step": 6235
    },
    {
      "entropy": 6.434380197525025,
      "epoch": 0.7201384881708021,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000495858501649031,
      "loss": 6.3433,
      "mean_token_accuracy": 0.15384828001260759,
      "num_tokens": 15831992.0,
      "step": 6240
    },
    {
      "entropy": 6.3990226745605465,
      "epoch": 0.7207155222158107,
      "grad_norm": 1.171875,
      "learning_rate": 0.000495850617144348,
      "loss": 6.3123,
      "mean_token_accuracy": 0.16566115766763687,
      "num_tokens": 15845348.0,
      "step": 6245
    },
    {
      "entropy": 6.357735061645508,
      "epoch": 0.7212925562608193,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004958427252114643,
      "loss": 6.3328,
      "mean_token_accuracy": 0.1589762821793556,
      "num_tokens": 15857677.0,
      "step": 6250
    },
    {
      "entropy": 6.334104681015015,
      "epoch": 0.7218695903058281,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000495834825850645,
      "loss": 6.281,
      "mean_token_accuracy": 0.16771712750196457,
      "num_tokens": 15871626.0,
      "step": 6255
    },
    {
      "entropy": 6.4957112789154055,
      "epoch": 0.7224466243508367,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004958269190621562,
      "loss": 6.412,
      "mean_token_accuracy": 0.15334925055503845,
      "num_tokens": 15884372.0,
      "step": 6260
    },
    {
      "entropy": 6.297770977020264,
      "epoch": 0.7230236583958454,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004958190048462637,
      "loss": 6.2718,
      "mean_token_accuracy": 0.16347897797822952,
      "num_tokens": 15896079.0,
      "step": 6265
    },
    {
      "entropy": 6.387682294845581,
      "epoch": 0.723600692440854,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004958110832032336,
      "loss": 6.3748,
      "mean_token_accuracy": 0.15538979172706605,
      "num_tokens": 15909528.0,
      "step": 6270
    },
    {
      "entropy": 6.45740327835083,
      "epoch": 0.7241777264858626,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004958031541333322,
      "loss": 6.3095,
      "mean_token_accuracy": 0.16044122576713563,
      "num_tokens": 15921247.0,
      "step": 6275
    },
    {
      "entropy": 6.467119598388672,
      "epoch": 0.7247547605308713,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004957952176368267,
      "loss": 6.3678,
      "mean_token_accuracy": 0.1579172983765602,
      "num_tokens": 15933093.0,
      "step": 6280
    },
    {
      "entropy": 6.354182529449463,
      "epoch": 0.7253317945758799,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004957872737139836,
      "loss": 6.3356,
      "mean_token_accuracy": 0.1578660324215889,
      "num_tokens": 15944888.0,
      "step": 6285
    },
    {
      "entropy": 6.489511966705322,
      "epoch": 0.7259088286208887,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004957793223650702,
      "loss": 6.373,
      "mean_token_accuracy": 0.1526200830936432,
      "num_tokens": 15957793.0,
      "step": 6290
    },
    {
      "entropy": 6.477022743225097,
      "epoch": 0.7264858626658973,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000495771363590354,
      "loss": 6.3584,
      "mean_token_accuracy": 0.14766902029514312,
      "num_tokens": 15970152.0,
      "step": 6295
    },
    {
      "entropy": 6.369326972961426,
      "epoch": 0.727062896710906,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004957633973901027,
      "loss": 6.3461,
      "mean_token_accuracy": 0.1584468200802803,
      "num_tokens": 15983347.0,
      "step": 6300
    },
    {
      "entropy": 6.353056287765503,
      "epoch": 0.7276399307559146,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004957554237645841,
      "loss": 6.3317,
      "mean_token_accuracy": 0.16123766601085662,
      "num_tokens": 15996148.0,
      "step": 6305
    },
    {
      "entropy": 6.465492105484008,
      "epoch": 0.7282169648009232,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004957474427140665,
      "loss": 6.3255,
      "mean_token_accuracy": 0.1620597556233406,
      "num_tokens": 16008833.0,
      "step": 6310
    },
    {
      "entropy": 6.3144227981567385,
      "epoch": 0.7287939988459319,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004957394542388182,
      "loss": 6.2602,
      "mean_token_accuracy": 0.16248877495527267,
      "num_tokens": 16020674.0,
      "step": 6315
    },
    {
      "entropy": 6.335475444793701,
      "epoch": 0.7293710328909405,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004957314583391082,
      "loss": 6.2317,
      "mean_token_accuracy": 0.16915848404169082,
      "num_tokens": 16033229.0,
      "step": 6320
    },
    {
      "entropy": 6.442795991897583,
      "epoch": 0.7299480669359493,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004957234550152052,
      "loss": 6.2427,
      "mean_token_accuracy": 0.15739369317889212,
      "num_tokens": 16045646.0,
      "step": 6325
    },
    {
      "entropy": 6.36692476272583,
      "epoch": 0.7305251009809579,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004957154442673784,
      "loss": 6.317,
      "mean_token_accuracy": 0.157645845413208,
      "num_tokens": 16058400.0,
      "step": 6330
    },
    {
      "entropy": 6.2602691650390625,
      "epoch": 0.7311021350259665,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004957074260958972,
      "loss": 6.3198,
      "mean_token_accuracy": 0.1575326144695282,
      "num_tokens": 16071067.0,
      "step": 6335
    },
    {
      "entropy": 6.474333477020264,
      "epoch": 0.7316791690709752,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004956994005010315,
      "loss": 6.3204,
      "mean_token_accuracy": 0.1546676605939865,
      "num_tokens": 16083760.0,
      "step": 6340
    },
    {
      "entropy": 6.3913445472717285,
      "epoch": 0.7322562031159838,
      "grad_norm": 1.1875,
      "learning_rate": 0.000495691367483051,
      "loss": 6.2693,
      "mean_token_accuracy": 0.16848120093345642,
      "num_tokens": 16095469.0,
      "step": 6345
    },
    {
      "entropy": 6.313849258422851,
      "epoch": 0.7328332371609925,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004956833270422259,
      "loss": 6.3063,
      "mean_token_accuracy": 0.15668762624263763,
      "num_tokens": 16108215.0,
      "step": 6350
    },
    {
      "entropy": 6.494552421569824,
      "epoch": 0.7334102712060011,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004956752791788269,
      "loss": 6.3482,
      "mean_token_accuracy": 0.15165616124868392,
      "num_tokens": 16121050.0,
      "step": 6355
    },
    {
      "entropy": 6.471996784210205,
      "epoch": 0.7339873052510099,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004956672238931244,
      "loss": 6.3505,
      "mean_token_accuracy": 0.15684145987033843,
      "num_tokens": 16133603.0,
      "step": 6360
    },
    {
      "entropy": 6.300800943374634,
      "epoch": 0.7345643392960185,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004956591611853894,
      "loss": 6.2531,
      "mean_token_accuracy": 0.16184316128492354,
      "num_tokens": 16146797.0,
      "step": 6365
    },
    {
      "entropy": 6.460974407196045,
      "epoch": 0.7351413733410271,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004956510910558931,
      "loss": 6.4096,
      "mean_token_accuracy": 0.1571536049246788,
      "num_tokens": 16159223.0,
      "step": 6370
    },
    {
      "entropy": 6.41186637878418,
      "epoch": 0.7357184073860358,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004956430135049071,
      "loss": 6.3288,
      "mean_token_accuracy": 0.1522546589374542,
      "num_tokens": 16172305.0,
      "step": 6375
    },
    {
      "entropy": 6.442885541915894,
      "epoch": 0.7362954414310444,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004956349285327028,
      "loss": 6.3384,
      "mean_token_accuracy": 0.15485752522945403,
      "num_tokens": 16185575.0,
      "step": 6380
    },
    {
      "entropy": 6.453862142562866,
      "epoch": 0.7368724754760531,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004956268361395523,
      "loss": 6.2689,
      "mean_token_accuracy": 0.164891554415226,
      "num_tokens": 16198014.0,
      "step": 6385
    },
    {
      "entropy": 6.376306581497192,
      "epoch": 0.7374495095210617,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004956187363257278,
      "loss": 6.3041,
      "mean_token_accuracy": 0.15524324849247934,
      "num_tokens": 16211717.0,
      "step": 6390
    },
    {
      "entropy": 6.427841424942017,
      "epoch": 0.7380265435660704,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004956106290915017,
      "loss": 6.3495,
      "mean_token_accuracy": 0.16113682538270951,
      "num_tokens": 16225039.0,
      "step": 6395
    },
    {
      "entropy": 6.350244951248169,
      "epoch": 0.7386035776110791,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004956025144371467,
      "loss": 6.2859,
      "mean_token_accuracy": 0.15418801456689835,
      "num_tokens": 16236832.0,
      "step": 6400
    },
    {
      "entropy": 6.391508340835571,
      "epoch": 0.7391806116560877,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004955943923629356,
      "loss": 6.2895,
      "mean_token_accuracy": 0.1572732299566269,
      "num_tokens": 16250093.0,
      "step": 6405
    },
    {
      "entropy": 6.367796468734741,
      "epoch": 0.7397576457010964,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004955862628691417,
      "loss": 6.2809,
      "mean_token_accuracy": 0.15725187063217164,
      "num_tokens": 16263051.0,
      "step": 6410
    },
    {
      "entropy": 6.3537661075592045,
      "epoch": 0.740334679746105,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004955781259560386,
      "loss": 6.3209,
      "mean_token_accuracy": 0.15947764217853547,
      "num_tokens": 16276545.0,
      "step": 6415
    },
    {
      "entropy": 6.474180889129639,
      "epoch": 0.7409117137911136,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004955699816238995,
      "loss": 6.3305,
      "mean_token_accuracy": 0.16403343975543977,
      "num_tokens": 16288900.0,
      "step": 6420
    },
    {
      "entropy": 6.399309730529785,
      "epoch": 0.7414887478361223,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004955618298729988,
      "loss": 6.3204,
      "mean_token_accuracy": 0.1557662233710289,
      "num_tokens": 16302066.0,
      "step": 6425
    },
    {
      "entropy": 6.464933729171753,
      "epoch": 0.742065781881131,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004955536707036105,
      "loss": 6.4118,
      "mean_token_accuracy": 0.15210218578577042,
      "num_tokens": 16316255.0,
      "step": 6430
    },
    {
      "entropy": 6.431001615524292,
      "epoch": 0.7426428159261397,
      "grad_norm": 1.09375,
      "learning_rate": 0.000495545504116009,
      "loss": 6.3203,
      "mean_token_accuracy": 0.15718675553798675,
      "num_tokens": 16328419.0,
      "step": 6435
    },
    {
      "entropy": 6.334261083602906,
      "epoch": 0.7432198499711483,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004955373301104691,
      "loss": 6.2601,
      "mean_token_accuracy": 0.16228911131620408,
      "num_tokens": 16339864.0,
      "step": 6440
    },
    {
      "entropy": 6.3899516582489015,
      "epoch": 0.743796884016157,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004955291486872657,
      "loss": 6.345,
      "mean_token_accuracy": 0.1576900452375412,
      "num_tokens": 16354364.0,
      "step": 6445
    },
    {
      "entropy": 6.437332439422607,
      "epoch": 0.7443739180611656,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004955209598466738,
      "loss": 6.3167,
      "mean_token_accuracy": 0.15318332314491273,
      "num_tokens": 16366118.0,
      "step": 6450
    },
    {
      "entropy": 6.403056859970093,
      "epoch": 0.7449509521061742,
      "grad_norm": 1.25,
      "learning_rate": 0.000495512763588969,
      "loss": 6.3466,
      "mean_token_accuracy": 0.15583230555057526,
      "num_tokens": 16378805.0,
      "step": 6455
    },
    {
      "entropy": 6.443750190734863,
      "epoch": 0.7455279861511829,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004955045599144269,
      "loss": 6.309,
      "mean_token_accuracy": 0.1636185646057129,
      "num_tokens": 16391202.0,
      "step": 6460
    },
    {
      "entropy": 6.352822160720825,
      "epoch": 0.7461050201961916,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004954963488233235,
      "loss": 6.2521,
      "mean_token_accuracy": 0.15978027135133743,
      "num_tokens": 16402409.0,
      "step": 6465
    },
    {
      "entropy": 6.411400270462036,
      "epoch": 0.7466820542412003,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000495488130315935,
      "loss": 6.29,
      "mean_token_accuracy": 0.1578140839934349,
      "num_tokens": 16415710.0,
      "step": 6470
    },
    {
      "entropy": 6.373956632614136,
      "epoch": 0.7472590882862089,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004954799043925377,
      "loss": 6.2975,
      "mean_token_accuracy": 0.1564439430832863,
      "num_tokens": 16428135.0,
      "step": 6475
    },
    {
      "entropy": 6.388805627822876,
      "epoch": 0.7478361223312175,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004954716710534082,
      "loss": 6.3847,
      "mean_token_accuracy": 0.15414656400680543,
      "num_tokens": 16440052.0,
      "step": 6480
    },
    {
      "entropy": 6.375358581542969,
      "epoch": 0.7484131563762262,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004954634302988237,
      "loss": 6.3246,
      "mean_token_accuracy": 0.15686732158064842,
      "num_tokens": 16453188.0,
      "step": 6485
    },
    {
      "entropy": 6.4655996799469,
      "epoch": 0.7489901904212348,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004954551821290612,
      "loss": 6.3876,
      "mean_token_accuracy": 0.15350718349218367,
      "num_tokens": 16464946.0,
      "step": 6490
    },
    {
      "entropy": 6.44443941116333,
      "epoch": 0.7495672244662435,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004954469265443981,
      "loss": 6.3764,
      "mean_token_accuracy": 0.1545371264219284,
      "num_tokens": 16476957.0,
      "step": 6495
    },
    {
      "entropy": 6.394003057479859,
      "epoch": 0.7501442585112522,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004954386635451123,
      "loss": 6.2422,
      "mean_token_accuracy": 0.16569167524576187,
      "num_tokens": 16488847.0,
      "step": 6500
    },
    {
      "entropy": 6.391976833343506,
      "epoch": 0.7507212925562609,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004954303931314814,
      "loss": 6.3052,
      "mean_token_accuracy": 0.16465448588132858,
      "num_tokens": 16502497.0,
      "step": 6505
    },
    {
      "entropy": 6.390478944778442,
      "epoch": 0.7512983266012695,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004954221153037837,
      "loss": 6.3531,
      "mean_token_accuracy": 0.15633823424577714,
      "num_tokens": 16514799.0,
      "step": 6510
    },
    {
      "entropy": 6.431984615325928,
      "epoch": 0.7518753606462781,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004954138300622978,
      "loss": 6.3868,
      "mean_token_accuracy": 0.1524726554751396,
      "num_tokens": 16528051.0,
      "step": 6515
    },
    {
      "entropy": 6.464916610717774,
      "epoch": 0.7524523946912868,
      "grad_norm": 1.4375,
      "learning_rate": 0.000495405537407302,
      "loss": 6.3051,
      "mean_token_accuracy": 0.15390395522117614,
      "num_tokens": 16540127.0,
      "step": 6520
    },
    {
      "entropy": 6.366085958480835,
      "epoch": 0.7530294287362954,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004953972373390755,
      "loss": 6.2958,
      "mean_token_accuracy": 0.16691721826791764,
      "num_tokens": 16553419.0,
      "step": 6525
    },
    {
      "entropy": 6.424267721176148,
      "epoch": 0.753606462781304,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004953889298578975,
      "loss": 6.3736,
      "mean_token_accuracy": 0.16062992215156555,
      "num_tokens": 16567068.0,
      "step": 6530
    },
    {
      "entropy": 6.413002920150757,
      "epoch": 0.7541834968263128,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004953806149640473,
      "loss": 6.223,
      "mean_token_accuracy": 0.15969938933849334,
      "num_tokens": 16579516.0,
      "step": 6535
    },
    {
      "entropy": 6.40973916053772,
      "epoch": 0.7547605308713214,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004953722926578045,
      "loss": 6.361,
      "mean_token_accuracy": 0.1577667474746704,
      "num_tokens": 16592489.0,
      "step": 6540
    },
    {
      "entropy": 6.335503387451172,
      "epoch": 0.7553375649163301,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004953639629394492,
      "loss": 6.2838,
      "mean_token_accuracy": 0.1580977886915207,
      "num_tokens": 16604848.0,
      "step": 6545
    },
    {
      "entropy": 6.480976963043213,
      "epoch": 0.7559145989613387,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004953556258092613,
      "loss": 6.3215,
      "mean_token_accuracy": 0.15820895731449128,
      "num_tokens": 16617426.0,
      "step": 6550
    },
    {
      "entropy": 6.425509929656982,
      "epoch": 0.7564916330063474,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004953472812675216,
      "loss": 6.3331,
      "mean_token_accuracy": 0.1553524613380432,
      "num_tokens": 16629676.0,
      "step": 6555
    },
    {
      "entropy": 6.431772375106812,
      "epoch": 0.757068667051356,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004953389293145104,
      "loss": 6.366,
      "mean_token_accuracy": 0.14746260792016982,
      "num_tokens": 16642113.0,
      "step": 6560
    },
    {
      "entropy": 6.362658548355102,
      "epoch": 0.7576457010963646,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004953305699505088,
      "loss": 6.3007,
      "mean_token_accuracy": 0.15562691688537597,
      "num_tokens": 16653866.0,
      "step": 6565
    },
    {
      "entropy": 6.329480743408203,
      "epoch": 0.7582227351413734,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004953222031757979,
      "loss": 6.3209,
      "mean_token_accuracy": 0.15877506732940674,
      "num_tokens": 16666917.0,
      "step": 6570
    },
    {
      "entropy": 6.364426517486573,
      "epoch": 0.758799769186382,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004953138289906592,
      "loss": 6.2548,
      "mean_token_accuracy": 0.15817622244358062,
      "num_tokens": 16678752.0,
      "step": 6575
    },
    {
      "entropy": 6.409604454040528,
      "epoch": 0.7593768032313907,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004953054473953742,
      "loss": 6.2703,
      "mean_token_accuracy": 0.16493133902549745,
      "num_tokens": 16690942.0,
      "step": 6580
    },
    {
      "entropy": 6.376152753829956,
      "epoch": 0.7599538372763993,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004952970583902251,
      "loss": 6.2837,
      "mean_token_accuracy": 0.15724904239177703,
      "num_tokens": 16703009.0,
      "step": 6585
    },
    {
      "entropy": 6.361568355560303,
      "epoch": 0.760530871321408,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004952886619754937,
      "loss": 6.2608,
      "mean_token_accuracy": 0.1616979017853737,
      "num_tokens": 16714881.0,
      "step": 6590
    },
    {
      "entropy": 6.340665197372436,
      "epoch": 0.7611079053664166,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004952802581514625,
      "loss": 6.3314,
      "mean_token_accuracy": 0.15557588189840316,
      "num_tokens": 16729892.0,
      "step": 6595
    },
    {
      "entropy": 6.457778739929199,
      "epoch": 0.7616849394114252,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004952718469184144,
      "loss": 6.3139,
      "mean_token_accuracy": 0.157939513027668,
      "num_tokens": 16742974.0,
      "step": 6600
    },
    {
      "entropy": 6.387112045288086,
      "epoch": 0.762261973456434,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004952634282766322,
      "loss": 6.3682,
      "mean_token_accuracy": 0.1535436376929283,
      "num_tokens": 16756758.0,
      "step": 6605
    },
    {
      "entropy": 6.448778247833252,
      "epoch": 0.7628390075014426,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004952550022263988,
      "loss": 6.3539,
      "mean_token_accuracy": 0.15587857812643052,
      "num_tokens": 16768875.0,
      "step": 6610
    },
    {
      "entropy": 6.372657394409179,
      "epoch": 0.7634160415464513,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004952465687679979,
      "loss": 6.243,
      "mean_token_accuracy": 0.15659031867980958,
      "num_tokens": 16780526.0,
      "step": 6615
    },
    {
      "entropy": 6.418962192535401,
      "epoch": 0.7639930755914599,
      "grad_norm": 1.265625,
      "learning_rate": 0.000495238127901713,
      "loss": 6.264,
      "mean_token_accuracy": 0.16757059693336487,
      "num_tokens": 16793778.0,
      "step": 6620
    },
    {
      "entropy": 6.3180074214935305,
      "epoch": 0.7645701096364685,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004952296796278282,
      "loss": 6.2334,
      "mean_token_accuracy": 0.1606579229235649,
      "num_tokens": 16805367.0,
      "step": 6625
    },
    {
      "entropy": 6.378176689147949,
      "epoch": 0.7651471436814772,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004952212239466274,
      "loss": 6.2722,
      "mean_token_accuracy": 0.16312642097473146,
      "num_tokens": 16818298.0,
      "step": 6630
    },
    {
      "entropy": 6.405220556259155,
      "epoch": 0.7657241777264858,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004952127608583953,
      "loss": 6.3588,
      "mean_token_accuracy": 0.15494537502527236,
      "num_tokens": 16831031.0,
      "step": 6635
    },
    {
      "entropy": 6.379863309860229,
      "epoch": 0.7663012117714946,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004952042903634162,
      "loss": 6.3168,
      "mean_token_accuracy": 0.16437758803367614,
      "num_tokens": 16844143.0,
      "step": 6640
    },
    {
      "entropy": 6.356973361968994,
      "epoch": 0.7668782458165032,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004951958124619752,
      "loss": 6.2298,
      "mean_token_accuracy": 0.17034306824207307,
      "num_tokens": 16857639.0,
      "step": 6645
    },
    {
      "entropy": 6.360614061355591,
      "epoch": 0.7674552798615119,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004951873271543573,
      "loss": 6.2954,
      "mean_token_accuracy": 0.16176145970821382,
      "num_tokens": 16870479.0,
      "step": 6650
    },
    {
      "entropy": 6.442333984375,
      "epoch": 0.7680323139065205,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004951788344408483,
      "loss": 6.2956,
      "mean_token_accuracy": 0.16062431335449218,
      "num_tokens": 16883876.0,
      "step": 6655
    },
    {
      "entropy": 6.402741479873657,
      "epoch": 0.7686093479515291,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004951703343217335,
      "loss": 6.2827,
      "mean_token_accuracy": 0.1593102291226387,
      "num_tokens": 16895516.0,
      "step": 6660
    },
    {
      "entropy": 6.401149892807007,
      "epoch": 0.7691863819965378,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004951618267972987,
      "loss": 6.3177,
      "mean_token_accuracy": 0.16399899423122405,
      "num_tokens": 16906558.0,
      "step": 6665
    },
    {
      "entropy": 6.381659460067749,
      "epoch": 0.7697634160415464,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004951533118678304,
      "loss": 6.2912,
      "mean_token_accuracy": 0.16513835042715072,
      "num_tokens": 16919951.0,
      "step": 6670
    },
    {
      "entropy": 6.458241510391235,
      "epoch": 0.7703404500865552,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004951447895336147,
      "loss": 6.3214,
      "mean_token_accuracy": 0.1566588670015335,
      "num_tokens": 16932885.0,
      "step": 6675
    },
    {
      "entropy": 6.4028833389282225,
      "epoch": 0.7709174841315638,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004951362597949384,
      "loss": 6.3149,
      "mean_token_accuracy": 0.15482009053230286,
      "num_tokens": 16944769.0,
      "step": 6680
    },
    {
      "entropy": 6.375073099136353,
      "epoch": 0.7714945181765724,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004951277226520883,
      "loss": 6.3065,
      "mean_token_accuracy": 0.15622059404850006,
      "num_tokens": 16958652.0,
      "step": 6685
    },
    {
      "entropy": 6.357271957397461,
      "epoch": 0.7720715522215811,
      "grad_norm": 1.25,
      "learning_rate": 0.0004951191781053517,
      "loss": 6.1796,
      "mean_token_accuracy": 0.17576172947883606,
      "num_tokens": 16971850.0,
      "step": 6690
    },
    {
      "entropy": 6.379316711425782,
      "epoch": 0.7726485862665897,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004951106261550157,
      "loss": 6.3052,
      "mean_token_accuracy": 0.15468476861715316,
      "num_tokens": 16984164.0,
      "step": 6695
    },
    {
      "entropy": 6.406053733825684,
      "epoch": 0.7732256203115984,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004951020668013683,
      "loss": 6.4075,
      "mean_token_accuracy": 0.14664715826511382,
      "num_tokens": 16999140.0,
      "step": 6700
    },
    {
      "entropy": 6.42271933555603,
      "epoch": 0.773802654356607,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004950935000446972,
      "loss": 6.2839,
      "mean_token_accuracy": 0.1637265384197235,
      "num_tokens": 17012137.0,
      "step": 6705
    },
    {
      "entropy": 6.360930299758911,
      "epoch": 0.7743796884016156,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004950849258852905,
      "loss": 6.2402,
      "mean_token_accuracy": 0.16348935067653655,
      "num_tokens": 17024594.0,
      "step": 6710
    },
    {
      "entropy": 6.384844589233398,
      "epoch": 0.7749567224466244,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004950763443234366,
      "loss": 6.3589,
      "mean_token_accuracy": 0.1582717314362526,
      "num_tokens": 17036927.0,
      "step": 6715
    },
    {
      "entropy": 6.3573465824127195,
      "epoch": 0.775533756491633,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004950677553594243,
      "loss": 6.2769,
      "mean_token_accuracy": 0.16219737231731415,
      "num_tokens": 17049393.0,
      "step": 6720
    },
    {
      "entropy": 6.420328903198242,
      "epoch": 0.7761107905366417,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004950591589935422,
      "loss": 6.3368,
      "mean_token_accuracy": 0.15120098516345024,
      "num_tokens": 17063053.0,
      "step": 6725
    },
    {
      "entropy": 6.535327768325805,
      "epoch": 0.7766878245816503,
      "grad_norm": 3.125,
      "learning_rate": 0.0004950505552260798,
      "loss": 6.3339,
      "mean_token_accuracy": 0.1562727525830269,
      "num_tokens": 17075700.0,
      "step": 6730
    },
    {
      "entropy": 6.371122217178344,
      "epoch": 0.777264858626659,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004950419440573261,
      "loss": 6.2287,
      "mean_token_accuracy": 0.15924327373504638,
      "num_tokens": 17088089.0,
      "step": 6735
    },
    {
      "entropy": 6.3832090377807615,
      "epoch": 0.7778418926716676,
      "grad_norm": 1.140625,
      "learning_rate": 0.000495033325487571,
      "loss": 6.3173,
      "mean_token_accuracy": 0.1595449075102806,
      "num_tokens": 17100769.0,
      "step": 6740
    },
    {
      "entropy": 6.370034694671631,
      "epoch": 0.7784189267166762,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004950246995171042,
      "loss": 6.3434,
      "mean_token_accuracy": 0.1615496963262558,
      "num_tokens": 17113107.0,
      "step": 6745
    },
    {
      "entropy": 6.361813163757324,
      "epoch": 0.778995960761685,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000495016066146216,
      "loss": 6.2944,
      "mean_token_accuracy": 0.16585608720779418,
      "num_tokens": 17125856.0,
      "step": 6750
    },
    {
      "entropy": 6.433376502990723,
      "epoch": 0.7795729948066936,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004950074253751968,
      "loss": 6.2896,
      "mean_token_accuracy": 0.16024880707263947,
      "num_tokens": 17139931.0,
      "step": 6755
    },
    {
      "entropy": 6.341174221038818,
      "epoch": 0.7801500288517023,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000494998777204337,
      "loss": 6.2373,
      "mean_token_accuracy": 0.1614516481757164,
      "num_tokens": 17151669.0,
      "step": 6760
    },
    {
      "entropy": 6.354077243804932,
      "epoch": 0.7807270628967109,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004949901216339276,
      "loss": 6.2524,
      "mean_token_accuracy": 0.16256770193576814,
      "num_tokens": 17164434.0,
      "step": 6765
    },
    {
      "entropy": 6.35262656211853,
      "epoch": 0.7813040969417195,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004949814586642598,
      "loss": 6.2998,
      "mean_token_accuracy": 0.159908227622509,
      "num_tokens": 17176732.0,
      "step": 6770
    },
    {
      "entropy": 6.37272424697876,
      "epoch": 0.7818811309867282,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000494972788295625,
      "loss": 6.2037,
      "mean_token_accuracy": 0.1622636005282402,
      "num_tokens": 17188698.0,
      "step": 6775
    },
    {
      "entropy": 6.355993938446045,
      "epoch": 0.7824581650317368,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004949641105283144,
      "loss": 6.2454,
      "mean_token_accuracy": 0.16349631547927856,
      "num_tokens": 17201219.0,
      "step": 6780
    },
    {
      "entropy": 6.437637662887573,
      "epoch": 0.7830351990767456,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004949554253626205,
      "loss": 6.3504,
      "mean_token_accuracy": 0.1554308459162712,
      "num_tokens": 17213318.0,
      "step": 6785
    },
    {
      "entropy": 6.387994813919067,
      "epoch": 0.7836122331217542,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004949467327988351,
      "loss": 6.2775,
      "mean_token_accuracy": 0.1567632406949997,
      "num_tokens": 17226080.0,
      "step": 6790
    },
    {
      "entropy": 6.3465128421783445,
      "epoch": 0.7841892671667628,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004949380328372505,
      "loss": 6.3002,
      "mean_token_accuracy": 0.15576818138360976,
      "num_tokens": 17237840.0,
      "step": 6795
    },
    {
      "entropy": 6.422131013870239,
      "epoch": 0.7847663012117715,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004949293254781595,
      "loss": 6.2205,
      "mean_token_accuracy": 0.1624345824122429,
      "num_tokens": 17249972.0,
      "step": 6800
    },
    {
      "entropy": 6.4173565864562985,
      "epoch": 0.7853433352567801,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004949206107218547,
      "loss": 6.294,
      "mean_token_accuracy": 0.16312530189752578,
      "num_tokens": 17261605.0,
      "step": 6805
    },
    {
      "entropy": 6.336081600189209,
      "epoch": 0.7859203693017888,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004949118885686296,
      "loss": 6.2349,
      "mean_token_accuracy": 0.1613142505288124,
      "num_tokens": 17273992.0,
      "step": 6810
    },
    {
      "entropy": 6.471020841598511,
      "epoch": 0.7864974033467974,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004949031590187774,
      "loss": 6.3011,
      "mean_token_accuracy": 0.16081756204366685,
      "num_tokens": 17286978.0,
      "step": 6815
    },
    {
      "entropy": 6.296640920639038,
      "epoch": 0.7870744373918062,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004948944220725915,
      "loss": 6.3009,
      "mean_token_accuracy": 0.16438583433628082,
      "num_tokens": 17298998.0,
      "step": 6820
    },
    {
      "entropy": 6.386342334747314,
      "epoch": 0.7876514714368148,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000494885677730366,
      "loss": 6.3079,
      "mean_token_accuracy": 0.16172787845134734,
      "num_tokens": 17311614.0,
      "step": 6825
    },
    {
      "entropy": 6.403314208984375,
      "epoch": 0.7882285054818234,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000494876925992395,
      "loss": 6.3093,
      "mean_token_accuracy": 0.15615005046129227,
      "num_tokens": 17323603.0,
      "step": 6830
    },
    {
      "entropy": 6.383903455734253,
      "epoch": 0.7888055395268321,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004948681668589728,
      "loss": 6.2719,
      "mean_token_accuracy": 0.16416805535554885,
      "num_tokens": 17334916.0,
      "step": 6835
    },
    {
      "entropy": 6.286671304702759,
      "epoch": 0.7893825735718407,
      "grad_norm": 1.0625,
      "learning_rate": 0.000494859400330394,
      "loss": 6.2988,
      "mean_token_accuracy": 0.1619895502924919,
      "num_tokens": 17346777.0,
      "step": 6840
    },
    {
      "entropy": 6.476213169097901,
      "epoch": 0.7899596076168494,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004948506264069535,
      "loss": 6.3201,
      "mean_token_accuracy": 0.15922749489545823,
      "num_tokens": 17359144.0,
      "step": 6845
    },
    {
      "entropy": 6.480512714385986,
      "epoch": 0.790536641661858,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004948418450889464,
      "loss": 6.3522,
      "mean_token_accuracy": 0.15333817154169083,
      "num_tokens": 17372992.0,
      "step": 6850
    },
    {
      "entropy": 6.244059085845947,
      "epoch": 0.7911136757068667,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000494833056376668,
      "loss": 6.1968,
      "mean_token_accuracy": 0.16934754550457,
      "num_tokens": 17385413.0,
      "step": 6855
    },
    {
      "entropy": 6.417005968093872,
      "epoch": 0.7916907097518754,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004948242602704139,
      "loss": 6.3337,
      "mean_token_accuracy": 0.16410370767116547,
      "num_tokens": 17397882.0,
      "step": 6860
    },
    {
      "entropy": 6.330115127563476,
      "epoch": 0.792267743796884,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004948154567704801,
      "loss": 6.2257,
      "mean_token_accuracy": 0.16090456694364547,
      "num_tokens": 17410662.0,
      "step": 6865
    },
    {
      "entropy": 6.369202423095703,
      "epoch": 0.7928447778418927,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004948066458771625,
      "loss": 6.3151,
      "mean_token_accuracy": 0.15470239520072937,
      "num_tokens": 17423301.0,
      "step": 6870
    },
    {
      "entropy": 6.423096418380737,
      "epoch": 0.7934218118869013,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004947978275907577,
      "loss": 6.2818,
      "mean_token_accuracy": 0.1664452314376831,
      "num_tokens": 17435733.0,
      "step": 6875
    },
    {
      "entropy": 6.433144426345825,
      "epoch": 0.79399884593191,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000494789001911562,
      "loss": 6.3624,
      "mean_token_accuracy": 0.15425862669944762,
      "num_tokens": 17448516.0,
      "step": 6880
    },
    {
      "entropy": 6.335712432861328,
      "epoch": 0.7945758799769186,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004947801688398725,
      "loss": 6.2691,
      "mean_token_accuracy": 0.15862028300762177,
      "num_tokens": 17462199.0,
      "step": 6885
    },
    {
      "entropy": 6.494321823120117,
      "epoch": 0.7951529140219273,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004947713283759862,
      "loss": 6.3377,
      "mean_token_accuracy": 0.15648676753044127,
      "num_tokens": 17475651.0,
      "step": 6890
    },
    {
      "entropy": 6.4883825302124025,
      "epoch": 0.795729948066936,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004947624805202003,
      "loss": 6.363,
      "mean_token_accuracy": 0.14902667999267577,
      "num_tokens": 17488264.0,
      "step": 6895
    },
    {
      "entropy": 6.401565933227539,
      "epoch": 0.7963069821119446,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004947536252728126,
      "loss": 6.3087,
      "mean_token_accuracy": 0.15726197510957718,
      "num_tokens": 17500691.0,
      "step": 6900
    },
    {
      "entropy": 6.34444670677185,
      "epoch": 0.7968840161569533,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004947447626341209,
      "loss": 6.3073,
      "mean_token_accuracy": 0.1573370113968849,
      "num_tokens": 17513957.0,
      "step": 6905
    },
    {
      "entropy": 6.509504508972168,
      "epoch": 0.7974610502019619,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004947358926044232,
      "loss": 6.3302,
      "mean_token_accuracy": 0.15750703364610671,
      "num_tokens": 17527057.0,
      "step": 6910
    },
    {
      "entropy": 6.380242729187012,
      "epoch": 0.7980380842469705,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004947270151840179,
      "loss": 6.3237,
      "mean_token_accuracy": 0.14655912965536116,
      "num_tokens": 17541215.0,
      "step": 6915
    },
    {
      "entropy": 6.381529903411865,
      "epoch": 0.7986151182919792,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004947181303732038,
      "loss": 6.312,
      "mean_token_accuracy": 0.15930712819099427,
      "num_tokens": 17553116.0,
      "step": 6920
    },
    {
      "entropy": 6.406609773635864,
      "epoch": 0.7991921523369879,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004947092381722793,
      "loss": 6.2427,
      "mean_token_accuracy": 0.16228737235069274,
      "num_tokens": 17565480.0,
      "step": 6925
    },
    {
      "entropy": 6.293425607681274,
      "epoch": 0.7997691863819966,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004947003385815438,
      "loss": 6.2973,
      "mean_token_accuracy": 0.16063206940889357,
      "num_tokens": 17577706.0,
      "step": 6930
    },
    {
      "entropy": 6.486158084869385,
      "epoch": 0.8003462204270052,
      "grad_norm": 1.125,
      "learning_rate": 0.0004946914316012964,
      "loss": 6.3487,
      "mean_token_accuracy": 0.15436617881059647,
      "num_tokens": 17591469.0,
      "step": 6935
    },
    {
      "entropy": 6.467647933959961,
      "epoch": 0.8009232544720138,
      "grad_norm": 1.09375,
      "learning_rate": 0.000494682517231837,
      "loss": 6.2669,
      "mean_token_accuracy": 0.163911335170269,
      "num_tokens": 17605225.0,
      "step": 6940
    },
    {
      "entropy": 6.277217245101928,
      "epoch": 0.8015002885170225,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004946735954734652,
      "loss": 6.1701,
      "mean_token_accuracy": 0.16724748462438582,
      "num_tokens": 17618509.0,
      "step": 6945
    },
    {
      "entropy": 6.408233976364135,
      "epoch": 0.8020773225620311,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004946646663264811,
      "loss": 6.3476,
      "mean_token_accuracy": 0.1609940305352211,
      "num_tokens": 17630652.0,
      "step": 6950
    },
    {
      "entropy": 6.401758766174316,
      "epoch": 0.8026543566070398,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004946557297911852,
      "loss": 6.1889,
      "mean_token_accuracy": 0.16586418002843856,
      "num_tokens": 17641935.0,
      "step": 6955
    },
    {
      "entropy": 6.388518810272217,
      "epoch": 0.8032313906520485,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004946467858678777,
      "loss": 6.4026,
      "mean_token_accuracy": 0.15096247494220733,
      "num_tokens": 17654043.0,
      "step": 6960
    },
    {
      "entropy": 6.485033464431763,
      "epoch": 0.8038084246970572,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004946378345568597,
      "loss": 6.3237,
      "mean_token_accuracy": 0.15735958218574525,
      "num_tokens": 17666566.0,
      "step": 6965
    },
    {
      "entropy": 6.272105741500854,
      "epoch": 0.8043854587420658,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004946288758584324,
      "loss": 6.1533,
      "mean_token_accuracy": 0.16458268761634826,
      "num_tokens": 17678463.0,
      "step": 6970
    },
    {
      "entropy": 6.44692029953003,
      "epoch": 0.8049624927870744,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004946199097728968,
      "loss": 6.3761,
      "mean_token_accuracy": 0.1529249370098114,
      "num_tokens": 17691663.0,
      "step": 6975
    },
    {
      "entropy": 6.474437618255616,
      "epoch": 0.8055395268320831,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004946109363005548,
      "loss": 6.3591,
      "mean_token_accuracy": 0.15547811090946198,
      "num_tokens": 17705048.0,
      "step": 6980
    },
    {
      "entropy": 6.387838554382324,
      "epoch": 0.8061165608770917,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000494601955441708,
      "loss": 6.3193,
      "mean_token_accuracy": 0.1601235330104828,
      "num_tokens": 17717865.0,
      "step": 6985
    },
    {
      "entropy": 6.4041938304901125,
      "epoch": 0.8066935949221004,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004945929671966585,
      "loss": 6.3548,
      "mean_token_accuracy": 0.15371118485927582,
      "num_tokens": 17730568.0,
      "step": 6990
    },
    {
      "entropy": 6.353836345672607,
      "epoch": 0.8072706289671091,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004945839715657085,
      "loss": 6.3052,
      "mean_token_accuracy": 0.15674142241477967,
      "num_tokens": 17743303.0,
      "step": 6995
    },
    {
      "entropy": 6.485065174102783,
      "epoch": 0.8078476630121177,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004945749685491607,
      "loss": 6.3725,
      "mean_token_accuracy": 0.15282811969518661,
      "num_tokens": 17756552.0,
      "step": 7000
    },
    {
      "entropy": 6.365577363967896,
      "epoch": 0.8084246970571264,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004945659581473181,
      "loss": 6.3927,
      "mean_token_accuracy": 0.15238141268491745,
      "num_tokens": 17770783.0,
      "step": 7005
    },
    {
      "entropy": 6.422938871383667,
      "epoch": 0.809001731102135,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004945569403604833,
      "loss": 6.2726,
      "mean_token_accuracy": 0.15890175998210906,
      "num_tokens": 17784573.0,
      "step": 7010
    },
    {
      "entropy": 6.467655897140503,
      "epoch": 0.8095787651471437,
      "grad_norm": 1.375,
      "learning_rate": 0.0004945479151889601,
      "loss": 6.3096,
      "mean_token_accuracy": 0.1511345088481903,
      "num_tokens": 17797304.0,
      "step": 7015
    },
    {
      "entropy": 6.31410665512085,
      "epoch": 0.8101557991921523,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004945388826330517,
      "loss": 6.1673,
      "mean_token_accuracy": 0.16778973639011383,
      "num_tokens": 17810401.0,
      "step": 7020
    },
    {
      "entropy": 6.310051012039184,
      "epoch": 0.8107328332371609,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004945298426930621,
      "loss": 6.2426,
      "mean_token_accuracy": 0.16368699818849564,
      "num_tokens": 17824120.0,
      "step": 7025
    },
    {
      "entropy": 6.389438819885254,
      "epoch": 0.8113098672821697,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004945207953692952,
      "loss": 6.3004,
      "mean_token_accuracy": 0.1567479759454727,
      "num_tokens": 17836865.0,
      "step": 7030
    },
    {
      "entropy": 6.340084505081177,
      "epoch": 0.8118869013271783,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004945117406620556,
      "loss": 6.325,
      "mean_token_accuracy": 0.15732779800891877,
      "num_tokens": 17849128.0,
      "step": 7035
    },
    {
      "entropy": 6.435843896865845,
      "epoch": 0.812463935372187,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004945026785716474,
      "loss": 6.2961,
      "mean_token_accuracy": 0.160021048784256,
      "num_tokens": 17861151.0,
      "step": 7040
    },
    {
      "entropy": 6.386323118209839,
      "epoch": 0.8130409694171956,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004944936090983757,
      "loss": 6.2636,
      "mean_token_accuracy": 0.16412176191806793,
      "num_tokens": 17874043.0,
      "step": 7045
    },
    {
      "entropy": 6.359679889678955,
      "epoch": 0.8136180034622043,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004944845322425455,
      "loss": 6.2905,
      "mean_token_accuracy": 0.15718530863523483,
      "num_tokens": 17887176.0,
      "step": 7050
    },
    {
      "entropy": 6.458650541305542,
      "epoch": 0.8141950375072129,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004944754480044621,
      "loss": 6.3397,
      "mean_token_accuracy": 0.15000032186508178,
      "num_tokens": 17899708.0,
      "step": 7055
    },
    {
      "entropy": 6.402772617340088,
      "epoch": 0.8147720715522215,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004944663563844311,
      "loss": 6.249,
      "mean_token_accuracy": 0.16167229413986206,
      "num_tokens": 17912303.0,
      "step": 7060
    },
    {
      "entropy": 6.393413019180298,
      "epoch": 0.8153491055972303,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004944572573827581,
      "loss": 6.3756,
      "mean_token_accuracy": 0.1472133457660675,
      "num_tokens": 17925963.0,
      "step": 7065
    },
    {
      "entropy": 6.396823883056641,
      "epoch": 0.8159261396422389,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004944481509997494,
      "loss": 6.3547,
      "mean_token_accuracy": 0.15728743374347687,
      "num_tokens": 17938786.0,
      "step": 7070
    },
    {
      "entropy": 6.489611387252808,
      "epoch": 0.8165031736872476,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000494439037235711,
      "loss": 6.2669,
      "mean_token_accuracy": 0.16249716579914092,
      "num_tokens": 17950646.0,
      "step": 7075
    },
    {
      "entropy": 6.4137121677398685,
      "epoch": 0.8170802077322562,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004944299160909495,
      "loss": 6.3163,
      "mean_token_accuracy": 0.155899378657341,
      "num_tokens": 17963115.0,
      "step": 7080
    },
    {
      "entropy": 6.329668474197388,
      "epoch": 0.8176572417772648,
      "grad_norm": 2.28125,
      "learning_rate": 0.000494420787565772,
      "loss": 6.3135,
      "mean_token_accuracy": 0.1589229092001915,
      "num_tokens": 17976068.0,
      "step": 7085
    },
    {
      "entropy": 6.411213445663452,
      "epoch": 0.8182342758222735,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004944116516604852,
      "loss": 6.2522,
      "mean_token_accuracy": 0.16533806920051575,
      "num_tokens": 17988487.0,
      "step": 7090
    },
    {
      "entropy": 6.363603591918945,
      "epoch": 0.8188113098672821,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004944025083753965,
      "loss": 6.309,
      "mean_token_accuracy": 0.1558293379843235,
      "num_tokens": 18001664.0,
      "step": 7095
    },
    {
      "entropy": 6.397566413879394,
      "epoch": 0.8193883439122909,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004943933577108133,
      "loss": 6.327,
      "mean_token_accuracy": 0.15606159269809722,
      "num_tokens": 18014384.0,
      "step": 7100
    },
    {
      "entropy": 6.403047323226929,
      "epoch": 0.8199653779572995,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004943841996670436,
      "loss": 6.2777,
      "mean_token_accuracy": 0.15832700282335282,
      "num_tokens": 18027516.0,
      "step": 7105
    },
    {
      "entropy": 6.362675762176513,
      "epoch": 0.8205424120023082,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004943750342443953,
      "loss": 6.1968,
      "mean_token_accuracy": 0.1685020849108696,
      "num_tokens": 18039771.0,
      "step": 7110
    },
    {
      "entropy": 6.356399583816528,
      "epoch": 0.8211194460473168,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004943658614431767,
      "loss": 6.3739,
      "mean_token_accuracy": 0.1547293670475483,
      "num_tokens": 18052620.0,
      "step": 7115
    },
    {
      "entropy": 6.317503356933594,
      "epoch": 0.8216964800923254,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004943566812636963,
      "loss": 6.1187,
      "mean_token_accuracy": 0.17016562521457673,
      "num_tokens": 18065723.0,
      "step": 7120
    },
    {
      "entropy": 6.375060653686523,
      "epoch": 0.8222735141373341,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000494347493706263,
      "loss": 6.2059,
      "mean_token_accuracy": 0.16255648285150529,
      "num_tokens": 18078335.0,
      "step": 7125
    },
    {
      "entropy": 6.301021528244019,
      "epoch": 0.8228505481823427,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004943382987711856,
      "loss": 6.2426,
      "mean_token_accuracy": 0.16367732137441635,
      "num_tokens": 18089943.0,
      "step": 7130
    },
    {
      "entropy": 6.3740767478942875,
      "epoch": 0.8234275822273515,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004943290964587734,
      "loss": 6.2212,
      "mean_token_accuracy": 0.15678982436656952,
      "num_tokens": 18101843.0,
      "step": 7135
    },
    {
      "entropy": 6.2509630680084225,
      "epoch": 0.8240046162723601,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004943198867693361,
      "loss": 6.1034,
      "mean_token_accuracy": 0.17667673975229264,
      "num_tokens": 18113170.0,
      "step": 7140
    },
    {
      "entropy": 6.34347767829895,
      "epoch": 0.8245816503173687,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004943106697031833,
      "loss": 6.3042,
      "mean_token_accuracy": 0.1592097744345665,
      "num_tokens": 18124936.0,
      "step": 7145
    },
    {
      "entropy": 6.348465394973755,
      "epoch": 0.8251586843623774,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004943014452606251,
      "loss": 6.2356,
      "mean_token_accuracy": 0.15629333555698394,
      "num_tokens": 18137894.0,
      "step": 7150
    },
    {
      "entropy": 6.3925800800323485,
      "epoch": 0.825735718407386,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004942922134419717,
      "loss": 6.3089,
      "mean_token_accuracy": 0.15640681982040405,
      "num_tokens": 18149170.0,
      "step": 7155
    },
    {
      "entropy": 6.375568866729736,
      "epoch": 0.8263127524523947,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004942829742475336,
      "loss": 6.3178,
      "mean_token_accuracy": 0.15552882552146913,
      "num_tokens": 18160694.0,
      "step": 7160
    },
    {
      "entropy": 6.443007850646973,
      "epoch": 0.8268897864974033,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004942737276776217,
      "loss": 6.43,
      "mean_token_accuracy": 0.1548667162656784,
      "num_tokens": 18173619.0,
      "step": 7165
    },
    {
      "entropy": 6.475993967056274,
      "epoch": 0.827466820542412,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004942644737325468,
      "loss": 6.385,
      "mean_token_accuracy": 0.15922041982412338,
      "num_tokens": 18186751.0,
      "step": 7170
    },
    {
      "entropy": 6.323366594314575,
      "epoch": 0.8280438545874207,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004942552124126202,
      "loss": 6.2678,
      "mean_token_accuracy": 0.16112520396709443,
      "num_tokens": 18198701.0,
      "step": 7175
    },
    {
      "entropy": 6.404819440841675,
      "epoch": 0.8286208886324293,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004942459437181535,
      "loss": 6.2347,
      "mean_token_accuracy": 0.16615718603134155,
      "num_tokens": 18211422.0,
      "step": 7180
    },
    {
      "entropy": 6.363730382919312,
      "epoch": 0.829197922677438,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004942366676494584,
      "loss": 6.3052,
      "mean_token_accuracy": 0.15867968648672104,
      "num_tokens": 18223427.0,
      "step": 7185
    },
    {
      "entropy": 6.3306841373443605,
      "epoch": 0.8297749567224466,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004942273842068469,
      "loss": 6.2706,
      "mean_token_accuracy": 0.16639561653137208,
      "num_tokens": 18235234.0,
      "step": 7190
    },
    {
      "entropy": 6.392367649078369,
      "epoch": 0.8303519907674553,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004942180933906311,
      "loss": 6.2083,
      "mean_token_accuracy": 0.16739328056573868,
      "num_tokens": 18247669.0,
      "step": 7195
    },
    {
      "entropy": 6.2721107482910154,
      "epoch": 0.8309290248124639,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004942087952011237,
      "loss": 6.1831,
      "mean_token_accuracy": 0.16879565715789796,
      "num_tokens": 18260987.0,
      "step": 7200
    },
    {
      "entropy": 6.351057481765747,
      "epoch": 0.8315060588574726,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004941994896386374,
      "loss": 6.3138,
      "mean_token_accuracy": 0.14855986833572388,
      "num_tokens": 18274805.0,
      "step": 7205
    },
    {
      "entropy": 6.455946922302246,
      "epoch": 0.8320830929024813,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004941901767034851,
      "loss": 6.2897,
      "mean_token_accuracy": 0.15959215462207793,
      "num_tokens": 18286438.0,
      "step": 7210
    },
    {
      "entropy": 6.420863437652588,
      "epoch": 0.8326601269474899,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004941808563959802,
      "loss": 6.307,
      "mean_token_accuracy": 0.1556537330150604,
      "num_tokens": 18299310.0,
      "step": 7215
    },
    {
      "entropy": 6.298717403411866,
      "epoch": 0.8332371609924986,
      "grad_norm": 1.125,
      "learning_rate": 0.0004941715287164359,
      "loss": 6.2355,
      "mean_token_accuracy": 0.15646998286247255,
      "num_tokens": 18311286.0,
      "step": 7220
    },
    {
      "entropy": 6.393199682235718,
      "epoch": 0.8338141950375072,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004941621936651661,
      "loss": 6.2977,
      "mean_token_accuracy": 0.160914871096611,
      "num_tokens": 18325759.0,
      "step": 7225
    },
    {
      "entropy": 6.3602536678314205,
      "epoch": 0.8343912290825158,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004941528512424849,
      "loss": 6.2431,
      "mean_token_accuracy": 0.16606508940458298,
      "num_tokens": 18338907.0,
      "step": 7230
    },
    {
      "entropy": 6.397316598892212,
      "epoch": 0.8349682631275245,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004941435014487064,
      "loss": 6.3076,
      "mean_token_accuracy": 0.1613747224211693,
      "num_tokens": 18351760.0,
      "step": 7235
    },
    {
      "entropy": 6.358635520935058,
      "epoch": 0.8355452971725332,
      "grad_norm": 1.078125,
      "learning_rate": 0.000494134144284145,
      "loss": 6.2839,
      "mean_token_accuracy": 0.15316848158836366,
      "num_tokens": 18363348.0,
      "step": 7240
    },
    {
      "entropy": 6.391732501983642,
      "epoch": 0.8361223312175419,
      "grad_norm": 1.125,
      "learning_rate": 0.0004941247797491156,
      "loss": 6.3251,
      "mean_token_accuracy": 0.15529790818691253,
      "num_tokens": 18375233.0,
      "step": 7245
    },
    {
      "entropy": 6.438053464889526,
      "epoch": 0.8366993652625505,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004941154078439329,
      "loss": 6.2465,
      "mean_token_accuracy": 0.1566495940089226,
      "num_tokens": 18388001.0,
      "step": 7250
    },
    {
      "entropy": 6.263370561599731,
      "epoch": 0.8372763993075591,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004941060285689126,
      "loss": 6.1174,
      "mean_token_accuracy": 0.1767331138253212,
      "num_tokens": 18399647.0,
      "step": 7255
    },
    {
      "entropy": 6.269815683364868,
      "epoch": 0.8378534333525678,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004940966419243695,
      "loss": 6.2706,
      "mean_token_accuracy": 0.16255548298358918,
      "num_tokens": 18412441.0,
      "step": 7260
    },
    {
      "entropy": 6.3893648147583,
      "epoch": 0.8384304673975764,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00049408724791062,
      "loss": 6.29,
      "mean_token_accuracy": 0.16324699223041533,
      "num_tokens": 18425232.0,
      "step": 7265
    },
    {
      "entropy": 6.459998893737793,
      "epoch": 0.8390075014425851,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004940778465279797,
      "loss": 6.2799,
      "mean_token_accuracy": 0.1541198819875717,
      "num_tokens": 18438148.0,
      "step": 7270
    },
    {
      "entropy": 6.2868123054504395,
      "epoch": 0.8395845354875938,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004940684377767647,
      "loss": 6.123,
      "mean_token_accuracy": 0.16577878296375276,
      "num_tokens": 18449955.0,
      "step": 7275
    },
    {
      "entropy": 6.300487852096557,
      "epoch": 0.8401615695326025,
      "grad_norm": 1.125,
      "learning_rate": 0.0004940590216572916,
      "loss": 6.2357,
      "mean_token_accuracy": 0.16564981490373612,
      "num_tokens": 18463125.0,
      "step": 7280
    },
    {
      "entropy": 6.4203471660614015,
      "epoch": 0.8407386035776111,
      "grad_norm": 1.125,
      "learning_rate": 0.0004940495981698772,
      "loss": 6.3227,
      "mean_token_accuracy": 0.15704980492591858,
      "num_tokens": 18475280.0,
      "step": 7285
    },
    {
      "entropy": 6.409898471832276,
      "epoch": 0.8413156376226197,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004940401673148384,
      "loss": 6.2833,
      "mean_token_accuracy": 0.15684410482645034,
      "num_tokens": 18487406.0,
      "step": 7290
    },
    {
      "entropy": 6.34295072555542,
      "epoch": 0.8418926716676284,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004940307290924923,
      "loss": 6.2256,
      "mean_token_accuracy": 0.16209163069725036,
      "num_tokens": 18500288.0,
      "step": 7295
    },
    {
      "entropy": 6.375508832931518,
      "epoch": 0.842469705712637,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004940212835031565,
      "loss": 6.3145,
      "mean_token_accuracy": 0.15740822851657868,
      "num_tokens": 18514254.0,
      "step": 7300
    },
    {
      "entropy": 6.40157299041748,
      "epoch": 0.8430467397576457,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004940118305471486,
      "loss": 6.3565,
      "mean_token_accuracy": 0.1491801179945469,
      "num_tokens": 18526536.0,
      "step": 7305
    },
    {
      "entropy": 6.41963005065918,
      "epoch": 0.8436237738026544,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004940023702247866,
      "loss": 6.252,
      "mean_token_accuracy": 0.1566472515463829,
      "num_tokens": 18540137.0,
      "step": 7310
    },
    {
      "entropy": 6.396530866622925,
      "epoch": 0.844200807847663,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004939929025363886,
      "loss": 6.2589,
      "mean_token_accuracy": 0.16515292823314667,
      "num_tokens": 18552975.0,
      "step": 7315
    },
    {
      "entropy": 6.235803699493408,
      "epoch": 0.8447778418926717,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004939834274822731,
      "loss": 6.1897,
      "mean_token_accuracy": 0.16500602960586547,
      "num_tokens": 18564653.0,
      "step": 7320
    },
    {
      "entropy": 6.427172470092773,
      "epoch": 0.8453548759376803,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004939739450627588,
      "loss": 6.3164,
      "mean_token_accuracy": 0.16018296778202057,
      "num_tokens": 18577226.0,
      "step": 7325
    },
    {
      "entropy": 6.380255937576294,
      "epoch": 0.845931909982689,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004939644552781647,
      "loss": 6.2441,
      "mean_token_accuracy": 0.15562110543251037,
      "num_tokens": 18591206.0,
      "step": 7330
    },
    {
      "entropy": 6.31131796836853,
      "epoch": 0.8465089440276976,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004939549581288099,
      "loss": 6.259,
      "mean_token_accuracy": 0.1633705511689186,
      "num_tokens": 18604182.0,
      "step": 7335
    },
    {
      "entropy": 6.37572751045227,
      "epoch": 0.8470859780727062,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004939454536150138,
      "loss": 6.3373,
      "mean_token_accuracy": 0.15385680496692658,
      "num_tokens": 18615830.0,
      "step": 7340
    },
    {
      "entropy": 6.348571872711181,
      "epoch": 0.847663012117715,
      "grad_norm": 1.015625,
      "learning_rate": 0.000493935941737096,
      "loss": 6.3064,
      "mean_token_accuracy": 0.15477886348962783,
      "num_tokens": 18628372.0,
      "step": 7345
    },
    {
      "entropy": 6.39981780052185,
      "epoch": 0.8482400461627236,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004939264224953768,
      "loss": 6.3497,
      "mean_token_accuracy": 0.150855715572834,
      "num_tokens": 18641470.0,
      "step": 7350
    },
    {
      "entropy": 6.458249425888061,
      "epoch": 0.8488170802077323,
      "grad_norm": 1.0625,
      "learning_rate": 0.000493916895890176,
      "loss": 6.218,
      "mean_token_accuracy": 0.16879717856645585,
      "num_tokens": 18654660.0,
      "step": 7355
    },
    {
      "entropy": 6.2829173564910885,
      "epoch": 0.8493941142527409,
      "grad_norm": 1.125,
      "learning_rate": 0.000493907361921814,
      "loss": 6.3111,
      "mean_token_accuracy": 0.15594610571861267,
      "num_tokens": 18665626.0,
      "step": 7360
    },
    {
      "entropy": 6.315751218795777,
      "epoch": 0.8499711482977496,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004938978205906118,
      "loss": 6.2266,
      "mean_token_accuracy": 0.16837245374917983,
      "num_tokens": 18677175.0,
      "step": 7365
    },
    {
      "entropy": 6.42999153137207,
      "epoch": 0.8505481823427582,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004938882718968901,
      "loss": 6.2521,
      "mean_token_accuracy": 0.1613793730735779,
      "num_tokens": 18689005.0,
      "step": 7370
    },
    {
      "entropy": 6.358723068237305,
      "epoch": 0.8511252163877668,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004938787158409702,
      "loss": 6.2889,
      "mean_token_accuracy": 0.16648382395505906,
      "num_tokens": 18702085.0,
      "step": 7375
    },
    {
      "entropy": 6.335157108306885,
      "epoch": 0.8517022504327756,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004938691524231733,
      "loss": 6.3042,
      "mean_token_accuracy": 0.15819645375013353,
      "num_tokens": 18715064.0,
      "step": 7380
    },
    {
      "entropy": 6.4331118106842045,
      "epoch": 0.8522792844777842,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004938595816438212,
      "loss": 6.3053,
      "mean_token_accuracy": 0.15899132192134857,
      "num_tokens": 18728105.0,
      "step": 7385
    },
    {
      "entropy": 6.367902612686157,
      "epoch": 0.8528563185227929,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004938500035032358,
      "loss": 6.2359,
      "mean_token_accuracy": 0.16572393774986266,
      "num_tokens": 18741021.0,
      "step": 7390
    },
    {
      "entropy": 6.2689062595367435,
      "epoch": 0.8534333525678015,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004938404180017392,
      "loss": 6.2936,
      "mean_token_accuracy": 0.17114799171686174,
      "num_tokens": 18753271.0,
      "step": 7395
    },
    {
      "entropy": 6.341451978683471,
      "epoch": 0.8540103866128101,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004938308251396539,
      "loss": 6.2609,
      "mean_token_accuracy": 0.15689549297094346,
      "num_tokens": 18765631.0,
      "step": 7400
    },
    {
      "entropy": 6.388570833206177,
      "epoch": 0.8545874206578188,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004938212249173024,
      "loss": 6.2667,
      "mean_token_accuracy": 0.15942747443914412,
      "num_tokens": 18778472.0,
      "step": 7405
    },
    {
      "entropy": 6.437329912185669,
      "epoch": 0.8551644547028274,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004938116173350078,
      "loss": 6.2963,
      "mean_token_accuracy": 0.15542578101158142,
      "num_tokens": 18791564.0,
      "step": 7410
    },
    {
      "entropy": 6.306224679946899,
      "epoch": 0.8557414887478362,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004938020023930932,
      "loss": 6.26,
      "mean_token_accuracy": 0.1629945918917656,
      "num_tokens": 18804749.0,
      "step": 7415
    },
    {
      "entropy": 6.41823000907898,
      "epoch": 0.8563185227928448,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004937923800918817,
      "loss": 6.2663,
      "mean_token_accuracy": 0.1586477354168892,
      "num_tokens": 18816863.0,
      "step": 7420
    },
    {
      "entropy": 6.342240524291992,
      "epoch": 0.8568955568378535,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004937827504316974,
      "loss": 6.2479,
      "mean_token_accuracy": 0.15983584225177766,
      "num_tokens": 18831543.0,
      "step": 7425
    },
    {
      "entropy": 6.344615936279297,
      "epoch": 0.8574725908828621,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004937731134128639,
      "loss": 6.2841,
      "mean_token_accuracy": 0.15680270344018937,
      "num_tokens": 18844242.0,
      "step": 7430
    },
    {
      "entropy": 6.35708179473877,
      "epoch": 0.8580496249278707,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004937634690357054,
      "loss": 6.2986,
      "mean_token_accuracy": 0.15914062261581421,
      "num_tokens": 18856213.0,
      "step": 7435
    },
    {
      "entropy": 6.42404351234436,
      "epoch": 0.8586266589728794,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004937538173005462,
      "loss": 6.2936,
      "mean_token_accuracy": 0.1603648319840431,
      "num_tokens": 18869680.0,
      "step": 7440
    },
    {
      "entropy": 6.42676305770874,
      "epoch": 0.859203693017888,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004937441582077111,
      "loss": 6.2659,
      "mean_token_accuracy": 0.16333474069833756,
      "num_tokens": 18882360.0,
      "step": 7445
    },
    {
      "entropy": 6.329063272476196,
      "epoch": 0.8597807270628968,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004937344917575249,
      "loss": 6.2367,
      "mean_token_accuracy": 0.16282939016819,
      "num_tokens": 18895138.0,
      "step": 7450
    },
    {
      "entropy": 6.3382631778717045,
      "epoch": 0.8603577611079054,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004937248179503127,
      "loss": 6.29,
      "mean_token_accuracy": 0.15547815710306168,
      "num_tokens": 18908849.0,
      "step": 7455
    },
    {
      "entropy": 6.421544885635376,
      "epoch": 0.860934795152914,
      "grad_norm": 1.25,
      "learning_rate": 0.0004937151367863998,
      "loss": 6.2408,
      "mean_token_accuracy": 0.16327654123306273,
      "num_tokens": 18920757.0,
      "step": 7460
    },
    {
      "entropy": 6.279328489303589,
      "epoch": 0.8615118291979227,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000493705448266112,
      "loss": 6.2043,
      "mean_token_accuracy": 0.16269662827253342,
      "num_tokens": 18933239.0,
      "step": 7465
    },
    {
      "entropy": 6.36497688293457,
      "epoch": 0.8620888632429313,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004936957523897752,
      "loss": 6.2964,
      "mean_token_accuracy": 0.15578166395425797,
      "num_tokens": 18945174.0,
      "step": 7470
    },
    {
      "entropy": 6.409827804565429,
      "epoch": 0.86266589728794,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004936860491577153,
      "loss": 6.3631,
      "mean_token_accuracy": 0.1496423974633217,
      "num_tokens": 18957478.0,
      "step": 7475
    },
    {
      "entropy": 6.42185263633728,
      "epoch": 0.8632429313329486,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004936763385702588,
      "loss": 6.2966,
      "mean_token_accuracy": 0.15954808592796327,
      "num_tokens": 18969461.0,
      "step": 7480
    },
    {
      "entropy": 6.431867456436157,
      "epoch": 0.8638199653779572,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004936666206277322,
      "loss": 6.2852,
      "mean_token_accuracy": 0.15736413449048997,
      "num_tokens": 18981891.0,
      "step": 7485
    },
    {
      "entropy": 6.338661050796508,
      "epoch": 0.864396999422966,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004936568953304624,
      "loss": 6.1624,
      "mean_token_accuracy": 0.17259442210197448,
      "num_tokens": 18993376.0,
      "step": 7490
    },
    {
      "entropy": 6.271618986129761,
      "epoch": 0.8649740334679746,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004936471626787766,
      "loss": 6.2383,
      "mean_token_accuracy": 0.1652093455195427,
      "num_tokens": 19005485.0,
      "step": 7495
    },
    {
      "entropy": 6.287970161437988,
      "epoch": 0.8655510675129833,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004936374226730022,
      "loss": 6.2818,
      "mean_token_accuracy": 0.15577176511287688,
      "num_tokens": 19018509.0,
      "step": 7500
    },
    {
      "entropy": 6.498890399932861,
      "epoch": 0.8661281015579919,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004936276753134666,
      "loss": 6.2644,
      "mean_token_accuracy": 0.15970234721899032,
      "num_tokens": 19030595.0,
      "step": 7505
    },
    {
      "entropy": 6.386164999008178,
      "epoch": 0.8667051356030006,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004936179206004976,
      "loss": 6.3264,
      "mean_token_accuracy": 0.15769854336977004,
      "num_tokens": 19043441.0,
      "step": 7510
    },
    {
      "entropy": 6.300699472427368,
      "epoch": 0.8672821696480092,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004936081585344236,
      "loss": 6.2198,
      "mean_token_accuracy": 0.164374503493309,
      "num_tokens": 19055761.0,
      "step": 7515
    },
    {
      "entropy": 6.352464389801026,
      "epoch": 0.8678592036930178,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004935983891155727,
      "loss": 6.2849,
      "mean_token_accuracy": 0.15642343759536742,
      "num_tokens": 19068211.0,
      "step": 7520
    },
    {
      "entropy": 6.324388551712036,
      "epoch": 0.8684362377380266,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004935886123442737,
      "loss": 6.2409,
      "mean_token_accuracy": 0.15791550874710084,
      "num_tokens": 19081487.0,
      "step": 7525
    },
    {
      "entropy": 6.47940583229065,
      "epoch": 0.8690132717830352,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004935788282208551,
      "loss": 6.2703,
      "mean_token_accuracy": 0.15967122465372086,
      "num_tokens": 19094779.0,
      "step": 7530
    },
    {
      "entropy": 6.392388963699341,
      "epoch": 0.8695903058280439,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004935690367456464,
      "loss": 6.306,
      "mean_token_accuracy": 0.15618782192468644,
      "num_tokens": 19106999.0,
      "step": 7535
    },
    {
      "entropy": 6.351345252990723,
      "epoch": 0.8701673398730525,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004935592379189766,
      "loss": 6.2508,
      "mean_token_accuracy": 0.16028429567813873,
      "num_tokens": 19119570.0,
      "step": 7540
    },
    {
      "entropy": 6.376242208480835,
      "epoch": 0.8707443739180611,
      "grad_norm": 1.125,
      "learning_rate": 0.0004935494317411754,
      "loss": 6.2441,
      "mean_token_accuracy": 0.15781840831041336,
      "num_tokens": 19131101.0,
      "step": 7545
    },
    {
      "entropy": 6.30266227722168,
      "epoch": 0.8713214079630698,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004935396182125726,
      "loss": 6.2324,
      "mean_token_accuracy": 0.16473338305950164,
      "num_tokens": 19144159.0,
      "step": 7550
    },
    {
      "entropy": 6.373244285583496,
      "epoch": 0.8718984420080784,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004935297973334983,
      "loss": 6.2315,
      "mean_token_accuracy": 0.16389684826135636,
      "num_tokens": 19157034.0,
      "step": 7555
    },
    {
      "entropy": 6.315349435806274,
      "epoch": 0.8724754760530872,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004935199691042828,
      "loss": 6.2797,
      "mean_token_accuracy": 0.16112970411777497,
      "num_tokens": 19170813.0,
      "step": 7560
    },
    {
      "entropy": 6.298638486862183,
      "epoch": 0.8730525100980958,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004935101335252568,
      "loss": 6.2118,
      "mean_token_accuracy": 0.16997221261262893,
      "num_tokens": 19183060.0,
      "step": 7565
    },
    {
      "entropy": 6.305606985092163,
      "epoch": 0.8736295441431045,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004935002905967509,
      "loss": 6.1529,
      "mean_token_accuracy": 0.1693923220038414,
      "num_tokens": 19195615.0,
      "step": 7570
    },
    {
      "entropy": 6.37696943283081,
      "epoch": 0.8742065781881131,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004934904403190963,
      "loss": 6.3182,
      "mean_token_accuracy": 0.15287303552031517,
      "num_tokens": 19209531.0,
      "step": 7575
    },
    {
      "entropy": 6.347745609283447,
      "epoch": 0.8747836122331217,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004934805826926242,
      "loss": 6.2462,
      "mean_token_accuracy": 0.15828338414430618,
      "num_tokens": 19221658.0,
      "step": 7580
    },
    {
      "entropy": 6.386853885650635,
      "epoch": 0.8753606462781304,
      "grad_norm": 1.25,
      "learning_rate": 0.0004934707177176663,
      "loss": 6.2483,
      "mean_token_accuracy": 0.15808498859405518,
      "num_tokens": 19233491.0,
      "step": 7585
    },
    {
      "entropy": 6.389758014678955,
      "epoch": 0.875937680323139,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004934608453945543,
      "loss": 6.3564,
      "mean_token_accuracy": 0.1572024181485176,
      "num_tokens": 19246318.0,
      "step": 7590
    },
    {
      "entropy": 6.369310760498047,
      "epoch": 0.8765147143681478,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004934509657236203,
      "loss": 6.2864,
      "mean_token_accuracy": 0.15725102871656418,
      "num_tokens": 19258735.0,
      "step": 7595
    },
    {
      "entropy": 6.427819347381591,
      "epoch": 0.8770917484131564,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004934410787051965,
      "loss": 6.3279,
      "mean_token_accuracy": 0.15563227981328964,
      "num_tokens": 19271640.0,
      "step": 7600
    },
    {
      "entropy": 6.3711097240448,
      "epoch": 0.877668782458165,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004934311843396157,
      "loss": 6.2454,
      "mean_token_accuracy": 0.1667380928993225,
      "num_tokens": 19283824.0,
      "step": 7605
    },
    {
      "entropy": 6.362688446044922,
      "epoch": 0.8782458165031737,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004934212826272104,
      "loss": 6.2207,
      "mean_token_accuracy": 0.16352664679288864,
      "num_tokens": 19298072.0,
      "step": 7610
    },
    {
      "entropy": 6.378260946273803,
      "epoch": 0.8788228505481823,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004934113735683137,
      "loss": 6.2739,
      "mean_token_accuracy": 0.15958625674247742,
      "num_tokens": 19311132.0,
      "step": 7615
    },
    {
      "entropy": 6.383971881866455,
      "epoch": 0.879399884593191,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000493401457163259,
      "loss": 6.2526,
      "mean_token_accuracy": 0.1592327371239662,
      "num_tokens": 19324342.0,
      "step": 7620
    },
    {
      "entropy": 6.341959762573242,
      "epoch": 0.8799769186381996,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004933915334123798,
      "loss": 6.2437,
      "mean_token_accuracy": 0.15743998885154725,
      "num_tokens": 19336629.0,
      "step": 7625
    },
    {
      "entropy": 6.340284395217895,
      "epoch": 0.8805539526832084,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004933816023160099,
      "loss": 6.2805,
      "mean_token_accuracy": 0.1536065548658371,
      "num_tokens": 19348247.0,
      "step": 7630
    },
    {
      "entropy": 6.346293115615845,
      "epoch": 0.881130986728217,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004933716638744832,
      "loss": 6.1855,
      "mean_token_accuracy": 0.16694023162126542,
      "num_tokens": 19361230.0,
      "step": 7635
    },
    {
      "entropy": 6.375268411636353,
      "epoch": 0.8817080207732256,
      "grad_norm": 1.734375,
      "learning_rate": 0.000493361718088134,
      "loss": 6.2119,
      "mean_token_accuracy": 0.1663305252790451,
      "num_tokens": 19373803.0,
      "step": 7640
    },
    {
      "entropy": 6.339536094665528,
      "epoch": 0.8822850548182343,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000493351764957297,
      "loss": 6.1824,
      "mean_token_accuracy": 0.1633446916937828,
      "num_tokens": 19385960.0,
      "step": 7645
    },
    {
      "entropy": 6.336419582366943,
      "epoch": 0.8828620888632429,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004933418044823068,
      "loss": 6.3096,
      "mean_token_accuracy": 0.16220148354768754,
      "num_tokens": 19399649.0,
      "step": 7650
    },
    {
      "entropy": 6.406404209136963,
      "epoch": 0.8834391229082516,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004933318366634984,
      "loss": 6.2852,
      "mean_token_accuracy": 0.16190747767686844,
      "num_tokens": 19413964.0,
      "step": 7655
    },
    {
      "entropy": 6.419785070419311,
      "epoch": 0.8840161569532602,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004933218615012072,
      "loss": 6.297,
      "mean_token_accuracy": 0.1609895721077919,
      "num_tokens": 19425715.0,
      "step": 7660
    },
    {
      "entropy": 6.401238536834716,
      "epoch": 0.8845931909982689,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004933118789957687,
      "loss": 6.2846,
      "mean_token_accuracy": 0.16301700621843337,
      "num_tokens": 19438733.0,
      "step": 7665
    },
    {
      "entropy": 6.419570732116699,
      "epoch": 0.8851702250432776,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004933018891475186,
      "loss": 6.3611,
      "mean_token_accuracy": 0.1552258089184761,
      "num_tokens": 19451924.0,
      "step": 7670
    },
    {
      "entropy": 6.347110462188721,
      "epoch": 0.8857472590882862,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004932918919567927,
      "loss": 6.2186,
      "mean_token_accuracy": 0.1645390957593918,
      "num_tokens": 19464331.0,
      "step": 7675
    },
    {
      "entropy": 6.327801513671875,
      "epoch": 0.8863242931332949,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004932818874239275,
      "loss": 6.2452,
      "mean_token_accuracy": 0.15719466060400009,
      "num_tokens": 19477127.0,
      "step": 7680
    },
    {
      "entropy": 6.3188268661499025,
      "epoch": 0.8869013271783035,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004932718755492595,
      "loss": 6.2019,
      "mean_token_accuracy": 0.16648583114147186,
      "num_tokens": 19490281.0,
      "step": 7685
    },
    {
      "entropy": 6.295724391937256,
      "epoch": 0.8874783612233121,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004932618563331254,
      "loss": 6.2153,
      "mean_token_accuracy": 0.169368976354599,
      "num_tokens": 19503154.0,
      "step": 7690
    },
    {
      "entropy": 6.353931093215943,
      "epoch": 0.8880553952683208,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004932518297758622,
      "loss": 6.3289,
      "mean_token_accuracy": 0.16042528301477432,
      "num_tokens": 19517363.0,
      "step": 7695
    },
    {
      "entropy": 6.383228445053101,
      "epoch": 0.8886324293133295,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004932417958778071,
      "loss": 6.2313,
      "mean_token_accuracy": 0.1621527314186096,
      "num_tokens": 19529785.0,
      "step": 7700
    },
    {
      "entropy": 6.330796480178833,
      "epoch": 0.8892094633583382,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004932317546392976,
      "loss": 6.3056,
      "mean_token_accuracy": 0.16220931336283684,
      "num_tokens": 19542025.0,
      "step": 7705
    },
    {
      "entropy": 6.391780138015747,
      "epoch": 0.8897864974033468,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004932217060606714,
      "loss": 6.2915,
      "mean_token_accuracy": 0.15693818777799606,
      "num_tokens": 19553969.0,
      "step": 7710
    },
    {
      "entropy": 6.364717960357666,
      "epoch": 0.8903635314483554,
      "grad_norm": 1.25,
      "learning_rate": 0.0004932116501422665,
      "loss": 6.1745,
      "mean_token_accuracy": 0.16281207799911498,
      "num_tokens": 19566568.0,
      "step": 7715
    },
    {
      "entropy": 6.287318897247315,
      "epoch": 0.8909405654933641,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004932015868844211,
      "loss": 6.2707,
      "mean_token_accuracy": 0.15486115962266922,
      "num_tokens": 19579947.0,
      "step": 7720
    },
    {
      "entropy": 6.346429347991943,
      "epoch": 0.8915175995383727,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004931915162874738,
      "loss": 6.2674,
      "mean_token_accuracy": 0.1584528475999832,
      "num_tokens": 19592457.0,
      "step": 7725
    },
    {
      "entropy": 6.45083589553833,
      "epoch": 0.8920946335833814,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004931814383517632,
      "loss": 6.2622,
      "mean_token_accuracy": 0.15782222300767898,
      "num_tokens": 19604165.0,
      "step": 7730
    },
    {
      "entropy": 6.329137945175171,
      "epoch": 0.8926716676283901,
      "grad_norm": 1.5,
      "learning_rate": 0.0004931713530776284,
      "loss": 6.266,
      "mean_token_accuracy": 0.15937328934669495,
      "num_tokens": 19617060.0,
      "step": 7735
    },
    {
      "entropy": 6.350716257095337,
      "epoch": 0.8932487016733988,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004931612604654083,
      "loss": 6.3109,
      "mean_token_accuracy": 0.14904512763023375,
      "num_tokens": 19628724.0,
      "step": 7740
    },
    {
      "entropy": 6.469025182723999,
      "epoch": 0.8938257357184074,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004931511605154428,
      "loss": 6.3259,
      "mean_token_accuracy": 0.16162268668413163,
      "num_tokens": 19642281.0,
      "step": 7745
    },
    {
      "entropy": 6.2978919506072994,
      "epoch": 0.894402769763416,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004931410532280711,
      "loss": 6.2381,
      "mean_token_accuracy": 0.16269858926534653,
      "num_tokens": 19655503.0,
      "step": 7750
    },
    {
      "entropy": 6.34242811203003,
      "epoch": 0.8949798038084247,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004931309386036337,
      "loss": 6.2341,
      "mean_token_accuracy": 0.158306260406971,
      "num_tokens": 19667758.0,
      "step": 7755
    },
    {
      "entropy": 6.341038179397583,
      "epoch": 0.8955568378534333,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004931208166424704,
      "loss": 6.289,
      "mean_token_accuracy": 0.16237967163324357,
      "num_tokens": 19679878.0,
      "step": 7760
    },
    {
      "entropy": 6.443572187423706,
      "epoch": 0.896133871898442,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004931106873449219,
      "loss": 6.3402,
      "mean_token_accuracy": 0.15650419741868973,
      "num_tokens": 19692278.0,
      "step": 7765
    },
    {
      "entropy": 6.369319057464599,
      "epoch": 0.8967109059434507,
      "grad_norm": 1.25,
      "learning_rate": 0.0004931005507113285,
      "loss": 6.1531,
      "mean_token_accuracy": 0.17518105059862138,
      "num_tokens": 19705319.0,
      "step": 7770
    },
    {
      "entropy": 6.32936282157898,
      "epoch": 0.8972879399884593,
      "grad_norm": 1.375,
      "learning_rate": 0.0004930904067420317,
      "loss": 6.2385,
      "mean_token_accuracy": 0.15738262236118317,
      "num_tokens": 19717908.0,
      "step": 7775
    },
    {
      "entropy": 6.461233234405517,
      "epoch": 0.897864974033468,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004930802554373723,
      "loss": 6.3465,
      "mean_token_accuracy": 0.15694792270660402,
      "num_tokens": 19730810.0,
      "step": 7780
    },
    {
      "entropy": 6.423625707626343,
      "epoch": 0.8984420080784766,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004930700967976918,
      "loss": 6.269,
      "mean_token_accuracy": 0.16405607610940934,
      "num_tokens": 19744176.0,
      "step": 7785
    },
    {
      "entropy": 6.320487785339355,
      "epoch": 0.8990190421234853,
      "grad_norm": 1.234375,
      "learning_rate": 0.000493059930823332,
      "loss": 6.1856,
      "mean_token_accuracy": 0.161333566904068,
      "num_tokens": 19756072.0,
      "step": 7790
    },
    {
      "entropy": 6.289909696578979,
      "epoch": 0.8995960761684939,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004930497575146346,
      "loss": 6.3014,
      "mean_token_accuracy": 0.158029642701149,
      "num_tokens": 19768170.0,
      "step": 7795
    },
    {
      "entropy": 6.287299871444702,
      "epoch": 0.9001731102135025,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004930395768719421,
      "loss": 6.2627,
      "mean_token_accuracy": 0.16084418296813965,
      "num_tokens": 19782029.0,
      "step": 7800
    },
    {
      "entropy": 6.37008810043335,
      "epoch": 0.9007501442585113,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004930293888955966,
      "loss": 6.2044,
      "mean_token_accuracy": 0.16015943884849548,
      "num_tokens": 19793973.0,
      "step": 7805
    },
    {
      "entropy": 6.2776854038238525,
      "epoch": 0.9013271783035199,
      "grad_norm": 1.140625,
      "learning_rate": 0.000493019193585941,
      "loss": 6.1722,
      "mean_token_accuracy": 0.16408360302448272,
      "num_tokens": 19806806.0,
      "step": 7810
    },
    {
      "entropy": 6.385754156112671,
      "epoch": 0.9019042123485286,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000493008990943318,
      "loss": 6.3091,
      "mean_token_accuracy": 0.1537381663918495,
      "num_tokens": 19821025.0,
      "step": 7815
    },
    {
      "entropy": 6.3435084342956545,
      "epoch": 0.9024812463935372,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004929987809680709,
      "loss": 6.1212,
      "mean_token_accuracy": 0.17080078423023223,
      "num_tokens": 19832909.0,
      "step": 7820
    },
    {
      "entropy": 6.264006185531616,
      "epoch": 0.9030582804385459,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004929885636605432,
      "loss": 6.1573,
      "mean_token_accuracy": 0.16311961114406587,
      "num_tokens": 19846813.0,
      "step": 7825
    },
    {
      "entropy": 6.369507551193237,
      "epoch": 0.9036353144835545,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004929783390210784,
      "loss": 6.2999,
      "mean_token_accuracy": 0.15389039888978004,
      "num_tokens": 19860005.0,
      "step": 7830
    },
    {
      "entropy": 6.422854995727539,
      "epoch": 0.9042123485285631,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004929681070500204,
      "loss": 6.2172,
      "mean_token_accuracy": 0.1588687479496002,
      "num_tokens": 19871931.0,
      "step": 7835
    },
    {
      "entropy": 6.3597208023071286,
      "epoch": 0.9047893825735719,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004929578677477135,
      "loss": 6.1913,
      "mean_token_accuracy": 0.16449404507875443,
      "num_tokens": 19884066.0,
      "step": 7840
    },
    {
      "entropy": 6.294052934646606,
      "epoch": 0.9053664166185805,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004929476211145019,
      "loss": 6.2798,
      "mean_token_accuracy": 0.15670019686222075,
      "num_tokens": 19896416.0,
      "step": 7845
    },
    {
      "entropy": 6.439695835113525,
      "epoch": 0.9059434506635892,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004929373671507303,
      "loss": 6.2915,
      "mean_token_accuracy": 0.15320463627576827,
      "num_tokens": 19911036.0,
      "step": 7850
    },
    {
      "entropy": 6.393083143234253,
      "epoch": 0.9065204847085978,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004929271058567436,
      "loss": 6.2791,
      "mean_token_accuracy": 0.15976049602031708,
      "num_tokens": 19924516.0,
      "step": 7855
    },
    {
      "entropy": 6.288011169433593,
      "epoch": 0.9070975187536064,
      "grad_norm": 1.265625,
      "learning_rate": 0.000492916837232887,
      "loss": 6.1675,
      "mean_token_accuracy": 0.16321094781160356,
      "num_tokens": 19936994.0,
      "step": 7860
    },
    {
      "entropy": 6.299168395996094,
      "epoch": 0.9076745527986151,
      "grad_norm": 1.375,
      "learning_rate": 0.0004929065612795058,
      "loss": 6.2138,
      "mean_token_accuracy": 0.16462670713663102,
      "num_tokens": 19950304.0,
      "step": 7865
    },
    {
      "entropy": 6.2785231590271,
      "epoch": 0.9082515868436237,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004928962779969456,
      "loss": 6.2634,
      "mean_token_accuracy": 0.1546306237578392,
      "num_tokens": 19963842.0,
      "step": 7870
    },
    {
      "entropy": 6.3666106224060055,
      "epoch": 0.9088286208886325,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004928859873855524,
      "loss": 6.2819,
      "mean_token_accuracy": 0.15856581926345825,
      "num_tokens": 19975806.0,
      "step": 7875
    },
    {
      "entropy": 6.3063102722167965,
      "epoch": 0.9094056549336411,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004928756894456723,
      "loss": 6.2181,
      "mean_token_accuracy": 0.16397362798452378,
      "num_tokens": 19987647.0,
      "step": 7880
    },
    {
      "entropy": 6.425796890258789,
      "epoch": 0.9099826889786498,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004928653841776515,
      "loss": 6.2926,
      "mean_token_accuracy": 0.15438606441020966,
      "num_tokens": 20000591.0,
      "step": 7885
    },
    {
      "entropy": 6.429916858673096,
      "epoch": 0.9105597230236584,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004928550715818368,
      "loss": 6.1986,
      "mean_token_accuracy": 0.1631989985704422,
      "num_tokens": 20012657.0,
      "step": 7890
    },
    {
      "entropy": 6.375060510635376,
      "epoch": 0.911136757068667,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004928447516585749,
      "loss": 6.2906,
      "mean_token_accuracy": 0.15863371789455413,
      "num_tokens": 20024752.0,
      "step": 7895
    },
    {
      "entropy": 6.277575063705444,
      "epoch": 0.9117137911136757,
      "grad_norm": 1.171875,
      "learning_rate": 0.000492834424408213,
      "loss": 6.1999,
      "mean_token_accuracy": 0.16147911846637725,
      "num_tokens": 20036896.0,
      "step": 7900
    },
    {
      "entropy": 6.341118240356446,
      "epoch": 0.9122908251586843,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004928240898310984,
      "loss": 6.2725,
      "mean_token_accuracy": 0.16054617315530778,
      "num_tokens": 20049421.0,
      "step": 7905
    },
    {
      "entropy": 6.404549932479858,
      "epoch": 0.9128678592036931,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004928137479275789,
      "loss": 6.313,
      "mean_token_accuracy": 0.15654291659593583,
      "num_tokens": 20062298.0,
      "step": 7910
    },
    {
      "entropy": 6.39699501991272,
      "epoch": 0.9134448932487017,
      "grad_norm": 3.765625,
      "learning_rate": 0.000492803398698002,
      "loss": 6.1985,
      "mean_token_accuracy": 0.15904554724693298,
      "num_tokens": 20075954.0,
      "step": 7915
    },
    {
      "entropy": 6.274241542816162,
      "epoch": 0.9140219272937103,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004927930421427161,
      "loss": 6.164,
      "mean_token_accuracy": 0.16510989367961884,
      "num_tokens": 20088035.0,
      "step": 7920
    },
    {
      "entropy": 6.3246208190917965,
      "epoch": 0.914598961338719,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004927826782620694,
      "loss": 6.1896,
      "mean_token_accuracy": 0.16373167783021927,
      "num_tokens": 20100081.0,
      "step": 7925
    },
    {
      "entropy": 6.3862145900726315,
      "epoch": 0.9151759953837276,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004927723070564104,
      "loss": 6.2705,
      "mean_token_accuracy": 0.15828072875738144,
      "num_tokens": 20113159.0,
      "step": 7930
    },
    {
      "entropy": 6.3315966606140135,
      "epoch": 0.9157530294287363,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004927619285260881,
      "loss": 6.2159,
      "mean_token_accuracy": 0.16448910236358644,
      "num_tokens": 20125741.0,
      "step": 7935
    },
    {
      "entropy": 6.340218925476075,
      "epoch": 0.9163300634737449,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004927515426714515,
      "loss": 6.1935,
      "mean_token_accuracy": 0.16135872304439544,
      "num_tokens": 20137481.0,
      "step": 7940
    },
    {
      "entropy": 6.3401132106781,
      "epoch": 0.9169070975187537,
      "grad_norm": 1.546875,
      "learning_rate": 0.00049274114949285,
      "loss": 6.2718,
      "mean_token_accuracy": 0.16108395904302597,
      "num_tokens": 20150188.0,
      "step": 7945
    },
    {
      "entropy": 6.400226259231568,
      "epoch": 0.9174841315637623,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000492730748990633,
      "loss": 6.3139,
      "mean_token_accuracy": 0.1535964459180832,
      "num_tokens": 20163012.0,
      "step": 7950
    },
    {
      "entropy": 6.359333896636963,
      "epoch": 0.9180611656087709,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004927203411651504,
      "loss": 6.2355,
      "mean_token_accuracy": 0.16038236618041993,
      "num_tokens": 20175468.0,
      "step": 7955
    },
    {
      "entropy": 6.323717975616455,
      "epoch": 0.9186381996537796,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004927099260167523,
      "loss": 6.2475,
      "mean_token_accuracy": 0.16237496733665466,
      "num_tokens": 20188382.0,
      "step": 7960
    },
    {
      "entropy": 6.401786851882934,
      "epoch": 0.9192152336987882,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004926995035457889,
      "loss": 6.237,
      "mean_token_accuracy": 0.1669002279639244,
      "num_tokens": 20200091.0,
      "step": 7965
    },
    {
      "entropy": 6.327285194396973,
      "epoch": 0.9197922677437969,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000492689073752611,
      "loss": 6.2547,
      "mean_token_accuracy": 0.1570654809474945,
      "num_tokens": 20212004.0,
      "step": 7970
    },
    {
      "entropy": 6.294810771942139,
      "epoch": 0.9203693017888055,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004926786366375691,
      "loss": 6.2795,
      "mean_token_accuracy": 0.16519549190998079,
      "num_tokens": 20225214.0,
      "step": 7975
    },
    {
      "entropy": 6.331897640228272,
      "epoch": 0.9209463358338142,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004926681922010145,
      "loss": 6.2041,
      "mean_token_accuracy": 0.16371652483940125,
      "num_tokens": 20236595.0,
      "step": 7980
    },
    {
      "entropy": 6.303496217727661,
      "epoch": 0.9215233698788229,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004926577404432982,
      "loss": 6.2132,
      "mean_token_accuracy": 0.15896149724721909,
      "num_tokens": 20248601.0,
      "step": 7985
    },
    {
      "entropy": 6.384378385543823,
      "epoch": 0.9221004039238315,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004926472813647721,
      "loss": 6.233,
      "mean_token_accuracy": 0.15437018722295762,
      "num_tokens": 20261314.0,
      "step": 7990
    },
    {
      "entropy": 6.42190899848938,
      "epoch": 0.9226774379688402,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004926368149657876,
      "loss": 6.285,
      "mean_token_accuracy": 0.16078990399837495,
      "num_tokens": 20274797.0,
      "step": 7995
    },
    {
      "entropy": 6.336082315444946,
      "epoch": 0.9232544720138488,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004926263412466972,
      "loss": 6.316,
      "mean_token_accuracy": 0.15320343524217606,
      "num_tokens": 20287031.0,
      "step": 8000
    },
    {
      "entropy": 6.317395257949829,
      "epoch": 0.9238315060588574,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004926158602078527,
      "loss": 6.2133,
      "mean_token_accuracy": 0.15914132446050644,
      "num_tokens": 20298716.0,
      "step": 8005
    },
    {
      "entropy": 6.323201513290405,
      "epoch": 0.9244085401038661,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004926053718496069,
      "loss": 6.1843,
      "mean_token_accuracy": 0.17220845371484755,
      "num_tokens": 20311129.0,
      "step": 8010
    },
    {
      "entropy": 6.176615142822266,
      "epoch": 0.9249855741488748,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004925948761723126,
      "loss": 6.092,
      "mean_token_accuracy": 0.17000402957201005,
      "num_tokens": 20323067.0,
      "step": 8015
    },
    {
      "entropy": 6.3518040657043455,
      "epoch": 0.9255626081938835,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004925843731763226,
      "loss": 6.2543,
      "mean_token_accuracy": 0.16006439179182053,
      "num_tokens": 20335710.0,
      "step": 8020
    },
    {
      "entropy": 6.435591554641723,
      "epoch": 0.9261396422388921,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004925738628619904,
      "loss": 6.3063,
      "mean_token_accuracy": 0.15671738535165786,
      "num_tokens": 20348436.0,
      "step": 8025
    },
    {
      "entropy": 6.3119806289672855,
      "epoch": 0.9267166762839008,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004925633452296693,
      "loss": 6.2343,
      "mean_token_accuracy": 0.1559642806649208,
      "num_tokens": 20362182.0,
      "step": 8030
    },
    {
      "entropy": 6.341772365570068,
      "epoch": 0.9272937103289094,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004925528202797131,
      "loss": 6.2552,
      "mean_token_accuracy": 0.15676630288362503,
      "num_tokens": 20374674.0,
      "step": 8035
    },
    {
      "entropy": 6.321902847290039,
      "epoch": 0.927870744373918,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004925422880124761,
      "loss": 6.2003,
      "mean_token_accuracy": 0.16159740835428238,
      "num_tokens": 20388918.0,
      "step": 8040
    },
    {
      "entropy": 6.306786632537841,
      "epoch": 0.9284477784189267,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004925317484283121,
      "loss": 6.1905,
      "mean_token_accuracy": 0.16130646765232087,
      "num_tokens": 20401428.0,
      "step": 8045
    },
    {
      "entropy": 6.364667177200317,
      "epoch": 0.9290248124639354,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004925212015275758,
      "loss": 6.2321,
      "mean_token_accuracy": 0.16572524309158326,
      "num_tokens": 20414276.0,
      "step": 8050
    },
    {
      "entropy": 6.3438468933105465,
      "epoch": 0.9296018465089441,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000492510647310622,
      "loss": 6.2014,
      "mean_token_accuracy": 0.16274796277284623,
      "num_tokens": 20425515.0,
      "step": 8055
    },
    {
      "entropy": 6.3759650707244875,
      "epoch": 0.9301788805539527,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004925000857778055,
      "loss": 6.2607,
      "mean_token_accuracy": 0.16023021340370178,
      "num_tokens": 20439137.0,
      "step": 8060
    },
    {
      "entropy": 6.240764379501343,
      "epoch": 0.9307559145989613,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004924895169294818,
      "loss": 6.1491,
      "mean_token_accuracy": 0.1668257012963295,
      "num_tokens": 20452059.0,
      "step": 8065
    },
    {
      "entropy": 6.392715072631836,
      "epoch": 0.93133294864397,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004924789407660062,
      "loss": 6.3269,
      "mean_token_accuracy": 0.1515945538878441,
      "num_tokens": 20465110.0,
      "step": 8070
    },
    {
      "entropy": 6.246210718154908,
      "epoch": 0.9319099826889786,
      "grad_norm": 1.125,
      "learning_rate": 0.0004924683572877345,
      "loss": 6.0972,
      "mean_token_accuracy": 0.1627405032515526,
      "num_tokens": 20477615.0,
      "step": 8075
    },
    {
      "entropy": 6.293426895141602,
      "epoch": 0.9324870167339873,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004924577664950225,
      "loss": 6.1122,
      "mean_token_accuracy": 0.17363643050193786,
      "num_tokens": 20490054.0,
      "step": 8080
    },
    {
      "entropy": 6.340121221542359,
      "epoch": 0.933064050778996,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004924471683882266,
      "loss": 6.2882,
      "mean_token_accuracy": 0.1604973256587982,
      "num_tokens": 20502439.0,
      "step": 8085
    },
    {
      "entropy": 6.325512266159057,
      "epoch": 0.9336410848240047,
      "grad_norm": 1.125,
      "learning_rate": 0.0004924365629677031,
      "loss": 6.2498,
      "mean_token_accuracy": 0.1647322356700897,
      "num_tokens": 20515094.0,
      "step": 8090
    },
    {
      "entropy": 6.379918050765991,
      "epoch": 0.9342181188690133,
      "grad_norm": 1.25,
      "learning_rate": 0.000492425950233809,
      "loss": 6.1739,
      "mean_token_accuracy": 0.16960037648677825,
      "num_tokens": 20527025.0,
      "step": 8095
    },
    {
      "entropy": 6.366863346099853,
      "epoch": 0.9347951529140219,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000492415330186901,
      "loss": 6.3087,
      "mean_token_accuracy": 0.1571563497185707,
      "num_tokens": 20540793.0,
      "step": 8100
    },
    {
      "entropy": 6.3741455078125,
      "epoch": 0.9353721869590306,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004924047028273364,
      "loss": 6.2886,
      "mean_token_accuracy": 0.1573895201086998,
      "num_tokens": 20553425.0,
      "step": 8105
    },
    {
      "entropy": 6.362455797195435,
      "epoch": 0.9359492210040392,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004923940681554725,
      "loss": 6.2596,
      "mean_token_accuracy": 0.15950031876564025,
      "num_tokens": 20566794.0,
      "step": 8110
    },
    {
      "entropy": 6.373832893371582,
      "epoch": 0.9365262550490479,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004923834261716672,
      "loss": 6.3115,
      "mean_token_accuracy": 0.15442514270544053,
      "num_tokens": 20579756.0,
      "step": 8115
    },
    {
      "entropy": 6.3996296405792235,
      "epoch": 0.9371032890940566,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004923727768762782,
      "loss": 6.2943,
      "mean_token_accuracy": 0.15450644940137864,
      "num_tokens": 20592626.0,
      "step": 8120
    },
    {
      "entropy": 6.332824230194092,
      "epoch": 0.9376803231390652,
      "grad_norm": 1.078125,
      "learning_rate": 0.000492362120269664,
      "loss": 6.1993,
      "mean_token_accuracy": 0.16827336251735686,
      "num_tokens": 20605023.0,
      "step": 8125
    },
    {
      "entropy": 6.36417064666748,
      "epoch": 0.9382573571840739,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004923514563521827,
      "loss": 6.341,
      "mean_token_accuracy": 0.15492560118436813,
      "num_tokens": 20617668.0,
      "step": 8130
    },
    {
      "entropy": 6.358595943450927,
      "epoch": 0.9388343912290825,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004923407851241933,
      "loss": 6.2039,
      "mean_token_accuracy": 0.16324663013219834,
      "num_tokens": 20629848.0,
      "step": 8135
    },
    {
      "entropy": 6.281935834884644,
      "epoch": 0.9394114252740912,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004923301065860545,
      "loss": 6.2222,
      "mean_token_accuracy": 0.16378238350152968,
      "num_tokens": 20642695.0,
      "step": 8140
    },
    {
      "entropy": 6.38076639175415,
      "epoch": 0.9399884593190998,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004923194207381254,
      "loss": 6.2582,
      "mean_token_accuracy": 0.16250781565904618,
      "num_tokens": 20655844.0,
      "step": 8145
    },
    {
      "entropy": 6.396403360366821,
      "epoch": 0.9405654933641084,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004923087275807656,
      "loss": 6.3275,
      "mean_token_accuracy": 0.15893930941820145,
      "num_tokens": 20668380.0,
      "step": 8150
    },
    {
      "entropy": 6.385412645339966,
      "epoch": 0.9411425274091172,
      "grad_norm": 1.0,
      "learning_rate": 0.0004922980271143347,
      "loss": 6.219,
      "mean_token_accuracy": 0.1706279993057251,
      "num_tokens": 20680766.0,
      "step": 8155
    },
    {
      "entropy": 6.3168497562408445,
      "epoch": 0.9417195614541258,
      "grad_norm": 1.625,
      "learning_rate": 0.0004922873193391927,
      "loss": 6.2181,
      "mean_token_accuracy": 0.15875881314277648,
      "num_tokens": 20692538.0,
      "step": 8160
    },
    {
      "entropy": 6.3164561748504635,
      "epoch": 0.9422965954991345,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004922766042556994,
      "loss": 6.2219,
      "mean_token_accuracy": 0.16642269045114516,
      "num_tokens": 20705832.0,
      "step": 8165
    },
    {
      "entropy": 6.337274217605591,
      "epoch": 0.9428736295441431,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004922658818642156,
      "loss": 6.2724,
      "mean_token_accuracy": 0.16034751683473586,
      "num_tokens": 20718741.0,
      "step": 8170
    },
    {
      "entropy": 6.379843282699585,
      "epoch": 0.9434506635891518,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004922551521651018,
      "loss": 6.2302,
      "mean_token_accuracy": 0.16260662525892258,
      "num_tokens": 20731516.0,
      "step": 8175
    },
    {
      "entropy": 6.414315986633301,
      "epoch": 0.9440276976341604,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004922444151587189,
      "loss": 6.2883,
      "mean_token_accuracy": 0.15603816956281663,
      "num_tokens": 20744835.0,
      "step": 8180
    },
    {
      "entropy": 6.345683002471924,
      "epoch": 0.944604731679169,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004922336708454279,
      "loss": 6.2603,
      "mean_token_accuracy": 0.15430638641119004,
      "num_tokens": 20758967.0,
      "step": 8185
    },
    {
      "entropy": 6.32820692062378,
      "epoch": 0.9451817657241778,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004922229192255903,
      "loss": 6.2026,
      "mean_token_accuracy": 0.16443230360746383,
      "num_tokens": 20771660.0,
      "step": 8190
    },
    {
      "entropy": 6.344119882583618,
      "epoch": 0.9457587997691864,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004922121602995678,
      "loss": 6.3039,
      "mean_token_accuracy": 0.15712254643440246,
      "num_tokens": 20785366.0,
      "step": 8195
    },
    {
      "entropy": 6.477925062179565,
      "epoch": 0.9463358338141951,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004922013940677221,
      "loss": 6.3098,
      "mean_token_accuracy": 0.15785575956106185,
      "num_tokens": 20797996.0,
      "step": 8200
    },
    {
      "entropy": 6.339235830307007,
      "epoch": 0.9469128678592037,
      "grad_norm": 1.5,
      "learning_rate": 0.0004921906205304155,
      "loss": 6.2596,
      "mean_token_accuracy": 0.1587812900543213,
      "num_tokens": 20810921.0,
      "step": 8205
    },
    {
      "entropy": 6.334531021118164,
      "epoch": 0.9474899019042123,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004921798396880101,
      "loss": 6.1416,
      "mean_token_accuracy": 0.1633761078119278,
      "num_tokens": 20823143.0,
      "step": 8210
    },
    {
      "entropy": 6.338975191116333,
      "epoch": 0.948066935949221,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004921690515408688,
      "loss": 6.2495,
      "mean_token_accuracy": 0.15717073380947114,
      "num_tokens": 20834604.0,
      "step": 8215
    },
    {
      "entropy": 6.343840265274048,
      "epoch": 0.9486439699942296,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004921582560893543,
      "loss": 6.2808,
      "mean_token_accuracy": 0.16047024428844453,
      "num_tokens": 20847224.0,
      "step": 8220
    },
    {
      "entropy": 6.403699731826782,
      "epoch": 0.9492210040392384,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004921474533338297,
      "loss": 6.2816,
      "mean_token_accuracy": 0.16321684718132018,
      "num_tokens": 20859477.0,
      "step": 8225
    },
    {
      "entropy": 6.35962929725647,
      "epoch": 0.949798038084247,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004921366432746585,
      "loss": 6.2311,
      "mean_token_accuracy": 0.15869871377944947,
      "num_tokens": 20872144.0,
      "step": 8230
    },
    {
      "entropy": 6.340493297576904,
      "epoch": 0.9503750721292556,
      "grad_norm": 1.078125,
      "learning_rate": 0.000492125825912204,
      "loss": 6.1934,
      "mean_token_accuracy": 0.16306080669164658,
      "num_tokens": 20885168.0,
      "step": 8235
    },
    {
      "entropy": 6.34116849899292,
      "epoch": 0.9509521061742643,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004921150012468302,
      "loss": 6.2351,
      "mean_token_accuracy": 0.15923132449388505,
      "num_tokens": 20896650.0,
      "step": 8240
    },
    {
      "entropy": 6.353695058822632,
      "epoch": 0.9515291402192729,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004921041692789013,
      "loss": 6.3356,
      "mean_token_accuracy": 0.15304399877786637,
      "num_tokens": 20908764.0,
      "step": 8245
    },
    {
      "entropy": 6.344550323486328,
      "epoch": 0.9521061742642816,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004920933300087813,
      "loss": 6.2556,
      "mean_token_accuracy": 0.15438825786113738,
      "num_tokens": 20920852.0,
      "step": 8250
    },
    {
      "entropy": 6.387536191940308,
      "epoch": 0.9526832083092902,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004920824834368353,
      "loss": 6.211,
      "mean_token_accuracy": 0.1632569134235382,
      "num_tokens": 20933747.0,
      "step": 8255
    },
    {
      "entropy": 6.347301054000854,
      "epoch": 0.953260242354299,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004920716295634275,
      "loss": 6.2518,
      "mean_token_accuracy": 0.16736212372779846,
      "num_tokens": 20945563.0,
      "step": 8260
    },
    {
      "entropy": 6.29720549583435,
      "epoch": 0.9538372763993076,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004920607683889235,
      "loss": 6.2133,
      "mean_token_accuracy": 0.15855415910482407,
      "num_tokens": 20957449.0,
      "step": 8265
    },
    {
      "entropy": 6.344671249389648,
      "epoch": 0.9544143104443162,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004920498999136882,
      "loss": 6.1728,
      "mean_token_accuracy": 0.1661275342106819,
      "num_tokens": 20969548.0,
      "step": 8270
    },
    {
      "entropy": 6.2877216815948485,
      "epoch": 0.9549913444893249,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004920390241380874,
      "loss": 6.1678,
      "mean_token_accuracy": 0.16137129962444305,
      "num_tokens": 20981712.0,
      "step": 8275
    },
    {
      "entropy": 6.309497261047364,
      "epoch": 0.9555683785343335,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004920281410624868,
      "loss": 6.2668,
      "mean_token_accuracy": 0.16037406176328659,
      "num_tokens": 20993997.0,
      "step": 8280
    },
    {
      "entropy": 6.430056142807007,
      "epoch": 0.9561454125793422,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004920172506872525,
      "loss": 6.3048,
      "mean_token_accuracy": 0.15896093845367432,
      "num_tokens": 21007158.0,
      "step": 8285
    },
    {
      "entropy": 6.308279275894165,
      "epoch": 0.9567224466243508,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004920063530127508,
      "loss": 6.1932,
      "mean_token_accuracy": 0.15953774899244308,
      "num_tokens": 21019592.0,
      "step": 8290
    },
    {
      "entropy": 6.38461012840271,
      "epoch": 0.9572994806693594,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004919954480393482,
      "loss": 6.2649,
      "mean_token_accuracy": 0.166105055809021,
      "num_tokens": 21031850.0,
      "step": 8295
    },
    {
      "entropy": 6.367042350769043,
      "epoch": 0.9578765147143682,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004919845357674114,
      "loss": 6.1896,
      "mean_token_accuracy": 0.16621782779693603,
      "num_tokens": 21044348.0,
      "step": 8300
    },
    {
      "entropy": 6.3342314720153805,
      "epoch": 0.9584535487593768,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004919736161973076,
      "loss": 6.2924,
      "mean_token_accuracy": 0.1633678510785103,
      "num_tokens": 21058017.0,
      "step": 8305
    },
    {
      "entropy": 6.404982948303223,
      "epoch": 0.9590305828043855,
      "grad_norm": 1.203125,
      "learning_rate": 0.000491962689329404,
      "loss": 6.2038,
      "mean_token_accuracy": 0.15929311662912368,
      "num_tokens": 21070443.0,
      "step": 8310
    },
    {
      "entropy": 6.354428720474243,
      "epoch": 0.9596076168493941,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004919517551640681,
      "loss": 6.3089,
      "mean_token_accuracy": 0.16186899542808533,
      "num_tokens": 21084033.0,
      "step": 8315
    },
    {
      "entropy": 6.409770393371582,
      "epoch": 0.9601846508944027,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004919408137016677,
      "loss": 6.2299,
      "mean_token_accuracy": 0.15859173089265824,
      "num_tokens": 21095737.0,
      "step": 8320
    },
    {
      "entropy": 6.409395027160644,
      "epoch": 0.9607616849394114,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004919298649425708,
      "loss": 6.2089,
      "mean_token_accuracy": 0.164140485227108,
      "num_tokens": 21107769.0,
      "step": 8325
    },
    {
      "entropy": 6.222255373001099,
      "epoch": 0.96133871898442,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004919189088871456,
      "loss": 6.1459,
      "mean_token_accuracy": 0.16910335868597032,
      "num_tokens": 21119008.0,
      "step": 8330
    },
    {
      "entropy": 6.320226526260376,
      "epoch": 0.9619157530294288,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004919079455357608,
      "loss": 6.2656,
      "mean_token_accuracy": 0.16041919440031052,
      "num_tokens": 21131433.0,
      "step": 8335
    },
    {
      "entropy": 6.4347902774810795,
      "epoch": 0.9624927870744374,
      "grad_norm": 1.25,
      "learning_rate": 0.0004918969748887847,
      "loss": 6.3153,
      "mean_token_accuracy": 0.15121374577283858,
      "num_tokens": 21144237.0,
      "step": 8340
    },
    {
      "entropy": 6.428329229354858,
      "epoch": 0.9630698211194461,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004918859969465868,
      "loss": 6.3252,
      "mean_token_accuracy": 0.1559145763516426,
      "num_tokens": 21157795.0,
      "step": 8345
    },
    {
      "entropy": 6.323662614822387,
      "epoch": 0.9636468551644547,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004918750117095361,
      "loss": 6.2426,
      "mean_token_accuracy": 0.15779153555631636,
      "num_tokens": 21169943.0,
      "step": 8350
    },
    {
      "entropy": 6.304287052154541,
      "epoch": 0.9642238892094633,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004918640191780021,
      "loss": 6.15,
      "mean_token_accuracy": 0.16351599246263504,
      "num_tokens": 21184432.0,
      "step": 8355
    },
    {
      "entropy": 6.3206535339355465,
      "epoch": 0.964800923254472,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004918530193523546,
      "loss": 6.1913,
      "mean_token_accuracy": 0.16340121179819106,
      "num_tokens": 21196539.0,
      "step": 8360
    },
    {
      "entropy": 6.317800664901734,
      "epoch": 0.9653779572994806,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004918420122329634,
      "loss": 6.2461,
      "mean_token_accuracy": 0.15788242518901824,
      "num_tokens": 21209550.0,
      "step": 8365
    },
    {
      "entropy": 6.272250461578369,
      "epoch": 0.9659549913444894,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004918309978201989,
      "loss": 6.1267,
      "mean_token_accuracy": 0.16811427772045134,
      "num_tokens": 21220747.0,
      "step": 8370
    },
    {
      "entropy": 6.320551919937134,
      "epoch": 0.966532025389498,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004918199761144315,
      "loss": 6.0951,
      "mean_token_accuracy": 0.1732812151312828,
      "num_tokens": 21233229.0,
      "step": 8375
    },
    {
      "entropy": 6.2873331069946286,
      "epoch": 0.9671090594345066,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004918089471160318,
      "loss": 6.1428,
      "mean_token_accuracy": 0.16874834895133972,
      "num_tokens": 21244656.0,
      "step": 8380
    },
    {
      "entropy": 6.216814994812012,
      "epoch": 0.9676860934795153,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004917979108253709,
      "loss": 6.1974,
      "mean_token_accuracy": 0.15955638885498047,
      "num_tokens": 21257602.0,
      "step": 8385
    },
    {
      "entropy": 6.246206188201905,
      "epoch": 0.9682631275245239,
      "grad_norm": 1.34375,
      "learning_rate": 0.00049178686724282,
      "loss": 6.2014,
      "mean_token_accuracy": 0.16660555005073546,
      "num_tokens": 21270098.0,
      "step": 8390
    },
    {
      "entropy": 6.374057102203369,
      "epoch": 0.9688401615695326,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004917758163687506,
      "loss": 6.2461,
      "mean_token_accuracy": 0.15347524657845496,
      "num_tokens": 21283516.0,
      "step": 8395
    },
    {
      "entropy": 6.366953897476196,
      "epoch": 0.9694171956145412,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004917647582035341,
      "loss": 6.1651,
      "mean_token_accuracy": 0.1628739908337593,
      "num_tokens": 21296146.0,
      "step": 8400
    },
    {
      "entropy": 6.3211194515228275,
      "epoch": 0.96999422965955,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004917536927475428,
      "loss": 6.2524,
      "mean_token_accuracy": 0.16070077568292618,
      "num_tokens": 21308537.0,
      "step": 8405
    },
    {
      "entropy": 6.372178459167481,
      "epoch": 0.9705712637045586,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004917426200011486,
      "loss": 6.2715,
      "mean_token_accuracy": 0.16082310527563096,
      "num_tokens": 21322375.0,
      "step": 8410
    },
    {
      "entropy": 6.397395229339599,
      "epoch": 0.9711482977495672,
      "grad_norm": 1.046875,
      "learning_rate": 0.000491731539964724,
      "loss": 6.2409,
      "mean_token_accuracy": 0.1612374186515808,
      "num_tokens": 21335884.0,
      "step": 8415
    },
    {
      "entropy": 6.241150045394898,
      "epoch": 0.9717253317945759,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004917204526386418,
      "loss": 6.1115,
      "mean_token_accuracy": 0.16565245836973191,
      "num_tokens": 21349106.0,
      "step": 8420
    },
    {
      "entropy": 6.316123104095459,
      "epoch": 0.9723023658395845,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004917093580232748,
      "loss": 6.2414,
      "mean_token_accuracy": 0.16426511108875275,
      "num_tokens": 21362306.0,
      "step": 8425
    },
    {
      "entropy": 6.38222303390503,
      "epoch": 0.9728793998845932,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004916982561189962,
      "loss": 6.2486,
      "mean_token_accuracy": 0.1573673501610756,
      "num_tokens": 21375672.0,
      "step": 8430
    },
    {
      "entropy": 6.323233127593994,
      "epoch": 0.9734564339296018,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004916871469261794,
      "loss": 6.1773,
      "mean_token_accuracy": 0.17354203164577484,
      "num_tokens": 21389174.0,
      "step": 8435
    },
    {
      "entropy": 6.340617418289185,
      "epoch": 0.9740334679746105,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000491676030445198,
      "loss": 6.2898,
      "mean_token_accuracy": 0.15847488045692443,
      "num_tokens": 21400027.0,
      "step": 8440
    },
    {
      "entropy": 6.362000417709351,
      "epoch": 0.9746105020196192,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004916649066764259,
      "loss": 6.2342,
      "mean_token_accuracy": 0.15929642617702483,
      "num_tokens": 21412843.0,
      "step": 8445
    },
    {
      "entropy": 6.412792491912842,
      "epoch": 0.9751875360646278,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004916537756202375,
      "loss": 6.3397,
      "mean_token_accuracy": 0.15305596292018891,
      "num_tokens": 21425253.0,
      "step": 8450
    },
    {
      "entropy": 6.321174192428589,
      "epoch": 0.9757645701096365,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004916426372770069,
      "loss": 6.1258,
      "mean_token_accuracy": 0.17601382732391357,
      "num_tokens": 21436428.0,
      "step": 8455
    },
    {
      "entropy": 6.2951771259307865,
      "epoch": 0.9763416041546451,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004916314916471087,
      "loss": 6.2067,
      "mean_token_accuracy": 0.16642078757286072,
      "num_tokens": 21448680.0,
      "step": 8460
    },
    {
      "entropy": 6.264667272567749,
      "epoch": 0.9769186381996537,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004916203387309179,
      "loss": 6.204,
      "mean_token_accuracy": 0.1652453675866127,
      "num_tokens": 21460935.0,
      "step": 8465
    },
    {
      "entropy": 6.334538221359253,
      "epoch": 0.9774956722446624,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004916091785288096,
      "loss": 6.2299,
      "mean_token_accuracy": 0.16360235959291458,
      "num_tokens": 21473624.0,
      "step": 8470
    },
    {
      "entropy": 6.377848100662232,
      "epoch": 0.9780727062896711,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004915980110411593,
      "loss": 6.3284,
      "mean_token_accuracy": 0.15689454078674317,
      "num_tokens": 21486684.0,
      "step": 8475
    },
    {
      "entropy": 6.319025707244873,
      "epoch": 0.9786497403346798,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004915868362683425,
      "loss": 6.2004,
      "mean_token_accuracy": 0.1698843389749527,
      "num_tokens": 21500955.0,
      "step": 8480
    },
    {
      "entropy": 6.328025913238525,
      "epoch": 0.9792267743796884,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004915756542107349,
      "loss": 6.2506,
      "mean_token_accuracy": 0.15968093425035476,
      "num_tokens": 21512632.0,
      "step": 8485
    },
    {
      "entropy": 6.295421600341797,
      "epoch": 0.979803808424697,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004915644648687127,
      "loss": 6.1978,
      "mean_token_accuracy": 0.16501185446977615,
      "num_tokens": 21526455.0,
      "step": 8490
    },
    {
      "entropy": 6.326158094406128,
      "epoch": 0.9803808424697057,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004915532682426524,
      "loss": 6.2454,
      "mean_token_accuracy": 0.15768125504255295,
      "num_tokens": 21538594.0,
      "step": 8495
    },
    {
      "entropy": 6.243759202957153,
      "epoch": 0.9809578765147143,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004915420643329306,
      "loss": 6.1349,
      "mean_token_accuracy": 0.16712158620357515,
      "num_tokens": 21551071.0,
      "step": 8500
    },
    {
      "entropy": 6.39003357887268,
      "epoch": 0.981534910559723,
      "grad_norm": 1.21875,
      "learning_rate": 0.000491530853139924,
      "loss": 6.1551,
      "mean_token_accuracy": 0.16680120676755905,
      "num_tokens": 21563241.0,
      "step": 8505
    },
    {
      "entropy": 6.2129449367523195,
      "epoch": 0.9821119446047317,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004915196346640095,
      "loss": 6.1448,
      "mean_token_accuracy": 0.16390320360660554,
      "num_tokens": 21575634.0,
      "step": 8510
    },
    {
      "entropy": 6.2650598049163815,
      "epoch": 0.9826889786497404,
      "grad_norm": 1.125,
      "learning_rate": 0.0004915084089055648,
      "loss": 6.2168,
      "mean_token_accuracy": 0.16073915213346482,
      "num_tokens": 21587703.0,
      "step": 8515
    },
    {
      "entropy": 6.3416365623474125,
      "epoch": 0.983266012694749,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004914971758649673,
      "loss": 6.2315,
      "mean_token_accuracy": 0.15882400423288345,
      "num_tokens": 21599048.0,
      "step": 8520
    },
    {
      "entropy": 6.288963365554809,
      "epoch": 0.9838430467397576,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004914859355425948,
      "loss": 6.1319,
      "mean_token_accuracy": 0.16714104413986205,
      "num_tokens": 21611547.0,
      "step": 8525
    },
    {
      "entropy": 6.240753984451294,
      "epoch": 0.9844200807847663,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004914746879388255,
      "loss": 6.2007,
      "mean_token_accuracy": 0.17427244037389755,
      "num_tokens": 21623343.0,
      "step": 8530
    },
    {
      "entropy": 6.260480070114136,
      "epoch": 0.9849971148297749,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004914634330540373,
      "loss": 6.2289,
      "mean_token_accuracy": 0.1607411563396454,
      "num_tokens": 21635318.0,
      "step": 8535
    },
    {
      "entropy": 6.319377660751343,
      "epoch": 0.9855741488747836,
      "grad_norm": 1.25,
      "learning_rate": 0.0004914521708886093,
      "loss": 6.2315,
      "mean_token_accuracy": 0.15914136916399002,
      "num_tokens": 21647539.0,
      "step": 8540
    },
    {
      "entropy": 6.334693527221679,
      "epoch": 0.9861511829197923,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004914409014429201,
      "loss": 6.2336,
      "mean_token_accuracy": 0.16057218611240387,
      "num_tokens": 21660587.0,
      "step": 8545
    },
    {
      "entropy": 6.355124998092651,
      "epoch": 0.986728216964801,
      "grad_norm": 1.125,
      "learning_rate": 0.0004914296247173486,
      "loss": 6.2232,
      "mean_token_accuracy": 0.1634837031364441,
      "num_tokens": 21672123.0,
      "step": 8550
    },
    {
      "entropy": 6.302836322784424,
      "epoch": 0.9873052510098096,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004914183407122741,
      "loss": 6.1796,
      "mean_token_accuracy": 0.1616678774356842,
      "num_tokens": 21685430.0,
      "step": 8555
    },
    {
      "entropy": 6.417318344116211,
      "epoch": 0.9878822850548182,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004914070494280763,
      "loss": 6.2717,
      "mean_token_accuracy": 0.15696694552898408,
      "num_tokens": 21698462.0,
      "step": 8560
    },
    {
      "entropy": 6.3537040710449215,
      "epoch": 0.9884593190998269,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004913957508651349,
      "loss": 6.3265,
      "mean_token_accuracy": 0.15328820794820786,
      "num_tokens": 21711741.0,
      "step": 8565
    },
    {
      "entropy": 6.331461095809937,
      "epoch": 0.9890363531448355,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004913844450238299,
      "loss": 6.2201,
      "mean_token_accuracy": 0.16047517210245132,
      "num_tokens": 21723983.0,
      "step": 8570
    },
    {
      "entropy": 6.386488676071167,
      "epoch": 0.9896133871898442,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004913731319045416,
      "loss": 6.2686,
      "mean_token_accuracy": 0.16281691789627076,
      "num_tokens": 21736399.0,
      "step": 8575
    },
    {
      "entropy": 6.364957475662232,
      "epoch": 0.9901904212348529,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004913618115076505,
      "loss": 6.263,
      "mean_token_accuracy": 0.15798835158348085,
      "num_tokens": 21748705.0,
      "step": 8580
    },
    {
      "entropy": 6.245175218582153,
      "epoch": 0.9907674552798615,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004913504838335372,
      "loss": 6.1552,
      "mean_token_accuracy": 0.17042683959007263,
      "num_tokens": 21761659.0,
      "step": 8585
    },
    {
      "entropy": 6.343247652053833,
      "epoch": 0.9913444893248702,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004913391488825829,
      "loss": 6.1897,
      "mean_token_accuracy": 0.1614478461444378,
      "num_tokens": 21774217.0,
      "step": 8590
    },
    {
      "entropy": 6.29022068977356,
      "epoch": 0.9919215233698788,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004913278066551689,
      "loss": 6.2356,
      "mean_token_accuracy": 0.16395027786493302,
      "num_tokens": 21787417.0,
      "step": 8595
    },
    {
      "entropy": 6.338612461090088,
      "epoch": 0.9924985574148875,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004913164571516765,
      "loss": 6.2745,
      "mean_token_accuracy": 0.16423107236623763,
      "num_tokens": 21798869.0,
      "step": 8600
    },
    {
      "entropy": 6.32934832572937,
      "epoch": 0.9930755914598961,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004913051003724876,
      "loss": 6.189,
      "mean_token_accuracy": 0.16142335683107376,
      "num_tokens": 21811622.0,
      "step": 8605
    },
    {
      "entropy": 6.331812238693237,
      "epoch": 0.9936526255049047,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004912937363179839,
      "loss": 6.1793,
      "mean_token_accuracy": 0.16208837628364564,
      "num_tokens": 21823401.0,
      "step": 8610
    },
    {
      "entropy": 6.285097980499268,
      "epoch": 0.9942296595499135,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000491282364988548,
      "loss": 6.1986,
      "mean_token_accuracy": 0.16630493700504304,
      "num_tokens": 21835460.0,
      "step": 8615
    },
    {
      "entropy": 6.2709862232208256,
      "epoch": 0.9948066935949221,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004912709863845621,
      "loss": 6.1456,
      "mean_token_accuracy": 0.17110171914100647,
      "num_tokens": 21847521.0,
      "step": 8620
    },
    {
      "entropy": 6.3518126010894775,
      "epoch": 0.9953837276399308,
      "grad_norm": 1.078125,
      "learning_rate": 0.000491259600506409,
      "loss": 6.3013,
      "mean_token_accuracy": 0.15523147135972976,
      "num_tokens": 21860861.0,
      "step": 8625
    },
    {
      "entropy": 6.321661376953125,
      "epoch": 0.9959607616849394,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004912482073544716,
      "loss": 6.1294,
      "mean_token_accuracy": 0.17137608230113982,
      "num_tokens": 21872380.0,
      "step": 8630
    },
    {
      "entropy": 6.254619026184082,
      "epoch": 0.996537795729948,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004912368069291333,
      "loss": 6.2209,
      "mean_token_accuracy": 0.1629670575261116,
      "num_tokens": 21885322.0,
      "step": 8635
    },
    {
      "entropy": 6.439844274520874,
      "epoch": 0.9971148297749567,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004912253992307773,
      "loss": 6.2893,
      "mean_token_accuracy": 0.15910178869962693,
      "num_tokens": 21898243.0,
      "step": 8640
    },
    {
      "entropy": 6.37636399269104,
      "epoch": 0.9976918638199653,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004912139842597875,
      "loss": 6.2389,
      "mean_token_accuracy": 0.15896237641572952,
      "num_tokens": 21911467.0,
      "step": 8645
    },
    {
      "entropy": 6.350759124755859,
      "epoch": 0.9982688978649741,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004912025620165477,
      "loss": 6.2533,
      "mean_token_accuracy": 0.16140036433935165,
      "num_tokens": 21925583.0,
      "step": 8650
    },
    {
      "entropy": 6.308727836608886,
      "epoch": 0.9988459319099827,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004911911325014421,
      "loss": 6.2706,
      "mean_token_accuracy": 0.15814359188079835,
      "num_tokens": 21937627.0,
      "step": 8655
    },
    {
      "entropy": 6.446664237976075,
      "epoch": 0.9994229659549914,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004911796957148552,
      "loss": 6.2842,
      "mean_token_accuracy": 0.15748122632503508,
      "num_tokens": 21950376.0,
      "step": 8660
    },
    {
      "entropy": 6.334322547912597,
      "epoch": 1.0,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004911682516571715,
      "loss": 6.2247,
      "mean_token_accuracy": 0.16553265005350112,
      "num_tokens": 21963117.0,
      "step": 8665
    },
    {
      "entropy": 6.309478378295898,
      "epoch": 1.0005770340450086,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004911568003287761,
      "loss": 6.0776,
      "mean_token_accuracy": 0.17455263435840607,
      "num_tokens": 21976256.0,
      "step": 8670
    },
    {
      "entropy": 6.346786022186279,
      "epoch": 1.0011540680900173,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004911453417300541,
      "loss": 6.095,
      "mean_token_accuracy": 0.15989951491355897,
      "num_tokens": 21988768.0,
      "step": 8675
    },
    {
      "entropy": 6.260541534423828,
      "epoch": 1.001731102135026,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004911338758613909,
      "loss": 6.0913,
      "mean_token_accuracy": 0.16880803257226945,
      "num_tokens": 22000570.0,
      "step": 8680
    },
    {
      "entropy": 6.31554388999939,
      "epoch": 1.0023081361800346,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004911224027231722,
      "loss": 6.1507,
      "mean_token_accuracy": 0.15978237390518188,
      "num_tokens": 22012914.0,
      "step": 8685
    },
    {
      "entropy": 6.359015321731567,
      "epoch": 1.0028851702250432,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004911109223157838,
      "loss": 6.1508,
      "mean_token_accuracy": 0.15542728900909425,
      "num_tokens": 22025474.0,
      "step": 8690
    },
    {
      "entropy": 6.219356346130371,
      "epoch": 1.0034622042700518,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004910994346396118,
      "loss": 6.0048,
      "mean_token_accuracy": 0.16527428179979325,
      "num_tokens": 22039228.0,
      "step": 8695
    },
    {
      "entropy": 6.224536752700805,
      "epoch": 1.0040392383150605,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004910879396950427,
      "loss": 6.1586,
      "mean_token_accuracy": 0.16529181897640227,
      "num_tokens": 22050153.0,
      "step": 8700
    },
    {
      "entropy": 6.333424806594849,
      "epoch": 1.0046162723600693,
      "grad_norm": 1.140625,
      "learning_rate": 0.000491076437482463,
      "loss": 6.1838,
      "mean_token_accuracy": 0.15986959934234618,
      "num_tokens": 22063214.0,
      "step": 8705
    },
    {
      "entropy": 6.265970230102539,
      "epoch": 1.005193306405078,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004910649280022599,
      "loss": 6.0611,
      "mean_token_accuracy": 0.16764540821313859,
      "num_tokens": 22076032.0,
      "step": 8710
    },
    {
      "entropy": 6.249867916107178,
      "epoch": 1.0057703404500866,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004910534112548201,
      "loss": 6.1,
      "mean_token_accuracy": 0.16600306779146196,
      "num_tokens": 22089134.0,
      "step": 8715
    },
    {
      "entropy": 6.306376981735229,
      "epoch": 1.0063473744950953,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004910418872405312,
      "loss": 6.1281,
      "mean_token_accuracy": 0.16220694929361343,
      "num_tokens": 22102772.0,
      "step": 8720
    },
    {
      "entropy": 6.278018474578857,
      "epoch": 1.006924408540104,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004910303559597806,
      "loss": 6.1305,
      "mean_token_accuracy": 0.16888279616832733,
      "num_tokens": 22117586.0,
      "step": 8725
    },
    {
      "entropy": 6.247881555557251,
      "epoch": 1.0075014425851125,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004910188174129564,
      "loss": 6.1823,
      "mean_token_accuracy": 0.16208623498678207,
      "num_tokens": 22129229.0,
      "step": 8730
    },
    {
      "entropy": 6.324610042572021,
      "epoch": 1.0080784766301212,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004910072716004466,
      "loss": 6.1288,
      "mean_token_accuracy": 0.1664586067199707,
      "num_tokens": 22142269.0,
      "step": 8735
    },
    {
      "entropy": 6.336810541152954,
      "epoch": 1.0086555106751298,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004909957185226394,
      "loss": 6.1291,
      "mean_token_accuracy": 0.16251941323280333,
      "num_tokens": 22153744.0,
      "step": 8740
    },
    {
      "entropy": 6.160395669937134,
      "epoch": 1.0092325447201385,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004909841581799236,
      "loss": 6.0261,
      "mean_token_accuracy": 0.16367970556020736,
      "num_tokens": 22166927.0,
      "step": 8745
    },
    {
      "entropy": 6.297188758850098,
      "epoch": 1.009809578765147,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004909725905726879,
      "loss": 6.1384,
      "mean_token_accuracy": 0.16302257776260376,
      "num_tokens": 22179003.0,
      "step": 8750
    },
    {
      "entropy": 6.3528547286987305,
      "epoch": 1.0103866128101557,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004909610157013214,
      "loss": 6.1809,
      "mean_token_accuracy": 0.1637558862566948,
      "num_tokens": 22191221.0,
      "step": 8755
    },
    {
      "entropy": 6.328153991699219,
      "epoch": 1.0109636468551644,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004909494335662134,
      "loss": 6.1504,
      "mean_token_accuracy": 0.16704724431037904,
      "num_tokens": 22203387.0,
      "step": 8760
    },
    {
      "entropy": 6.313574647903442,
      "epoch": 1.011540680900173,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004909378441677535,
      "loss": 6.0921,
      "mean_token_accuracy": 0.16964887231588363,
      "num_tokens": 22215225.0,
      "step": 8765
    },
    {
      "entropy": 6.300811529159546,
      "epoch": 1.0121177149451817,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004909262475063314,
      "loss": 6.0742,
      "mean_token_accuracy": 0.16994739025831224,
      "num_tokens": 22226390.0,
      "step": 8770
    },
    {
      "entropy": 6.25765323638916,
      "epoch": 1.0126947489901905,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004909146435823373,
      "loss": 6.1262,
      "mean_token_accuracy": 0.1657228872179985,
      "num_tokens": 22238884.0,
      "step": 8775
    },
    {
      "entropy": 6.267900133132935,
      "epoch": 1.0132717830351992,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004909030323961613,
      "loss": 6.2608,
      "mean_token_accuracy": 0.15375184863805771,
      "num_tokens": 22252214.0,
      "step": 8780
    },
    {
      "entropy": 6.364380025863648,
      "epoch": 1.0138488170802078,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004908914139481942,
      "loss": 6.1345,
      "mean_token_accuracy": 0.16241769790649413,
      "num_tokens": 22264814.0,
      "step": 8785
    },
    {
      "entropy": 6.255403661727906,
      "epoch": 1.0144258511252164,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004908797882388265,
      "loss": 6.0823,
      "mean_token_accuracy": 0.17427722215652466,
      "num_tokens": 22278737.0,
      "step": 8790
    },
    {
      "entropy": 6.252494049072266,
      "epoch": 1.015002885170225,
      "grad_norm": 1.125,
      "learning_rate": 0.0004908681552684495,
      "loss": 6.1359,
      "mean_token_accuracy": 0.16039522141218185,
      "num_tokens": 22290974.0,
      "step": 8795
    },
    {
      "entropy": 6.304461145401001,
      "epoch": 1.0155799192152337,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004908565150374542,
      "loss": 6.1196,
      "mean_token_accuracy": 0.1662422925233841,
      "num_tokens": 22303080.0,
      "step": 8800
    },
    {
      "entropy": 6.285967683792114,
      "epoch": 1.0161569532602424,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004908448675462323,
      "loss": 6.1761,
      "mean_token_accuracy": 0.16349056363105774,
      "num_tokens": 22315536.0,
      "step": 8805
    },
    {
      "entropy": 6.262104654312134,
      "epoch": 1.016733987305251,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004908332127951756,
      "loss": 6.1356,
      "mean_token_accuracy": 0.1671780079603195,
      "num_tokens": 22328114.0,
      "step": 8810
    },
    {
      "entropy": 6.292828702926636,
      "epoch": 1.0173110213502596,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004908215507846757,
      "loss": 6.0709,
      "mean_token_accuracy": 0.16945097744464874,
      "num_tokens": 22339608.0,
      "step": 8815
    },
    {
      "entropy": 6.317346429824829,
      "epoch": 1.0178880553952683,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004908098815151254,
      "loss": 6.1003,
      "mean_token_accuracy": 0.16731197535991668,
      "num_tokens": 22351594.0,
      "step": 8820
    },
    {
      "entropy": 6.26181468963623,
      "epoch": 1.018465089440277,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004907982049869168,
      "loss": 6.1898,
      "mean_token_accuracy": 0.1664214164018631,
      "num_tokens": 22363978.0,
      "step": 8825
    },
    {
      "entropy": 6.362544250488281,
      "epoch": 1.0190421234852856,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004907865212004428,
      "loss": 6.1849,
      "mean_token_accuracy": 0.16163639798760415,
      "num_tokens": 22377845.0,
      "step": 8830
    },
    {
      "entropy": 6.296783828735352,
      "epoch": 1.0196191575302942,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004907748301560963,
      "loss": 6.1164,
      "mean_token_accuracy": 0.1656057432293892,
      "num_tokens": 22392654.0,
      "step": 8835
    },
    {
      "entropy": 6.3022528171539305,
      "epoch": 1.0201961915753028,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004907631318542707,
      "loss": 6.1444,
      "mean_token_accuracy": 0.16330374479293824,
      "num_tokens": 22404713.0,
      "step": 8840
    },
    {
      "entropy": 6.352923202514648,
      "epoch": 1.0207732256203117,
      "grad_norm": 1.125,
      "learning_rate": 0.0004907514262953594,
      "loss": 6.1656,
      "mean_token_accuracy": 0.1557633936405182,
      "num_tokens": 22417879.0,
      "step": 8845
    },
    {
      "entropy": 6.2035558223724365,
      "epoch": 1.0213502596653203,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004907397134797559,
      "loss": 6.1195,
      "mean_token_accuracy": 0.16064710915088654,
      "num_tokens": 22430674.0,
      "step": 8850
    },
    {
      "entropy": 6.3108683109283445,
      "epoch": 1.021927293710329,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004907279934078543,
      "loss": 6.1097,
      "mean_token_accuracy": 0.16840347200632094,
      "num_tokens": 22443135.0,
      "step": 8855
    },
    {
      "entropy": 6.248948192596435,
      "epoch": 1.0225043277553376,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004907162660800488,
      "loss": 6.0981,
      "mean_token_accuracy": 0.16608906984329225,
      "num_tokens": 22455832.0,
      "step": 8860
    },
    {
      "entropy": 6.318018245697021,
      "epoch": 1.0230813618003463,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004907045314967338,
      "loss": 6.1316,
      "mean_token_accuracy": 0.16259942948818207,
      "num_tokens": 22467466.0,
      "step": 8865
    },
    {
      "entropy": 6.341202020645142,
      "epoch": 1.023658395845355,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004906927896583041,
      "loss": 6.1748,
      "mean_token_accuracy": 0.15842344015836715,
      "num_tokens": 22481127.0,
      "step": 8870
    },
    {
      "entropy": 6.282949018478393,
      "epoch": 1.0242354298903635,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004906810405651546,
      "loss": 6.1698,
      "mean_token_accuracy": 0.16103839129209518,
      "num_tokens": 22492894.0,
      "step": 8875
    },
    {
      "entropy": 6.4367516994476315,
      "epoch": 1.0248124639353722,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004906692842176803,
      "loss": 6.1719,
      "mean_token_accuracy": 0.15828175619244575,
      "num_tokens": 22505369.0,
      "step": 8880
    },
    {
      "entropy": 6.236190700531006,
      "epoch": 1.0253894979803808,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004906575206162769,
      "loss": 6.0738,
      "mean_token_accuracy": 0.16713444739580155,
      "num_tokens": 22517163.0,
      "step": 8885
    },
    {
      "entropy": 6.241188812255859,
      "epoch": 1.0259665320253895,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004906457497613399,
      "loss": 6.1704,
      "mean_token_accuracy": 0.16040320992469786,
      "num_tokens": 22531132.0,
      "step": 8890
    },
    {
      "entropy": 6.401804399490357,
      "epoch": 1.026543566070398,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004906339716532651,
      "loss": 6.1379,
      "mean_token_accuracy": 0.15718916058540344,
      "num_tokens": 22542663.0,
      "step": 8895
    },
    {
      "entropy": 6.286692190170288,
      "epoch": 1.0271206001154067,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004906221862924488,
      "loss": 6.1492,
      "mean_token_accuracy": 0.1643242284655571,
      "num_tokens": 22555397.0,
      "step": 8900
    },
    {
      "entropy": 6.157395076751709,
      "epoch": 1.0276976341604154,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004906103936792875,
      "loss": 6.0891,
      "mean_token_accuracy": 0.16359584778547287,
      "num_tokens": 22567373.0,
      "step": 8905
    },
    {
      "entropy": 6.354315376281738,
      "epoch": 1.028274668205424,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004905985938141777,
      "loss": 6.1054,
      "mean_token_accuracy": 0.16490954756736756,
      "num_tokens": 22579200.0,
      "step": 8910
    },
    {
      "entropy": 6.302970886230469,
      "epoch": 1.0288517022504329,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004905867866975163,
      "loss": 6.1621,
      "mean_token_accuracy": 0.16215419471263887,
      "num_tokens": 22592940.0,
      "step": 8915
    },
    {
      "entropy": 6.2957353591918945,
      "epoch": 1.0294287362954415,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004905749723297003,
      "loss": 6.1366,
      "mean_token_accuracy": 0.16920628994703293,
      "num_tokens": 22605440.0,
      "step": 8920
    },
    {
      "entropy": 6.305857753753662,
      "epoch": 1.0300057703404502,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004905631507111271,
      "loss": 6.1287,
      "mean_token_accuracy": 0.16817355304956436,
      "num_tokens": 22618976.0,
      "step": 8925
    },
    {
      "entropy": 6.24271330833435,
      "epoch": 1.0305828043854588,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004905513218421946,
      "loss": 6.0175,
      "mean_token_accuracy": 0.1673113763332367,
      "num_tokens": 22632356.0,
      "step": 8930
    },
    {
      "entropy": 6.238995218276978,
      "epoch": 1.0311598384304674,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004905394857233004,
      "loss": 6.1304,
      "mean_token_accuracy": 0.16247055381536485,
      "num_tokens": 22645272.0,
      "step": 8935
    },
    {
      "entropy": 6.303689956665039,
      "epoch": 1.031736872475476,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004905276423548426,
      "loss": 6.0957,
      "mean_token_accuracy": 0.1600155919790268,
      "num_tokens": 22657101.0,
      "step": 8940
    },
    {
      "entropy": 6.28863034248352,
      "epoch": 1.0323139065204847,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004905157917372197,
      "loss": 6.0686,
      "mean_token_accuracy": 0.16965386569499968,
      "num_tokens": 22669107.0,
      "step": 8945
    },
    {
      "entropy": 6.2883241176605225,
      "epoch": 1.0328909405654934,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004905039338708302,
      "loss": 6.1221,
      "mean_token_accuracy": 0.15880545377731323,
      "num_tokens": 22681633.0,
      "step": 8950
    },
    {
      "entropy": 6.325016736984253,
      "epoch": 1.033467974610502,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004904920687560728,
      "loss": 6.1793,
      "mean_token_accuracy": 0.16751186102628707,
      "num_tokens": 22694985.0,
      "step": 8955
    },
    {
      "entropy": 6.225956583023072,
      "epoch": 1.0340450086555106,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004904801963933469,
      "loss": 6.1221,
      "mean_token_accuracy": 0.16137125790119172,
      "num_tokens": 22707333.0,
      "step": 8960
    },
    {
      "entropy": 6.293618535995483,
      "epoch": 1.0346220427005193,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004904683167830515,
      "loss": 6.1102,
      "mean_token_accuracy": 0.16506971418857574,
      "num_tokens": 22721030.0,
      "step": 8965
    },
    {
      "entropy": 6.251954460144043,
      "epoch": 1.035199076745528,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004904564299255862,
      "loss": 6.0681,
      "mean_token_accuracy": 0.17179013937711715,
      "num_tokens": 22732711.0,
      "step": 8970
    },
    {
      "entropy": 6.21856575012207,
      "epoch": 1.0357761107905366,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004904445358213511,
      "loss": 6.0968,
      "mean_token_accuracy": 0.17304373681545257,
      "num_tokens": 22745815.0,
      "step": 8975
    },
    {
      "entropy": 6.272159576416016,
      "epoch": 1.0363531448355452,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004904326344707461,
      "loss": 6.1262,
      "mean_token_accuracy": 0.16413632929325103,
      "num_tokens": 22757590.0,
      "step": 8980
    },
    {
      "entropy": 6.380202913284302,
      "epoch": 1.036930178880554,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004904207258741712,
      "loss": 6.1848,
      "mean_token_accuracy": 0.15278932005167006,
      "num_tokens": 22769841.0,
      "step": 8985
    },
    {
      "entropy": 6.200816965103149,
      "epoch": 1.0375072129255627,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004904088100320274,
      "loss": 6.0861,
      "mean_token_accuracy": 0.17516712099313736,
      "num_tokens": 22781642.0,
      "step": 8990
    },
    {
      "entropy": 6.326822280883789,
      "epoch": 1.0380842469705713,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004903968869447151,
      "loss": 6.1696,
      "mean_token_accuracy": 0.16336700171232224,
      "num_tokens": 22794903.0,
      "step": 8995
    },
    {
      "entropy": 6.281322622299195,
      "epoch": 1.03866128101558,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004903849566126356,
      "loss": 6.1596,
      "mean_token_accuracy": 0.16943808048963546,
      "num_tokens": 22807839.0,
      "step": 9000
    },
    {
      "epoch": 1.03866128101558,
      "eval_entropy": 6.070269384131976,
      "eval_loss": 6.286910057067871,
      "eval_mean_token_accuracy": 0.16314482978601533,
      "eval_num_tokens": 22807839.0,
      "eval_runtime": 23.1489,
      "eval_samples_per_second": 1215.436,
      "eval_steps_per_second": 151.929,
      "step": 9000
    },
    {
      "entropy": 6.2383894443511965,
      "epoch": 1.0392383150605886,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004903730190361902,
      "loss": 6.197,
      "mean_token_accuracy": 0.15852217376232147,
      "num_tokens": 22821037.0,
      "step": 9005
    },
    {
      "entropy": 6.3174175262451175,
      "epoch": 1.0398153491055973,
      "grad_norm": 1.09375,
      "learning_rate": 0.00049036107421578,
      "loss": 6.1662,
      "mean_token_accuracy": 0.1641897365450859,
      "num_tokens": 22834225.0,
      "step": 9010
    },
    {
      "entropy": 6.289822626113891,
      "epoch": 1.040392383150606,
      "grad_norm": 1.125,
      "learning_rate": 0.0004903491221518073,
      "loss": 6.1314,
      "mean_token_accuracy": 0.17093408107757568,
      "num_tokens": 22847355.0,
      "step": 9015
    },
    {
      "entropy": 6.222614717483521,
      "epoch": 1.0409694171956145,
      "grad_norm": 1.25,
      "learning_rate": 0.0004903371628446737,
      "loss": 6.1047,
      "mean_token_accuracy": 0.16839305609464644,
      "num_tokens": 22859550.0,
      "step": 9020
    },
    {
      "entropy": 6.334166049957275,
      "epoch": 1.0415464512406232,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004903251962947817,
      "loss": 6.196,
      "mean_token_accuracy": 0.15899620652198793,
      "num_tokens": 22871690.0,
      "step": 9025
    },
    {
      "entropy": 6.329074335098267,
      "epoch": 1.0421234852856318,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004903132225025335,
      "loss": 6.2028,
      "mean_token_accuracy": 0.16065797060728074,
      "num_tokens": 22883838.0,
      "step": 9030
    },
    {
      "entropy": 6.208608150482178,
      "epoch": 1.0427005193306405,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004903012414683322,
      "loss": 6.0926,
      "mean_token_accuracy": 0.16543340235948562,
      "num_tokens": 22896566.0,
      "step": 9035
    },
    {
      "entropy": 6.32443528175354,
      "epoch": 1.043277553375649,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004902892531925805,
      "loss": 6.0914,
      "mean_token_accuracy": 0.1645181432366371,
      "num_tokens": 22908480.0,
      "step": 9040
    },
    {
      "entropy": 6.265196180343628,
      "epoch": 1.0438545874206577,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004902772576756818,
      "loss": 6.1623,
      "mean_token_accuracy": 0.16638606935739517,
      "num_tokens": 22920923.0,
      "step": 9045
    },
    {
      "entropy": 6.338992023468018,
      "epoch": 1.0444316214656664,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004902652549180394,
      "loss": 6.2397,
      "mean_token_accuracy": 0.156393101811409,
      "num_tokens": 22934991.0,
      "step": 9050
    },
    {
      "entropy": 6.254326629638672,
      "epoch": 1.0450086555106752,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004902532449200571,
      "loss": 6.0953,
      "mean_token_accuracy": 0.16595544517040253,
      "num_tokens": 22947223.0,
      "step": 9055
    },
    {
      "entropy": 6.283085680007934,
      "epoch": 1.0455856895556839,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004902412276821386,
      "loss": 6.1898,
      "mean_token_accuracy": 0.1656716287136078,
      "num_tokens": 22960320.0,
      "step": 9060
    },
    {
      "entropy": 6.302255249023437,
      "epoch": 1.0461627236006925,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004902292032046887,
      "loss": 6.1107,
      "mean_token_accuracy": 0.16879165321588516,
      "num_tokens": 22972373.0,
      "step": 9065
    },
    {
      "entropy": 6.34630012512207,
      "epoch": 1.0467397576457012,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004902171714881112,
      "loss": 6.1964,
      "mean_token_accuracy": 0.15887296944856644,
      "num_tokens": 22984936.0,
      "step": 9070
    },
    {
      "entropy": 6.321364545822144,
      "epoch": 1.0473167916907098,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004902051325328112,
      "loss": 6.1023,
      "mean_token_accuracy": 0.16771287918090821,
      "num_tokens": 22997027.0,
      "step": 9075
    },
    {
      "entropy": 6.288811445236206,
      "epoch": 1.0478938257357184,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004901930863391934,
      "loss": 6.1299,
      "mean_token_accuracy": 0.16606606245040895,
      "num_tokens": 23010861.0,
      "step": 9080
    },
    {
      "entropy": 6.261023235321045,
      "epoch": 1.048470859780727,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000490181032907663,
      "loss": 6.2237,
      "mean_token_accuracy": 0.15730314403772355,
      "num_tokens": 23023654.0,
      "step": 9085
    },
    {
      "entropy": 6.37444109916687,
      "epoch": 1.0490478938257357,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004901689722386255,
      "loss": 6.1958,
      "mean_token_accuracy": 0.16093336045742035,
      "num_tokens": 23036401.0,
      "step": 9090
    },
    {
      "entropy": 6.353020429611206,
      "epoch": 1.0496249278707444,
      "grad_norm": 1.25,
      "learning_rate": 0.0004901569043324864,
      "loss": 6.1324,
      "mean_token_accuracy": 0.16288792937994004,
      "num_tokens": 23048891.0,
      "step": 9095
    },
    {
      "entropy": 6.243642950057984,
      "epoch": 1.050201961915753,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004901448291896518,
      "loss": 6.1723,
      "mean_token_accuracy": 0.1600377231836319,
      "num_tokens": 23062170.0,
      "step": 9100
    },
    {
      "entropy": 6.298622751235962,
      "epoch": 1.0507789959607616,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004901327468105277,
      "loss": 6.118,
      "mean_token_accuracy": 0.16497347503900528,
      "num_tokens": 23075512.0,
      "step": 9105
    },
    {
      "entropy": 6.308530998229981,
      "epoch": 1.0513560300057703,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004901206571955205,
      "loss": 6.1879,
      "mean_token_accuracy": 0.16107307523488998,
      "num_tokens": 23086644.0,
      "step": 9110
    },
    {
      "entropy": 6.294368124008178,
      "epoch": 1.051933064050779,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004901085603450369,
      "loss": 6.0432,
      "mean_token_accuracy": 0.17177970707416534,
      "num_tokens": 23098596.0,
      "step": 9115
    },
    {
      "entropy": 6.25173134803772,
      "epoch": 1.0525100980957875,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004900964562594838,
      "loss": 6.1197,
      "mean_token_accuracy": 0.16303024291992188,
      "num_tokens": 23112746.0,
      "step": 9120
    },
    {
      "entropy": 6.274853372573853,
      "epoch": 1.0530871321407964,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000490084344939268,
      "loss": 6.0495,
      "mean_token_accuracy": 0.16825281083583832,
      "num_tokens": 23124297.0,
      "step": 9125
    },
    {
      "entropy": 6.277349328994751,
      "epoch": 1.053664166185805,
      "grad_norm": 1.125,
      "learning_rate": 0.0004900722263847973,
      "loss": 6.0671,
      "mean_token_accuracy": 0.16575167775154115,
      "num_tokens": 23135702.0,
      "step": 9130
    },
    {
      "entropy": 6.216898632049561,
      "epoch": 1.0542412002308137,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004900601005964791,
      "loss": 6.0688,
      "mean_token_accuracy": 0.1649308383464813,
      "num_tokens": 23149055.0,
      "step": 9135
    },
    {
      "entropy": 6.186308765411377,
      "epoch": 1.0548182342758223,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004900479675747212,
      "loss": 6.0754,
      "mean_token_accuracy": 0.1714058741927147,
      "num_tokens": 23162432.0,
      "step": 9140
    },
    {
      "entropy": 6.295329332351685,
      "epoch": 1.055395268320831,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004900358273199316,
      "loss": 6.1395,
      "mean_token_accuracy": 0.15748925060033797,
      "num_tokens": 23174187.0,
      "step": 9145
    },
    {
      "entropy": 6.251478481292724,
      "epoch": 1.0559723023658396,
      "grad_norm": 1.234375,
      "learning_rate": 0.000490023679832519,
      "loss": 6.0934,
      "mean_token_accuracy": 0.16925156861543655,
      "num_tokens": 23186584.0,
      "step": 9150
    },
    {
      "entropy": 6.159129667282104,
      "epoch": 1.0565493364108482,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004900115251128916,
      "loss": 6.049,
      "mean_token_accuracy": 0.16758615523576736,
      "num_tokens": 23197804.0,
      "step": 9155
    },
    {
      "entropy": 6.327369213104248,
      "epoch": 1.057126370455857,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004899993631614583,
      "loss": 6.0868,
      "mean_token_accuracy": 0.17150183916091918,
      "num_tokens": 23210262.0,
      "step": 9160
    },
    {
      "entropy": 6.282505846023559,
      "epoch": 1.0577034045008655,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004899871939786283,
      "loss": 6.1064,
      "mean_token_accuracy": 0.1674810379743576,
      "num_tokens": 23221922.0,
      "step": 9165
    },
    {
      "entropy": 6.205162906646729,
      "epoch": 1.0582804385458742,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004899750175648107,
      "loss": 6.0736,
      "mean_token_accuracy": 0.1726841449737549,
      "num_tokens": 23232983.0,
      "step": 9170
    },
    {
      "entropy": 6.269182205200195,
      "epoch": 1.0588574725908828,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004899628339204154,
      "loss": 6.1784,
      "mean_token_accuracy": 0.15772062093019484,
      "num_tokens": 23245692.0,
      "step": 9175
    },
    {
      "entropy": 6.369002723693848,
      "epoch": 1.0594345066358914,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004899506430458518,
      "loss": 6.0491,
      "mean_token_accuracy": 0.16568606942892075,
      "num_tokens": 23256963.0,
      "step": 9180
    },
    {
      "entropy": 6.190002202987671,
      "epoch": 1.0600115406809,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004899384449415302,
      "loss": 6.101,
      "mean_token_accuracy": 0.16143595725297927,
      "num_tokens": 23270226.0,
      "step": 9185
    },
    {
      "entropy": 6.249382543563843,
      "epoch": 1.0605885747259087,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004899262396078606,
      "loss": 6.2042,
      "mean_token_accuracy": 0.1554243430495262,
      "num_tokens": 23283666.0,
      "step": 9190
    },
    {
      "entropy": 6.35944676399231,
      "epoch": 1.0611656087709176,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004899140270452539,
      "loss": 6.1244,
      "mean_token_accuracy": 0.17235365360975266,
      "num_tokens": 23295499.0,
      "step": 9195
    },
    {
      "entropy": 6.268174695968628,
      "epoch": 1.0617426428159262,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004899018072541204,
      "loss": 6.1251,
      "mean_token_accuracy": 0.16058170050382614,
      "num_tokens": 23308876.0,
      "step": 9200
    },
    {
      "entropy": 6.288631200790405,
      "epoch": 1.0623196768609349,
      "grad_norm": 1.125,
      "learning_rate": 0.0004898895802348715,
      "loss": 6.1524,
      "mean_token_accuracy": 0.16507319062948228,
      "num_tokens": 23321678.0,
      "step": 9205
    },
    {
      "entropy": 6.2690895080566404,
      "epoch": 1.0628967109059435,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004898773459879183,
      "loss": 6.0968,
      "mean_token_accuracy": 0.17374747693538667,
      "num_tokens": 23334775.0,
      "step": 9210
    },
    {
      "entropy": 6.203541803359985,
      "epoch": 1.0634737449509521,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004898651045136724,
      "loss": 6.0134,
      "mean_token_accuracy": 0.17150746285915375,
      "num_tokens": 23347874.0,
      "step": 9215
    },
    {
      "entropy": 6.2491126537323,
      "epoch": 1.0640507789959608,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004898528558125453,
      "loss": 6.1168,
      "mean_token_accuracy": 0.16273091435432435,
      "num_tokens": 23360150.0,
      "step": 9220
    },
    {
      "entropy": 6.266243267059326,
      "epoch": 1.0646278130409694,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004898405998849492,
      "loss": 6.1648,
      "mean_token_accuracy": 0.16678090393543243,
      "num_tokens": 23373324.0,
      "step": 9225
    },
    {
      "entropy": 6.293151617050171,
      "epoch": 1.065204847085978,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004898283367312962,
      "loss": 6.1926,
      "mean_token_accuracy": 0.15916960835456848,
      "num_tokens": 23385576.0,
      "step": 9230
    },
    {
      "entropy": 6.2894689559936525,
      "epoch": 1.0657818811309867,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004898160663519988,
      "loss": 6.0836,
      "mean_token_accuracy": 0.1684999942779541,
      "num_tokens": 23396793.0,
      "step": 9235
    },
    {
      "entropy": 6.331593465805054,
      "epoch": 1.0663589151759953,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004898037887474697,
      "loss": 6.1565,
      "mean_token_accuracy": 0.16632169783115386,
      "num_tokens": 23410034.0,
      "step": 9240
    },
    {
      "entropy": 6.2295318126678465,
      "epoch": 1.066935949221004,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004897915039181219,
      "loss": 6.1089,
      "mean_token_accuracy": 0.16505092531442642,
      "num_tokens": 23423857.0,
      "step": 9245
    },
    {
      "entropy": 6.272892618179322,
      "epoch": 1.0675129832660126,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004897792118643685,
      "loss": 6.1677,
      "mean_token_accuracy": 0.16611825972795485,
      "num_tokens": 23437075.0,
      "step": 9250
    },
    {
      "entropy": 6.27429347038269,
      "epoch": 1.0680900173110213,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004897669125866231,
      "loss": 6.1801,
      "mean_token_accuracy": 0.1603154718875885,
      "num_tokens": 23450635.0,
      "step": 9255
    },
    {
      "entropy": 6.2924275398254395,
      "epoch": 1.06866705135603,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004897546060852993,
      "loss": 6.0858,
      "mean_token_accuracy": 0.16646239161491394,
      "num_tokens": 23464341.0,
      "step": 9260
    },
    {
      "entropy": 6.385766172409058,
      "epoch": 1.0692440854010385,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004897422923608108,
      "loss": 6.2204,
      "mean_token_accuracy": 0.15758515372872353,
      "num_tokens": 23477462.0,
      "step": 9265
    },
    {
      "entropy": 6.293824529647827,
      "epoch": 1.0698211194460474,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004897299714135721,
      "loss": 6.2119,
      "mean_token_accuracy": 0.15515773445367814,
      "num_tokens": 23489990.0,
      "step": 9270
    },
    {
      "entropy": 6.288429689407349,
      "epoch": 1.070398153491056,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004897176432439974,
      "loss": 6.1589,
      "mean_token_accuracy": 0.1638083577156067,
      "num_tokens": 23502503.0,
      "step": 9275
    },
    {
      "entropy": 6.284149837493897,
      "epoch": 1.0709751875360647,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004897053078525016,
      "loss": 6.0754,
      "mean_token_accuracy": 0.16358094960451125,
      "num_tokens": 23514903.0,
      "step": 9280
    },
    {
      "entropy": 6.306416320800781,
      "epoch": 1.0715522215810733,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004896929652394993,
      "loss": 6.12,
      "mean_token_accuracy": 0.16296200454235077,
      "num_tokens": 23528563.0,
      "step": 9285
    },
    {
      "entropy": 6.295992183685303,
      "epoch": 1.072129255626082,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004896806154054057,
      "loss": 6.204,
      "mean_token_accuracy": 0.1587452083826065,
      "num_tokens": 23541631.0,
      "step": 9290
    },
    {
      "entropy": 6.290651607513428,
      "epoch": 1.0727062896710906,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004896682583506363,
      "loss": 6.0365,
      "mean_token_accuracy": 0.16789826899766921,
      "num_tokens": 23553665.0,
      "step": 9295
    },
    {
      "entropy": 6.201587867736817,
      "epoch": 1.0732833237160992,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004896558940756067,
      "loss": 6.0811,
      "mean_token_accuracy": 0.16426270306110383,
      "num_tokens": 23567050.0,
      "step": 9300
    },
    {
      "entropy": 6.207051944732666,
      "epoch": 1.0738603577611079,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004896435225807327,
      "loss": 6.1137,
      "mean_token_accuracy": 0.16744877845048906,
      "num_tokens": 23580676.0,
      "step": 9305
    },
    {
      "entropy": 6.300777626037598,
      "epoch": 1.0744373918061165,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004896311438664304,
      "loss": 6.1092,
      "mean_token_accuracy": 0.16863760203123093,
      "num_tokens": 23592705.0,
      "step": 9310
    },
    {
      "entropy": 6.301867389678955,
      "epoch": 1.0750144258511252,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004896187579331163,
      "loss": 6.1321,
      "mean_token_accuracy": 0.1659500390291214,
      "num_tokens": 23604365.0,
      "step": 9315
    },
    {
      "entropy": 6.216816711425781,
      "epoch": 1.0755914598961338,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004896063647812068,
      "loss": 6.1415,
      "mean_token_accuracy": 0.16507016271352767,
      "num_tokens": 23617657.0,
      "step": 9320
    },
    {
      "entropy": 6.308189773559571,
      "epoch": 1.0761684939411424,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004895939644111189,
      "loss": 6.132,
      "mean_token_accuracy": 0.1621185377240181,
      "num_tokens": 23630215.0,
      "step": 9325
    },
    {
      "entropy": 6.381429290771484,
      "epoch": 1.076745527986151,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004895815568232694,
      "loss": 6.1193,
      "mean_token_accuracy": 0.1686142161488533,
      "num_tokens": 23643126.0,
      "step": 9330
    },
    {
      "entropy": 6.082545232772827,
      "epoch": 1.07732256203116,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004895691420180759,
      "loss": 6.0693,
      "mean_token_accuracy": 0.1625459000468254,
      "num_tokens": 23656156.0,
      "step": 9335
    },
    {
      "entropy": 6.22721037864685,
      "epoch": 1.0778995960761686,
      "grad_norm": 1.375,
      "learning_rate": 0.000489556719995956,
      "loss": 6.1119,
      "mean_token_accuracy": 0.16327948570251466,
      "num_tokens": 23668930.0,
      "step": 9340
    },
    {
      "entropy": 6.378998565673828,
      "epoch": 1.0784766301211772,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004895442907573274,
      "loss": 6.1656,
      "mean_token_accuracy": 0.15929747372865677,
      "num_tokens": 23681386.0,
      "step": 9345
    },
    {
      "entropy": 6.239406681060791,
      "epoch": 1.0790536641661859,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000489531854302608,
      "loss": 6.2099,
      "mean_token_accuracy": 0.15547912046313286,
      "num_tokens": 23695381.0,
      "step": 9350
    },
    {
      "entropy": 6.223207473754883,
      "epoch": 1.0796306982111945,
      "grad_norm": 1.125,
      "learning_rate": 0.0004895194106322164,
      "loss": 6.1419,
      "mean_token_accuracy": 0.16573455780744553,
      "num_tokens": 23707259.0,
      "step": 9355
    },
    {
      "entropy": 6.419655227661133,
      "epoch": 1.0802077322562031,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004895069597465709,
      "loss": 6.1877,
      "mean_token_accuracy": 0.15922195017337798,
      "num_tokens": 23720933.0,
      "step": 9360
    },
    {
      "entropy": 6.26898889541626,
      "epoch": 1.0807847663012118,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004894945016460904,
      "loss": 6.1123,
      "mean_token_accuracy": 0.16257226020097731,
      "num_tokens": 23733735.0,
      "step": 9365
    },
    {
      "entropy": 6.247603416442871,
      "epoch": 1.0813618003462204,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004894820363311938,
      "loss": 6.1601,
      "mean_token_accuracy": 0.1639579564332962,
      "num_tokens": 23746231.0,
      "step": 9370
    },
    {
      "entropy": 6.2967294216156,
      "epoch": 1.081938834391229,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004894695638023005,
      "loss": 6.0904,
      "mean_token_accuracy": 0.16889988780021667,
      "num_tokens": 23758817.0,
      "step": 9375
    },
    {
      "entropy": 6.3042731285095215,
      "epoch": 1.0825158684362377,
      "grad_norm": 1.640625,
      "learning_rate": 0.00048945708405983,
      "loss": 6.1311,
      "mean_token_accuracy": 0.16656047701835633,
      "num_tokens": 23771983.0,
      "step": 9380
    },
    {
      "entropy": 6.277537059783936,
      "epoch": 1.0830929024812463,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004894445971042019,
      "loss": 6.1552,
      "mean_token_accuracy": 0.1622651696205139,
      "num_tokens": 23784112.0,
      "step": 9385
    },
    {
      "entropy": 6.407172822952271,
      "epoch": 1.083669936526255,
      "grad_norm": 1.875,
      "learning_rate": 0.0004894321029358364,
      "loss": 6.1702,
      "mean_token_accuracy": 0.15601906031370164,
      "num_tokens": 23796854.0,
      "step": 9390
    },
    {
      "entropy": 6.343762683868408,
      "epoch": 1.0842469705712636,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004894196015551536,
      "loss": 6.1641,
      "mean_token_accuracy": 0.16759544909000396,
      "num_tokens": 23809450.0,
      "step": 9395
    },
    {
      "entropy": 6.30757417678833,
      "epoch": 1.0848240046162723,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000489407092962574,
      "loss": 6.2143,
      "mean_token_accuracy": 0.15676299482584,
      "num_tokens": 23821742.0,
      "step": 9400
    },
    {
      "entropy": 6.246688032150269,
      "epoch": 1.085401038661281,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004893945771585183,
      "loss": 6.06,
      "mean_token_accuracy": 0.16792974174022673,
      "num_tokens": 23834174.0,
      "step": 9405
    },
    {
      "entropy": 6.218525171279907,
      "epoch": 1.0859780727062898,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004893820541434075,
      "loss": 6.0874,
      "mean_token_accuracy": 0.16832081228494644,
      "num_tokens": 23847076.0,
      "step": 9410
    },
    {
      "entropy": 6.264400625228882,
      "epoch": 1.0865551067512984,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004893695239176629,
      "loss": 6.112,
      "mean_token_accuracy": 0.1674555942416191,
      "num_tokens": 23860682.0,
      "step": 9415
    },
    {
      "entropy": 6.158101320266724,
      "epoch": 1.087132140796307,
      "grad_norm": 1.125,
      "learning_rate": 0.0004893569864817057,
      "loss": 5.9683,
      "mean_token_accuracy": 0.1703802689909935,
      "num_tokens": 23873631.0,
      "step": 9420
    },
    {
      "entropy": 6.223286199569702,
      "epoch": 1.0877091748413157,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004893444418359579,
      "loss": 6.0569,
      "mean_token_accuracy": 0.1697439581155777,
      "num_tokens": 23885144.0,
      "step": 9425
    },
    {
      "entropy": 6.331843852996826,
      "epoch": 1.0882862088863243,
      "grad_norm": 1.0625,
      "learning_rate": 0.000489331889980841,
      "loss": 6.1574,
      "mean_token_accuracy": 0.16503539085388183,
      "num_tokens": 23898313.0,
      "step": 9430
    },
    {
      "entropy": 6.332405090332031,
      "epoch": 1.088863242931333,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004893193309167778,
      "loss": 6.1076,
      "mean_token_accuracy": 0.16618113219738007,
      "num_tokens": 23910002.0,
      "step": 9435
    },
    {
      "entropy": 6.276224517822266,
      "epoch": 1.0894402769763416,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004893067646441902,
      "loss": 6.1869,
      "mean_token_accuracy": 0.1596734806895256,
      "num_tokens": 23922770.0,
      "step": 9440
    },
    {
      "entropy": 6.295549011230468,
      "epoch": 1.0900173110213502,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000489294191163501,
      "loss": 6.1737,
      "mean_token_accuracy": 0.16593922227621077,
      "num_tokens": 23936218.0,
      "step": 9445
    },
    {
      "entropy": 6.2784405708312985,
      "epoch": 1.0905943450663589,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004892816104751331,
      "loss": 6.1654,
      "mean_token_accuracy": 0.16389482468366623,
      "num_tokens": 23949250.0,
      "step": 9450
    },
    {
      "entropy": 6.375916337966919,
      "epoch": 1.0911713791113675,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004892690225795096,
      "loss": 6.0871,
      "mean_token_accuracy": 0.1641623094677925,
      "num_tokens": 23961112.0,
      "step": 9455
    },
    {
      "entropy": 6.314445209503174,
      "epoch": 1.0917484131563762,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004892564274770542,
      "loss": 6.1839,
      "mean_token_accuracy": 0.15494174212217332,
      "num_tokens": 23974450.0,
      "step": 9460
    },
    {
      "entropy": 6.278241300582886,
      "epoch": 1.0923254472013848,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004892438251681901,
      "loss": 6.1137,
      "mean_token_accuracy": 0.16092193722724915,
      "num_tokens": 23986368.0,
      "step": 9465
    },
    {
      "entropy": 6.303944063186646,
      "epoch": 1.0929024812463934,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004892312156533413,
      "loss": 6.1634,
      "mean_token_accuracy": 0.16244134157896042,
      "num_tokens": 23998186.0,
      "step": 9470
    },
    {
      "entropy": 6.254711055755616,
      "epoch": 1.0934795152914023,
      "grad_norm": 1.125,
      "learning_rate": 0.0004892185989329321,
      "loss": 6.1859,
      "mean_token_accuracy": 0.15706839561462402,
      "num_tokens": 24011909.0,
      "step": 9475
    },
    {
      "entropy": 6.274745273590088,
      "epoch": 1.094056549336411,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004892059750073868,
      "loss": 6.1584,
      "mean_token_accuracy": 0.16165001988410949,
      "num_tokens": 24024142.0,
      "step": 9480
    },
    {
      "entropy": 6.169730186462402,
      "epoch": 1.0946335833814196,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004891933438771299,
      "loss": 6.0689,
      "mean_token_accuracy": 0.16412845402956008,
      "num_tokens": 24037579.0,
      "step": 9485
    },
    {
      "entropy": 6.3204652786254885,
      "epoch": 1.0952106174264282,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004891807055425862,
      "loss": 6.1092,
      "mean_token_accuracy": 0.16852035075426103,
      "num_tokens": 24050707.0,
      "step": 9490
    },
    {
      "entropy": 6.250433826446534,
      "epoch": 1.0957876514714369,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004891680600041809,
      "loss": 6.135,
      "mean_token_accuracy": 0.16853047609329225,
      "num_tokens": 24064148.0,
      "step": 9495
    },
    {
      "entropy": 6.197220993041992,
      "epoch": 1.0963646855164455,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004891554072623392,
      "loss": 6.1068,
      "mean_token_accuracy": 0.164017516374588,
      "num_tokens": 24077769.0,
      "step": 9500
    },
    {
      "entropy": 6.352374029159546,
      "epoch": 1.0969417195614541,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004891427473174869,
      "loss": 6.1616,
      "mean_token_accuracy": 0.16006930321455,
      "num_tokens": 24089364.0,
      "step": 9505
    },
    {
      "entropy": 6.304450368881225,
      "epoch": 1.0975187536064628,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004891300801700496,
      "loss": 6.0961,
      "mean_token_accuracy": 0.17309346944093704,
      "num_tokens": 24102890.0,
      "step": 9510
    },
    {
      "entropy": 6.151680612564087,
      "epoch": 1.0980957876514714,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004891174058204534,
      "loss": 6.1026,
      "mean_token_accuracy": 0.1652431905269623,
      "num_tokens": 24116290.0,
      "step": 9515
    },
    {
      "entropy": 6.246303462982178,
      "epoch": 1.09867282169648,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004891047242691246,
      "loss": 6.0723,
      "mean_token_accuracy": 0.1669682264328003,
      "num_tokens": 24128318.0,
      "step": 9520
    },
    {
      "entropy": 6.244490146636963,
      "epoch": 1.0992498557414887,
      "grad_norm": 1.25,
      "learning_rate": 0.0004890920355164897,
      "loss": 6.0683,
      "mean_token_accuracy": 0.1730115070939064,
      "num_tokens": 24140947.0,
      "step": 9525
    },
    {
      "entropy": 6.196271657943726,
      "epoch": 1.0998268897864973,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004890793395629756,
      "loss": 6.0977,
      "mean_token_accuracy": 0.16113903373479843,
      "num_tokens": 24153254.0,
      "step": 9530
    },
    {
      "entropy": 6.327161836624145,
      "epoch": 1.100403923831506,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004890666364090093,
      "loss": 6.1174,
      "mean_token_accuracy": 0.16286847591400147,
      "num_tokens": 24166115.0,
      "step": 9535
    },
    {
      "entropy": 6.240642261505127,
      "epoch": 1.1009809578765146,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004890539260550181,
      "loss": 6.0854,
      "mean_token_accuracy": 0.17154496163129807,
      "num_tokens": 24178725.0,
      "step": 9540
    },
    {
      "entropy": 6.2435588359832765,
      "epoch": 1.1015579919215233,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004890412085014292,
      "loss": 6.1025,
      "mean_token_accuracy": 0.1737593561410904,
      "num_tokens": 24191258.0,
      "step": 9545
    },
    {
      "entropy": 6.260970830917358,
      "epoch": 1.1021350259665321,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004890284837486706,
      "loss": 6.1047,
      "mean_token_accuracy": 0.15966490358114244,
      "num_tokens": 24203096.0,
      "step": 9550
    },
    {
      "entropy": 6.312766218185425,
      "epoch": 1.1027120600115408,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004890157517971704,
      "loss": 6.1476,
      "mean_token_accuracy": 0.1686030238866806,
      "num_tokens": 24214729.0,
      "step": 9555
    },
    {
      "entropy": 6.231602573394776,
      "epoch": 1.1032890940565494,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004890030126473566,
      "loss": 6.0827,
      "mean_token_accuracy": 0.1710619792342186,
      "num_tokens": 24227164.0,
      "step": 9560
    },
    {
      "entropy": 6.231527709960938,
      "epoch": 1.103866128101558,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004889902662996578,
      "loss": 6.1447,
      "mean_token_accuracy": 0.16064967066049576,
      "num_tokens": 24239240.0,
      "step": 9565
    },
    {
      "entropy": 6.312191915512085,
      "epoch": 1.1044431621465667,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004889775127545027,
      "loss": 6.0844,
      "mean_token_accuracy": 0.1606465607881546,
      "num_tokens": 24251562.0,
      "step": 9570
    },
    {
      "entropy": 6.233772850036621,
      "epoch": 1.1050201961915753,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004889647520123202,
      "loss": 6.1316,
      "mean_token_accuracy": 0.1686311200261116,
      "num_tokens": 24265439.0,
      "step": 9575
    },
    {
      "entropy": 6.210593795776367,
      "epoch": 1.105597230236584,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004889519840735396,
      "loss": 6.0266,
      "mean_token_accuracy": 0.16629702746868133,
      "num_tokens": 24278982.0,
      "step": 9580
    },
    {
      "entropy": 6.316622066497803,
      "epoch": 1.1061742642815926,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004889392089385903,
      "loss": 6.1462,
      "mean_token_accuracy": 0.15745373964309692,
      "num_tokens": 24291119.0,
      "step": 9585
    },
    {
      "entropy": 6.263356256484985,
      "epoch": 1.1067512983266012,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004889264266079019,
      "loss": 6.0996,
      "mean_token_accuracy": 0.1621411547064781,
      "num_tokens": 24302646.0,
      "step": 9590
    },
    {
      "entropy": 6.222952222824096,
      "epoch": 1.1073283323716099,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004889136370819045,
      "loss": 6.0788,
      "mean_token_accuracy": 0.163968588411808,
      "num_tokens": 24314754.0,
      "step": 9595
    },
    {
      "entropy": 6.320454788208008,
      "epoch": 1.1079053664166185,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004889008403610281,
      "loss": 6.0668,
      "mean_token_accuracy": 0.16170306652784347,
      "num_tokens": 24328241.0,
      "step": 9600
    },
    {
      "entropy": 6.202263021469117,
      "epoch": 1.1084824004616272,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004888880364457033,
      "loss": 6.1041,
      "mean_token_accuracy": 0.1690272197127342,
      "num_tokens": 24340319.0,
      "step": 9605
    },
    {
      "entropy": 6.301255512237549,
      "epoch": 1.1090594345066358,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004888752253363604,
      "loss": 6.1741,
      "mean_token_accuracy": 0.15546730756759644,
      "num_tokens": 24352457.0,
      "step": 9610
    },
    {
      "entropy": 6.2395905494689945,
      "epoch": 1.1096364685516447,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004888624070334308,
      "loss": 6.0514,
      "mean_token_accuracy": 0.16441725194454193,
      "num_tokens": 24363587.0,
      "step": 9615
    },
    {
      "entropy": 6.2865642547607425,
      "epoch": 1.1102135025966533,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004888495815373452,
      "loss": 6.1485,
      "mean_token_accuracy": 0.1660570427775383,
      "num_tokens": 24376555.0,
      "step": 9620
    },
    {
      "entropy": 6.312252807617187,
      "epoch": 1.110790536641662,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004888367488485351,
      "loss": 6.1218,
      "mean_token_accuracy": 0.16511175483465196,
      "num_tokens": 24388720.0,
      "step": 9625
    },
    {
      "entropy": 6.277844572067261,
      "epoch": 1.1113675706866706,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004888239089674323,
      "loss": 6.0529,
      "mean_token_accuracy": 0.1682996317744255,
      "num_tokens": 24400320.0,
      "step": 9630
    },
    {
      "entropy": 6.289002799987793,
      "epoch": 1.1119446047316792,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004888110618944686,
      "loss": 6.1398,
      "mean_token_accuracy": 0.17045128643512725,
      "num_tokens": 24414521.0,
      "step": 9635
    },
    {
      "entropy": 6.192933177947998,
      "epoch": 1.1125216387766879,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004887982076300759,
      "loss": 6.1264,
      "mean_token_accuracy": 0.16553627997636794,
      "num_tokens": 24426254.0,
      "step": 9640
    },
    {
      "entropy": 6.300319814682007,
      "epoch": 1.1130986728216965,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004887853461746867,
      "loss": 6.1495,
      "mean_token_accuracy": 0.1689577266573906,
      "num_tokens": 24438089.0,
      "step": 9645
    },
    {
      "entropy": 6.317306756973267,
      "epoch": 1.1136757068667051,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004887724775287336,
      "loss": 6.1763,
      "mean_token_accuracy": 0.16700106412172316,
      "num_tokens": 24450214.0,
      "step": 9650
    },
    {
      "entropy": 6.169091844558716,
      "epoch": 1.1142527409117138,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004887596016926494,
      "loss": 6.0438,
      "mean_token_accuracy": 0.17151311710476874,
      "num_tokens": 24463577.0,
      "step": 9655
    },
    {
      "entropy": 6.183196926116944,
      "epoch": 1.1148297749567224,
      "grad_norm": 1.125,
      "learning_rate": 0.0004887467186668673,
      "loss": 6.0296,
      "mean_token_accuracy": 0.17124876976013184,
      "num_tokens": 24475533.0,
      "step": 9660
    },
    {
      "entropy": 6.318900537490845,
      "epoch": 1.115406809001731,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004887338284518204,
      "loss": 6.1665,
      "mean_token_accuracy": 0.16185251176357268,
      "num_tokens": 24488779.0,
      "step": 9665
    },
    {
      "entropy": 6.180599164962769,
      "epoch": 1.1159838430467397,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004887209310479423,
      "loss": 5.9687,
      "mean_token_accuracy": 0.1751622125506401,
      "num_tokens": 24501967.0,
      "step": 9670
    },
    {
      "entropy": 6.230393743515014,
      "epoch": 1.1165608770917483,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004887080264556668,
      "loss": 6.1883,
      "mean_token_accuracy": 0.16221662610769272,
      "num_tokens": 24513743.0,
      "step": 9675
    },
    {
      "entropy": 6.331747245788574,
      "epoch": 1.117137911136757,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004886951146754282,
      "loss": 6.14,
      "mean_token_accuracy": 0.16292295306921006,
      "num_tokens": 24524782.0,
      "step": 9680
    },
    {
      "entropy": 6.299449062347412,
      "epoch": 1.1177149451817656,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004886821957076603,
      "loss": 6.1831,
      "mean_token_accuracy": 0.15988557934761047,
      "num_tokens": 24537598.0,
      "step": 9685
    },
    {
      "entropy": 6.259903907775879,
      "epoch": 1.1182919792267745,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000488669269552798,
      "loss": 6.1764,
      "mean_token_accuracy": 0.15695709437131883,
      "num_tokens": 24551100.0,
      "step": 9690
    },
    {
      "entropy": 6.330960845947265,
      "epoch": 1.1188690132717831,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000488656336211276,
      "loss": 6.1983,
      "mean_token_accuracy": 0.16324839890003204,
      "num_tokens": 24562604.0,
      "step": 9695
    },
    {
      "entropy": 6.30520076751709,
      "epoch": 1.1194460473167918,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004886433956835292,
      "loss": 6.1511,
      "mean_token_accuracy": 0.16451692581176758,
      "num_tokens": 24573990.0,
      "step": 9700
    },
    {
      "entropy": 6.304443454742431,
      "epoch": 1.1200230813618004,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004886304479699929,
      "loss": 6.1241,
      "mean_token_accuracy": 0.16457738280296325,
      "num_tokens": 24586320.0,
      "step": 9705
    },
    {
      "entropy": 6.227281999588013,
      "epoch": 1.120600115406809,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004886174930711027,
      "loss": 6.0529,
      "mean_token_accuracy": 0.17341021597385406,
      "num_tokens": 24597995.0,
      "step": 9710
    },
    {
      "entropy": 6.229392385482788,
      "epoch": 1.1211771494518177,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004886045309872941,
      "loss": 6.1459,
      "mean_token_accuracy": 0.16764940768480302,
      "num_tokens": 24609752.0,
      "step": 9715
    },
    {
      "entropy": 6.212685680389404,
      "epoch": 1.1217541834968263,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004885915617190034,
      "loss": 6.0605,
      "mean_token_accuracy": 0.17084028869867324,
      "num_tokens": 24622374.0,
      "step": 9720
    },
    {
      "entropy": 6.240245008468628,
      "epoch": 1.122331217541835,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004885785852666664,
      "loss": 5.9255,
      "mean_token_accuracy": 0.18366212546825408,
      "num_tokens": 24634877.0,
      "step": 9725
    },
    {
      "entropy": 6.171277046203613,
      "epoch": 1.1229082515868436,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004885656016307199,
      "loss": 6.0493,
      "mean_token_accuracy": 0.16829527765512467,
      "num_tokens": 24647498.0,
      "step": 9730
    },
    {
      "entropy": 6.098988246917725,
      "epoch": 1.1234852856318522,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004885526108116004,
      "loss": 6.0625,
      "mean_token_accuracy": 0.17167832255363463,
      "num_tokens": 24660331.0,
      "step": 9735
    },
    {
      "entropy": 6.361937999725342,
      "epoch": 1.1240623196768609,
      "grad_norm": 1.34375,
      "learning_rate": 0.000488539612809745,
      "loss": 6.1177,
      "mean_token_accuracy": 0.16555654108524323,
      "num_tokens": 24673169.0,
      "step": 9740
    },
    {
      "entropy": 6.2339081287384035,
      "epoch": 1.1246393537218695,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004885266076255907,
      "loss": 6.0156,
      "mean_token_accuracy": 0.16694944202899933,
      "num_tokens": 24684900.0,
      "step": 9745
    },
    {
      "entropy": 6.267468643188477,
      "epoch": 1.1252163877668782,
      "grad_norm": 2.765625,
      "learning_rate": 0.000488513595259575,
      "loss": 6.0052,
      "mean_token_accuracy": 0.17324970066547393,
      "num_tokens": 24699135.0,
      "step": 9750
    },
    {
      "entropy": 6.3071935176849365,
      "epoch": 1.125793421811887,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004885005757121357,
      "loss": 6.1425,
      "mean_token_accuracy": 0.16348030865192414,
      "num_tokens": 24711402.0,
      "step": 9755
    },
    {
      "entropy": 6.198820161819458,
      "epoch": 1.1263704558568954,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004884875489837105,
      "loss": 6.1187,
      "mean_token_accuracy": 0.16100963950157166,
      "num_tokens": 24723438.0,
      "step": 9760
    },
    {
      "entropy": 6.266419458389282,
      "epoch": 1.1269474899019043,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004884745150747378,
      "loss": 6.1047,
      "mean_token_accuracy": 0.17071303725242615,
      "num_tokens": 24737939.0,
      "step": 9765
    },
    {
      "entropy": 6.221473836898804,
      "epoch": 1.127524523946913,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004884614739856556,
      "loss": 6.0304,
      "mean_token_accuracy": 0.16827375292778016,
      "num_tokens": 24751120.0,
      "step": 9770
    },
    {
      "entropy": 6.243126344680786,
      "epoch": 1.1281015579919216,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004884484257169028,
      "loss": 6.021,
      "mean_token_accuracy": 0.1731890171766281,
      "num_tokens": 24763156.0,
      "step": 9775
    },
    {
      "entropy": 6.28215446472168,
      "epoch": 1.1286785920369302,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004884353702689183,
      "loss": 6.1377,
      "mean_token_accuracy": 0.16387613862752914,
      "num_tokens": 24775728.0,
      "step": 9780
    },
    {
      "entropy": 6.135974740982055,
      "epoch": 1.1292556260819389,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004884223076421411,
      "loss": 6.0112,
      "mean_token_accuracy": 0.17222613394260405,
      "num_tokens": 24789232.0,
      "step": 9785
    },
    {
      "entropy": 6.265791606903076,
      "epoch": 1.1298326601269475,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004884092378370106,
      "loss": 6.1123,
      "mean_token_accuracy": 0.16204613000154494,
      "num_tokens": 24801188.0,
      "step": 9790
    },
    {
      "entropy": 6.1921991348266605,
      "epoch": 1.1304096941719561,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004883961608539664,
      "loss": 6.0502,
      "mean_token_accuracy": 0.1745037019252777,
      "num_tokens": 24814482.0,
      "step": 9795
    },
    {
      "entropy": 6.240455627441406,
      "epoch": 1.1309867282169648,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004883830766934484,
      "loss": 6.1364,
      "mean_token_accuracy": 0.15735915303230286,
      "num_tokens": 24829042.0,
      "step": 9800
    },
    {
      "entropy": 6.3157586574554445,
      "epoch": 1.1315637622619734,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004883699853558965,
      "loss": 6.1222,
      "mean_token_accuracy": 0.16135212630033494,
      "num_tokens": 24842227.0,
      "step": 9805
    },
    {
      "entropy": 6.187296962738037,
      "epoch": 1.132140796306982,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004883568868417511,
      "loss": 6.0457,
      "mean_token_accuracy": 0.17535947412252426,
      "num_tokens": 24856225.0,
      "step": 9810
    },
    {
      "entropy": 6.208665943145752,
      "epoch": 1.1327178303519907,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004883437811514528,
      "loss": 6.0878,
      "mean_token_accuracy": 0.1653905138373375,
      "num_tokens": 24868889.0,
      "step": 9815
    },
    {
      "entropy": 6.23685393333435,
      "epoch": 1.1332948643969993,
      "grad_norm": 1.125,
      "learning_rate": 0.0004883306682854424,
      "loss": 6.0529,
      "mean_token_accuracy": 0.1685851439833641,
      "num_tokens": 24881738.0,
      "step": 9820
    },
    {
      "entropy": 6.294564199447632,
      "epoch": 1.133871898442008,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004883175482441611,
      "loss": 6.0911,
      "mean_token_accuracy": 0.16411995589733125,
      "num_tokens": 24893886.0,
      "step": 9825
    },
    {
      "entropy": 6.2085202693939205,
      "epoch": 1.1344489324870168,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004883044210280499,
      "loss": 6.1319,
      "mean_token_accuracy": 0.15953268557786943,
      "num_tokens": 24908213.0,
      "step": 9830
    },
    {
      "entropy": 6.3302061557769775,
      "epoch": 1.1350259665320255,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004882912866375505,
      "loss": 6.1882,
      "mean_token_accuracy": 0.15905737578868867,
      "num_tokens": 24921157.0,
      "step": 9835
    },
    {
      "entropy": 6.343192386627197,
      "epoch": 1.1356030005770341,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048827814507310455,
      "loss": 6.1061,
      "mean_token_accuracy": 0.1667718321084976,
      "num_tokens": 24933922.0,
      "step": 9840
    },
    {
      "entropy": 6.213817024230957,
      "epoch": 1.1361800346220428,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048826499633515416,
      "loss": 6.0687,
      "mean_token_accuracy": 0.17247530370950698,
      "num_tokens": 24945351.0,
      "step": 9845
    },
    {
      "entropy": 6.211613750457763,
      "epoch": 1.1367570686670514,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048825184042414156,
      "loss": 6.0927,
      "mean_token_accuracy": 0.1713977575302124,
      "num_tokens": 24958099.0,
      "step": 9850
    },
    {
      "entropy": 6.213688898086548,
      "epoch": 1.13733410271206,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004882386773405092,
      "loss": 6.1002,
      "mean_token_accuracy": 0.16913821697235107,
      "num_tokens": 24970622.0,
      "step": 9855
    },
    {
      "entropy": 6.333008432388306,
      "epoch": 1.1379111367570687,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004882255070846999,
      "loss": 6.2109,
      "mean_token_accuracy": 0.156184521317482,
      "num_tokens": 24983456.0,
      "step": 9860
    },
    {
      "entropy": 6.282187747955322,
      "epoch": 1.1384881708020773,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048821232965715663,
      "loss": 6.1126,
      "mean_token_accuracy": 0.17017486095428466,
      "num_tokens": 24995976.0,
      "step": 9865
    },
    {
      "entropy": 6.1477538585662845,
      "epoch": 1.139065204847086,
      "grad_norm": 1.25,
      "learning_rate": 0.0004881991450583225,
      "loss": 6.0927,
      "mean_token_accuracy": 0.17406780868768693,
      "num_tokens": 25008456.0,
      "step": 9870
    },
    {
      "entropy": 6.275055694580078,
      "epoch": 1.1396422388920946,
      "grad_norm": 1.109375,
      "learning_rate": 0.000488185953288641,
      "loss": 6.0778,
      "mean_token_accuracy": 0.16098694652318954,
      "num_tokens": 25021186.0,
      "step": 9875
    },
    {
      "entropy": 6.302857494354248,
      "epoch": 1.1402192729371032,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004881727543485559,
      "loss": 6.1691,
      "mean_token_accuracy": 0.1565827801823616,
      "num_tokens": 25034561.0,
      "step": 9880
    },
    {
      "entropy": 6.256385707855225,
      "epoch": 1.1407963069821119,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048815954823851107,
      "loss": 6.0649,
      "mean_token_accuracy": 0.16743325889110566,
      "num_tokens": 25046107.0,
      "step": 9885
    },
    {
      "entropy": 6.289016675949097,
      "epoch": 1.1413733410271205,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048814633495895067,
      "loss": 6.1324,
      "mean_token_accuracy": 0.16312570124864578,
      "num_tokens": 25058605.0,
      "step": 9890
    },
    {
      "entropy": 6.254915571212768,
      "epoch": 1.1419503750721294,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004881331145103192,
      "loss": 6.0507,
      "mean_token_accuracy": 0.1701641261577606,
      "num_tokens": 25070730.0,
      "step": 9895
    },
    {
      "entropy": 6.205109405517578,
      "epoch": 1.1425274091171378,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004881198868930614,
      "loss": 6.0935,
      "mean_token_accuracy": 0.16716530919075012,
      "num_tokens": 25084121.0,
      "step": 9900
    },
    {
      "entropy": 6.222742509841919,
      "epoch": 1.1431044431621467,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048810665210762195,
      "loss": 6.0924,
      "mean_token_accuracy": 0.1754786804318428,
      "num_tokens": 25096132.0,
      "step": 9905
    },
    {
      "entropy": 6.304066896438599,
      "epoch": 1.1436814772071553,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048809341015444615,
      "loss": 6.2353,
      "mean_token_accuracy": 0.16024893075227736,
      "num_tokens": 25108220.0,
      "step": 9910
    },
    {
      "entropy": 6.268913745880127,
      "epoch": 1.144258511252164,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048808016103397934,
      "loss": 6.1451,
      "mean_token_accuracy": 0.16343348920345308,
      "num_tokens": 25120955.0,
      "step": 9915
    },
    {
      "entropy": 6.3085408210754395,
      "epoch": 1.1448355452971726,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004880669047466671,
      "loss": 6.1462,
      "mean_token_accuracy": 0.16178373396396636,
      "num_tokens": 25133644.0,
      "step": 9920
    },
    {
      "entropy": 6.203466510772705,
      "epoch": 1.1454125793421812,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048805364129295554,
      "loss": 6.0228,
      "mean_token_accuracy": 0.17411303222179414,
      "num_tokens": 25147172.0,
      "step": 9925
    },
    {
      "entropy": 6.2808722972869875,
      "epoch": 1.1459896133871899,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048804037067329037,
      "loss": 6.0813,
      "mean_token_accuracy": 0.17082541435956955,
      "num_tokens": 25159293.0,
      "step": 9930
    },
    {
      "entropy": 6.24948935508728,
      "epoch": 1.1465666474321985,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004880270928881183,
      "loss": 6.1447,
      "mean_token_accuracy": 0.15984688550233841,
      "num_tokens": 25172149.0,
      "step": 9935
    },
    {
      "entropy": 6.2018743515014645,
      "epoch": 1.1471436814772071,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004880138079378857,
      "loss": 5.9932,
      "mean_token_accuracy": 0.16733321845531463,
      "num_tokens": 25184255.0,
      "step": 9940
    },
    {
      "entropy": 6.143245363235474,
      "epoch": 1.1477207155222158,
      "grad_norm": 1.125,
      "learning_rate": 0.0004880005158230395,
      "loss": 6.0576,
      "mean_token_accuracy": 0.17233417332172393,
      "num_tokens": 25197532.0,
      "step": 9945
    },
    {
      "entropy": 6.2314058303833,
      "epoch": 1.1482977495672244,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004879872165440268,
      "loss": 6.0936,
      "mean_token_accuracy": 0.16797082275152206,
      "num_tokens": 25210046.0,
      "step": 9950
    },
    {
      "entropy": 6.33461709022522,
      "epoch": 1.148874783612233,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004879739101012948,
      "loss": 6.1845,
      "mean_token_accuracy": 0.16109299212694167,
      "num_tokens": 25221123.0,
      "step": 9955
    },
    {
      "entropy": 6.203636598587036,
      "epoch": 1.1494518176572417,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004879605964952911,
      "loss": 6.0628,
      "mean_token_accuracy": 0.16972762048244477,
      "num_tokens": 25233196.0,
      "step": 9960
    },
    {
      "entropy": 6.247003412246704,
      "epoch": 1.1500288517022503,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048794727572646366,
      "loss": 6.0943,
      "mean_token_accuracy": 0.17366009056568146,
      "num_tokens": 25247921.0,
      "step": 9965
    },
    {
      "entropy": 6.310330057144165,
      "epoch": 1.1506058857472592,
      "grad_norm": 4.5,
      "learning_rate": 0.0004879339477952603,
      "loss": 6.1053,
      "mean_token_accuracy": 0.16640760600566865,
      "num_tokens": 25260999.0,
      "step": 9970
    },
    {
      "entropy": 6.171389865875244,
      "epoch": 1.1511829197922678,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048792061270212935,
      "loss": 6.0658,
      "mean_token_accuracy": 0.16953585296869278,
      "num_tokens": 25273435.0,
      "step": 9975
    },
    {
      "entropy": 6.268343830108643,
      "epoch": 1.1517599538372765,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004879072704475193,
      "loss": 6.1917,
      "mean_token_accuracy": 0.16562627851963044,
      "num_tokens": 25286071.0,
      "step": 9980
    },
    {
      "entropy": 6.350545406341553,
      "epoch": 1.1523369878822851,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048789392103187906,
      "loss": 6.1141,
      "mean_token_accuracy": 0.16076473146677017,
      "num_tokens": 25298083.0,
      "step": 9985
    },
    {
      "entropy": 6.284563159942627,
      "epoch": 1.1529140219272938,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004878805644556575,
      "loss": 6.166,
      "mean_token_accuracy": 0.16007014364004135,
      "num_tokens": 25310488.0,
      "step": 9990
    },
    {
      "entropy": 6.2582155704498295,
      "epoch": 1.1534910559723024,
      "grad_norm": 1.109375,
      "learning_rate": 0.000487867200719304,
      "loss": 6.0908,
      "mean_token_accuracy": 0.16960073858499528,
      "num_tokens": 25323135.0,
      "step": 9995
    },
    {
      "entropy": 6.243107795715332,
      "epoch": 1.154068090017311,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004878538298232678,
      "loss": 6.1422,
      "mean_token_accuracy": 0.1620587721467018,
      "num_tokens": 25335349.0,
      "step": 10000
    },
    {
      "entropy": 6.28153429031372,
      "epoch": 1.1546451240623197,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004878404517679988,
      "loss": 6.1426,
      "mean_token_accuracy": 0.16404337733983992,
      "num_tokens": 25347717.0,
      "step": 10005
    },
    {
      "entropy": 6.20682315826416,
      "epoch": 1.1552221581073283,
      "grad_norm": 1.125,
      "learning_rate": 0.00048782706655394695,
      "loss": 6.0249,
      "mean_token_accuracy": 0.17437740862369538,
      "num_tokens": 25360343.0,
      "step": 10010
    },
    {
      "entropy": 6.22298412322998,
      "epoch": 1.155799192152337,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004878136741815624,
      "loss": 6.1102,
      "mean_token_accuracy": 0.1684797078371048,
      "num_tokens": 25373516.0,
      "step": 10015
    },
    {
      "entropy": 6.34765043258667,
      "epoch": 1.1563762261973456,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004878002746512956,
      "loss": 6.149,
      "mean_token_accuracy": 0.1629353180527687,
      "num_tokens": 25386728.0,
      "step": 10020
    },
    {
      "entropy": 6.2509033203125,
      "epoch": 1.1569532602423542,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004877868679635974,
      "loss": 6.0563,
      "mean_token_accuracy": 0.17708538770675658,
      "num_tokens": 25399419.0,
      "step": 10025
    },
    {
      "entropy": 6.200060796737671,
      "epoch": 1.1575302942873629,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004877734541189185,
      "loss": 6.1092,
      "mean_token_accuracy": 0.16421406120061874,
      "num_tokens": 25411922.0,
      "step": 10030
    },
    {
      "entropy": 6.325707006454468,
      "epoch": 1.1581073283323717,
      "grad_norm": 1.921875,
      "learning_rate": 0.00048776003311771013,
      "loss": 6.1078,
      "mean_token_accuracy": 0.1685599520802498,
      "num_tokens": 25425430.0,
      "step": 10035
    },
    {
      "entropy": 6.348438787460327,
      "epoch": 1.1586843623773802,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004877466049604238,
      "loss": 6.2179,
      "mean_token_accuracy": 0.154144524037838,
      "num_tokens": 25437222.0,
      "step": 10040
    },
    {
      "entropy": 6.196552085876465,
      "epoch": 1.159261396422389,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048773316964751106,
      "loss": 6.0204,
      "mean_token_accuracy": 0.1722936064004898,
      "num_tokens": 25448173.0,
      "step": 10045
    },
    {
      "entropy": 6.174867582321167,
      "epoch": 1.1598384304673977,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004877197271794239,
      "loss": 6.0511,
      "mean_token_accuracy": 0.17796580046415328,
      "num_tokens": 25458868.0,
      "step": 10050
    },
    {
      "entropy": 6.262329244613648,
      "epoch": 1.1604154645124063,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004877062775566142,
      "loss": 6.1376,
      "mean_token_accuracy": 0.16303899139165878,
      "num_tokens": 25470986.0,
      "step": 10055
    },
    {
      "entropy": 6.378747224807739,
      "epoch": 1.160992498557415,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048769282077953464,
      "loss": 6.2259,
      "mean_token_accuracy": 0.15943869203329086,
      "num_tokens": 25483484.0,
      "step": 10060
    },
    {
      "entropy": 6.373404788970947,
      "epoch": 1.1615695326024236,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048767935684863775,
      "loss": 6.1554,
      "mean_token_accuracy": 0.16581090092658995,
      "num_tokens": 25496051.0,
      "step": 10065
    },
    {
      "entropy": 6.19994158744812,
      "epoch": 1.1621465666474322,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004876658857643762,
      "loss": 6.0527,
      "mean_token_accuracy": 0.1677911564707756,
      "num_tokens": 25508543.0,
      "step": 10070
    },
    {
      "entropy": 6.279211759567261,
      "epoch": 1.1627236006924409,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004876524075272034,
      "loss": 6.1834,
      "mean_token_accuracy": 0.16343737542629241,
      "num_tokens": 25521306.0,
      "step": 10075
    },
    {
      "entropy": 6.2529863834381105,
      "epoch": 1.1633006347374495,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048763892213757234,
      "loss": 6.0332,
      "mean_token_accuracy": 0.17185359299182892,
      "num_tokens": 25533428.0,
      "step": 10080
    },
    {
      "entropy": 6.281757068634033,
      "epoch": 1.1638776687824581,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048762542959593683,
      "loss": 6.0757,
      "mean_token_accuracy": 0.16308882385492324,
      "num_tokens": 25545747.0,
      "step": 10085
    },
    {
      "entropy": 6.238220310211181,
      "epoch": 1.1644547028274668,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004876119299027506,
      "loss": 6.1647,
      "mean_token_accuracy": 0.1584583640098572,
      "num_tokens": 25558460.0,
      "step": 10090
    },
    {
      "entropy": 6.300864171981812,
      "epoch": 1.1650317368724754,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048759842305846766,
      "loss": 6.1152,
      "mean_token_accuracy": 0.16493510603904724,
      "num_tokens": 25571583.0,
      "step": 10095
    },
    {
      "entropy": 6.298920202255249,
      "epoch": 1.165608770917484,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004875849090635425,
      "loss": 6.1204,
      "mean_token_accuracy": 0.1637258291244507,
      "num_tokens": 25586133.0,
      "step": 10100
    },
    {
      "entropy": 6.177766895294189,
      "epoch": 1.1661858049624927,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048757138791842943,
      "loss": 6.0378,
      "mean_token_accuracy": 0.16949152052402497,
      "num_tokens": 25598935.0,
      "step": 10105
    },
    {
      "entropy": 6.263977003097534,
      "epoch": 1.1667628390075016,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004875578596235832,
      "loss": 6.1259,
      "mean_token_accuracy": 0.16436364650726318,
      "num_tokens": 25611400.0,
      "step": 10110
    },
    {
      "entropy": 6.277270126342773,
      "epoch": 1.1673398730525102,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004875443241794591,
      "loss": 6.1053,
      "mean_token_accuracy": 0.1688404768705368,
      "num_tokens": 25623389.0,
      "step": 10115
    },
    {
      "entropy": 6.235487747192383,
      "epoch": 1.1679169070975188,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048753078158651227,
      "loss": 6.1312,
      "mean_token_accuracy": 0.16458784490823747,
      "num_tokens": 25636702.0,
      "step": 10120
    },
    {
      "entropy": 6.189268732070923,
      "epoch": 1.1684939411425275,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000487517231845198,
      "loss": 6.0561,
      "mean_token_accuracy": 0.1708905890583992,
      "num_tokens": 25650216.0,
      "step": 10125
    },
    {
      "entropy": 6.18388409614563,
      "epoch": 1.169070975187536,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004875036749559724,
      "loss": 6.0092,
      "mean_token_accuracy": 0.1782259464263916,
      "num_tokens": 25662948.0,
      "step": 10130
    },
    {
      "entropy": 6.288214063644409,
      "epoch": 1.1696480092325447,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048749011091929115,
      "loss": 6.0795,
      "mean_token_accuracy": 0.16321244239807128,
      "num_tokens": 25676984.0,
      "step": 10135
    },
    {
      "entropy": 6.194680833816529,
      "epoch": 1.1702250432775534,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004874765397356105,
      "loss": 6.0525,
      "mean_token_accuracy": 0.1706096351146698,
      "num_tokens": 25689422.0,
      "step": 10140
    },
    {
      "entropy": 6.191124677658081,
      "epoch": 1.170802077322562,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004874629614053871,
      "loss": 6.0931,
      "mean_token_accuracy": 0.1669308975338936,
      "num_tokens": 25701621.0,
      "step": 10145
    },
    {
      "entropy": 6.270998287200928,
      "epoch": 1.1713791113675707,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004874493759290775,
      "loss": 6.1341,
      "mean_token_accuracy": 0.16456253528594972,
      "num_tokens": 25716102.0,
      "step": 10150
    },
    {
      "entropy": 6.282492828369141,
      "epoch": 1.1719561454125793,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048743578330713854,
      "loss": 6.1704,
      "mean_token_accuracy": 0.16168964952230452,
      "num_tokens": 25728224.0,
      "step": 10155
    },
    {
      "entropy": 6.270199155807495,
      "epoch": 1.172533179457588,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004874221835400277,
      "loss": 6.126,
      "mean_token_accuracy": 0.16836394369602203,
      "num_tokens": 25741709.0,
      "step": 10160
    },
    {
      "entropy": 6.272929048538208,
      "epoch": 1.1731102135025966,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004874085766282022,
      "loss": 6.0588,
      "mean_token_accuracy": 0.1660297155380249,
      "num_tokens": 25754565.0,
      "step": 10165
    },
    {
      "entropy": 6.275498867034912,
      "epoch": 1.1736872475476052,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048739496257211966,
      "loss": 6.1557,
      "mean_token_accuracy": 0.1586190477013588,
      "num_tokens": 25767329.0,
      "step": 10170
    },
    {
      "entropy": 6.2716052532196045,
      "epoch": 1.1742642815926139,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048738134137223815,
      "loss": 6.0867,
      "mean_token_accuracy": 0.17010551542043686,
      "num_tokens": 25780253.0,
      "step": 10175
    },
    {
      "entropy": 6.2237632274627686,
      "epoch": 1.1748413156376225,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048736771302901566,
      "loss": 6.0603,
      "mean_token_accuracy": 0.16766718327999114,
      "num_tokens": 25792712.0,
      "step": 10180
    },
    {
      "entropy": 6.1873047828674315,
      "epoch": 1.1754183496826314,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048735407754291063,
      "loss": 6.0936,
      "mean_token_accuracy": 0.1663523480296135,
      "num_tokens": 25804735.0,
      "step": 10185
    },
    {
      "entropy": 6.259357833862305,
      "epoch": 1.17599538372764,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048734043491438175,
      "loss": 6.1105,
      "mean_token_accuracy": 0.1677462413907051,
      "num_tokens": 25816880.0,
      "step": 10190
    },
    {
      "entropy": 6.201093435287476,
      "epoch": 1.1765724177726486,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048732678514388773,
      "loss": 6.0379,
      "mean_token_accuracy": 0.16589249819517135,
      "num_tokens": 25829756.0,
      "step": 10195
    },
    {
      "entropy": 6.155019617080688,
      "epoch": 1.1771494518176573,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004873131282318878,
      "loss": 6.0052,
      "mean_token_accuracy": 0.17416342943906785,
      "num_tokens": 25842318.0,
      "step": 10200
    },
    {
      "entropy": 6.21946349143982,
      "epoch": 1.177726485862666,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048729946417884126,
      "loss": 6.0021,
      "mean_token_accuracy": 0.16925073713064193,
      "num_tokens": 25854716.0,
      "step": 10205
    },
    {
      "entropy": 6.276928567886353,
      "epoch": 1.1783035199076746,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004872857929852076,
      "loss": 6.0471,
      "mean_token_accuracy": 0.1733916789293289,
      "num_tokens": 25867678.0,
      "step": 10210
    },
    {
      "entropy": 6.205417490005493,
      "epoch": 1.1788805539526832,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004872721146514469,
      "loss": 6.0861,
      "mean_token_accuracy": 0.16271694898605346,
      "num_tokens": 25880845.0,
      "step": 10215
    },
    {
      "entropy": 6.141888380050659,
      "epoch": 1.1794575879976918,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000487258429178019,
      "loss": 6.0396,
      "mean_token_accuracy": 0.1721838817000389,
      "num_tokens": 25893693.0,
      "step": 10220
    },
    {
      "entropy": 6.298300409317017,
      "epoch": 1.1800346220427005,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048724473656538427,
      "loss": 6.1272,
      "mean_token_accuracy": 0.15936866849660875,
      "num_tokens": 25906752.0,
      "step": 10225
    },
    {
      "entropy": 6.273559617996216,
      "epoch": 1.1806116560877091,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004872310368140032,
      "loss": 6.1214,
      "mean_token_accuracy": 0.16647805720567704,
      "num_tokens": 25918343.0,
      "step": 10230
    },
    {
      "entropy": 6.315348052978516,
      "epoch": 1.1811886901327178,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004872173299243368,
      "loss": 6.1131,
      "mean_token_accuracy": 0.15925135165452958,
      "num_tokens": 25930521.0,
      "step": 10235
    },
    {
      "entropy": 6.283903694152832,
      "epoch": 1.1817657241777264,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048720361589684575,
      "loss": 6.1011,
      "mean_token_accuracy": 0.15997551679611205,
      "num_tokens": 25942618.0,
      "step": 10240
    },
    {
      "entropy": 6.2160004615783695,
      "epoch": 1.182342758222735,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048718989473199147,
      "loss": 6.1704,
      "mean_token_accuracy": 0.16155589371919632,
      "num_tokens": 25956068.0,
      "step": 10245
    },
    {
      "entropy": 6.2736945152282715,
      "epoch": 1.182919792267744,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004871761664302356,
      "loss": 6.07,
      "mean_token_accuracy": 0.1746903821825981,
      "num_tokens": 25967639.0,
      "step": 10250
    },
    {
      "entropy": 6.2149364948272705,
      "epoch": 1.1834968263127525,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004871624309920397,
      "loss": 6.0223,
      "mean_token_accuracy": 0.17545360177755356,
      "num_tokens": 25980493.0,
      "step": 10255
    },
    {
      "entropy": 6.297014141082764,
      "epoch": 1.1840738603577612,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048714868841786586,
      "loss": 6.1238,
      "mean_token_accuracy": 0.15757402926683425,
      "num_tokens": 25994414.0,
      "step": 10260
    },
    {
      "entropy": 6.323869085311889,
      "epoch": 1.1846508944027698,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048713493870817626,
      "loss": 6.1471,
      "mean_token_accuracy": 0.16203411743044854,
      "num_tokens": 26008510.0,
      "step": 10265
    },
    {
      "entropy": 6.266760921478271,
      "epoch": 1.1852279284477785,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048712118186343336,
      "loss": 6.1423,
      "mean_token_accuracy": 0.15995179116725922,
      "num_tokens": 26022682.0,
      "step": 10270
    },
    {
      "entropy": 6.271547889709472,
      "epoch": 1.185804962492787,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048710741788409985,
      "loss": 6.1139,
      "mean_token_accuracy": 0.166634601354599,
      "num_tokens": 26035476.0,
      "step": 10275
    },
    {
      "entropy": 6.2775876998901365,
      "epoch": 1.1863819965377957,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004870936467706387,
      "loss": 6.1212,
      "mean_token_accuracy": 0.16022894531488419,
      "num_tokens": 26049253.0,
      "step": 10280
    },
    {
      "entropy": 6.258497095108032,
      "epoch": 1.1869590305828044,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004870798685235131,
      "loss": 6.1207,
      "mean_token_accuracy": 0.1675887554883957,
      "num_tokens": 26061343.0,
      "step": 10285
    },
    {
      "entropy": 6.270390176773072,
      "epoch": 1.187536064627813,
      "grad_norm": 2.15625,
      "learning_rate": 0.00048706608314318654,
      "loss": 6.113,
      "mean_token_accuracy": 0.15879189521074294,
      "num_tokens": 26074133.0,
      "step": 10290
    },
    {
      "entropy": 6.29678258895874,
      "epoch": 1.1881130986728217,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004870522906301225,
      "loss": 6.0643,
      "mean_token_accuracy": 0.1635904222726822,
      "num_tokens": 26086706.0,
      "step": 10295
    },
    {
      "entropy": 6.240001153945923,
      "epoch": 1.1886901327178303,
      "grad_norm": 1.375,
      "learning_rate": 0.000487038490984785,
      "loss": 6.0388,
      "mean_token_accuracy": 0.16647442430257797,
      "num_tokens": 26100074.0,
      "step": 10300
    },
    {
      "entropy": 6.2251519680023195,
      "epoch": 1.189267166762839,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048702468420763826,
      "loss": 6.1345,
      "mean_token_accuracy": 0.1652000814676285,
      "num_tokens": 26112758.0,
      "step": 10305
    },
    {
      "entropy": 6.25880389213562,
      "epoch": 1.1898442008078476,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048701087029914657,
      "loss": 6.1429,
      "mean_token_accuracy": 0.164242784678936,
      "num_tokens": 26126088.0,
      "step": 10310
    },
    {
      "entropy": 6.288421106338501,
      "epoch": 1.1904212348528562,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004869970492597745,
      "loss": 6.1321,
      "mean_token_accuracy": 0.15779225826263427,
      "num_tokens": 26137793.0,
      "step": 10315
    },
    {
      "entropy": 6.27298526763916,
      "epoch": 1.1909982688978649,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004869832210899871,
      "loss": 6.0865,
      "mean_token_accuracy": 0.1696375072002411,
      "num_tokens": 26150975.0,
      "step": 10320
    },
    {
      "entropy": 6.272920036315918,
      "epoch": 1.1915753029428737,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048696938579024927,
      "loss": 6.0971,
      "mean_token_accuracy": 0.15676479637622834,
      "num_tokens": 26162658.0,
      "step": 10325
    },
    {
      "entropy": 6.294800424575806,
      "epoch": 1.1921523369878824,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048695554336102644,
      "loss": 6.1457,
      "mean_token_accuracy": 0.1658876806497574,
      "num_tokens": 26176089.0,
      "step": 10330
    },
    {
      "entropy": 6.296675777435302,
      "epoch": 1.192729371032891,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048694169380278417,
      "loss": 6.1649,
      "mean_token_accuracy": 0.16248101592063904,
      "num_tokens": 26188974.0,
      "step": 10335
    },
    {
      "entropy": 6.280240726470947,
      "epoch": 1.1933064050778996,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004869278371159884,
      "loss": 6.1013,
      "mean_token_accuracy": 0.16119649708271028,
      "num_tokens": 26201628.0,
      "step": 10340
    },
    {
      "entropy": 6.2404388904571535,
      "epoch": 1.1938834391229083,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048691397330110503,
      "loss": 5.9991,
      "mean_token_accuracy": 0.17240139693021775,
      "num_tokens": 26213540.0,
      "step": 10345
    },
    {
      "entropy": 6.291774702072144,
      "epoch": 1.194460473167917,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004869001023586005,
      "loss": 6.1687,
      "mean_token_accuracy": 0.1628555178642273,
      "num_tokens": 26226476.0,
      "step": 10350
    },
    {
      "entropy": 6.161723518371582,
      "epoch": 1.1950375072129256,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004868862242889413,
      "loss": 6.0501,
      "mean_token_accuracy": 0.16851481646299363,
      "num_tokens": 26238854.0,
      "step": 10355
    },
    {
      "entropy": 6.244963788986206,
      "epoch": 1.1956145412579342,
      "grad_norm": 1.078125,
      "learning_rate": 0.000486872339092594,
      "loss": 6.039,
      "mean_token_accuracy": 0.17083438485860825,
      "num_tokens": 26251279.0,
      "step": 10360
    },
    {
      "entropy": 6.369075584411621,
      "epoch": 1.1961915753029428,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000486858446770026,
      "loss": 6.1254,
      "mean_token_accuracy": 0.15898502990603447,
      "num_tokens": 26263412.0,
      "step": 10365
    },
    {
      "entropy": 6.1898352146148685,
      "epoch": 1.1967686093479515,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004868445473217043,
      "loss": 6.1093,
      "mean_token_accuracy": 0.16251132339239122,
      "num_tokens": 26276085.0,
      "step": 10370
    },
    {
      "entropy": 6.259449243545532,
      "epoch": 1.1973456433929601,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004868306407480965,
      "loss": 6.1561,
      "mean_token_accuracy": 0.1609532877802849,
      "num_tokens": 26288856.0,
      "step": 10375
    },
    {
      "entropy": 6.265676259994507,
      "epoch": 1.1979226774379688,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048681672704967036,
      "loss": 6.0517,
      "mean_token_accuracy": 0.173240265250206,
      "num_tokens": 26300346.0,
      "step": 10380
    },
    {
      "entropy": 6.248412179946899,
      "epoch": 1.1984997114829774,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004868028062268938,
      "loss": 6.105,
      "mean_token_accuracy": 0.1706072822213173,
      "num_tokens": 26313414.0,
      "step": 10385
    },
    {
      "entropy": 6.191400241851807,
      "epoch": 1.1990767455279863,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048678887828023504,
      "loss": 6.152,
      "mean_token_accuracy": 0.1635143905878067,
      "num_tokens": 26325779.0,
      "step": 10390
    },
    {
      "entropy": 6.324222803115845,
      "epoch": 1.1996537795729947,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004867749432101626,
      "loss": 6.1088,
      "mean_token_accuracy": 0.1593686670064926,
      "num_tokens": 26337513.0,
      "step": 10395
    },
    {
      "entropy": 6.30271315574646,
      "epoch": 1.2002308136180035,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004867610010171451,
      "loss": 6.1479,
      "mean_token_accuracy": 0.16110331267118455,
      "num_tokens": 26350569.0,
      "step": 10400
    },
    {
      "entropy": 6.273625278472901,
      "epoch": 1.2008078476630122,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048674705170165147,
      "loss": 6.1126,
      "mean_token_accuracy": 0.16242406368255616,
      "num_tokens": 26362989.0,
      "step": 10405
    },
    {
      "entropy": 6.203810501098633,
      "epoch": 1.2013848817080208,
      "grad_norm": 1.046875,
      "learning_rate": 0.000486733095264151,
      "loss": 6.0159,
      "mean_token_accuracy": 0.17562277168035506,
      "num_tokens": 26376671.0,
      "step": 10410
    },
    {
      "entropy": 6.259033441543579,
      "epoch": 1.2019619157530295,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048671913170511306,
      "loss": 6.1245,
      "mean_token_accuracy": 0.16710228472948074,
      "num_tokens": 26389437.0,
      "step": 10415
    },
    {
      "entropy": 6.269059085845948,
      "epoch": 1.202538949798038,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004867051610250073,
      "loss": 6.0952,
      "mean_token_accuracy": 0.16706571877002716,
      "num_tokens": 26401441.0,
      "step": 10420
    },
    {
      "entropy": 6.201774930953979,
      "epoch": 1.2031159838430467,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004866911832243035,
      "loss": 6.0622,
      "mean_token_accuracy": 0.1702714204788208,
      "num_tokens": 26413062.0,
      "step": 10425
    },
    {
      "entropy": 6.253553771972657,
      "epoch": 1.2036930178880554,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048667719830347203,
      "loss": 6.1039,
      "mean_token_accuracy": 0.1693117141723633,
      "num_tokens": 26425822.0,
      "step": 10430
    },
    {
      "entropy": 6.244201755523681,
      "epoch": 1.204270051933064,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048666320626298307,
      "loss": 6.0761,
      "mean_token_accuracy": 0.1654995821416378,
      "num_tokens": 26438226.0,
      "step": 10435
    },
    {
      "entropy": 6.245728349685669,
      "epoch": 1.2048470859780727,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048664920710330735,
      "loss": 6.1171,
      "mean_token_accuracy": 0.16892891228199006,
      "num_tokens": 26450851.0,
      "step": 10440
    },
    {
      "entropy": 6.30135407447815,
      "epoch": 1.2054241200230813,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048663520082491564,
      "loss": 6.2018,
      "mean_token_accuracy": 0.16059584319591522,
      "num_tokens": 26463292.0,
      "step": 10445
    },
    {
      "entropy": 6.319532775878907,
      "epoch": 1.20600115406809,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004866211874282791,
      "loss": 6.15,
      "mean_token_accuracy": 0.16495329290628433,
      "num_tokens": 26476731.0,
      "step": 10450
    },
    {
      "entropy": 6.198712778091431,
      "epoch": 1.2065781881130986,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000486607166913869,
      "loss": 6.0847,
      "mean_token_accuracy": 0.16421911269426345,
      "num_tokens": 26489295.0,
      "step": 10455
    },
    {
      "entropy": 6.247257566452026,
      "epoch": 1.2071552221581072,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048659313928215705,
      "loss": 6.1015,
      "mean_token_accuracy": 0.16043473184108734,
      "num_tokens": 26502729.0,
      "step": 10460
    },
    {
      "entropy": 6.263458347320556,
      "epoch": 1.207732256203116,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004865791045336149,
      "loss": 6.0843,
      "mean_token_accuracy": 0.16971976011991502,
      "num_tokens": 26517442.0,
      "step": 10465
    },
    {
      "entropy": 6.333164072036743,
      "epoch": 1.2083092902481247,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004865650626687146,
      "loss": 6.0683,
      "mean_token_accuracy": 0.16716044619679452,
      "num_tokens": 26530980.0,
      "step": 10470
    },
    {
      "entropy": 6.211981630325317,
      "epoch": 1.2088863242931334,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048655101368792866,
      "loss": 6.0671,
      "mean_token_accuracy": 0.16627950817346573,
      "num_tokens": 26542143.0,
      "step": 10475
    },
    {
      "entropy": 6.244042015075683,
      "epoch": 1.209463358338142,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004865369575917293,
      "loss": 6.1528,
      "mean_token_accuracy": 0.15941849797964097,
      "num_tokens": 26555193.0,
      "step": 10480
    },
    {
      "entropy": 6.296036672592163,
      "epoch": 1.2100403923831506,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048652289438058953,
      "loss": 6.1,
      "mean_token_accuracy": 0.17024942636489868,
      "num_tokens": 26567015.0,
      "step": 10485
    },
    {
      "entropy": 6.269211673736573,
      "epoch": 1.2106174264281593,
      "grad_norm": 1.125,
      "learning_rate": 0.00048650882405498226,
      "loss": 6.1278,
      "mean_token_accuracy": 0.16495791524648667,
      "num_tokens": 26579155.0,
      "step": 10490
    },
    {
      "entropy": 6.273886013031006,
      "epoch": 1.211194460473168,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004864947466153808,
      "loss": 6.113,
      "mean_token_accuracy": 0.16628367602825164,
      "num_tokens": 26592552.0,
      "step": 10495
    },
    {
      "entropy": 6.306654548645019,
      "epoch": 1.2117714945181766,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004864806620622585,
      "loss": 6.1224,
      "mean_token_accuracy": 0.16425371915102005,
      "num_tokens": 26604820.0,
      "step": 10500
    },
    {
      "entropy": 6.277959966659546,
      "epoch": 1.2123485285631852,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004864665703960892,
      "loss": 6.1136,
      "mean_token_accuracy": 0.16500537544488908,
      "num_tokens": 26617218.0,
      "step": 10505
    },
    {
      "entropy": 6.1786338806152346,
      "epoch": 1.2129255626081938,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004864524716173469,
      "loss": 6.029,
      "mean_token_accuracy": 0.16454496830701829,
      "num_tokens": 26628929.0,
      "step": 10510
    },
    {
      "entropy": 6.24074420928955,
      "epoch": 1.2135025966532025,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004864383657265058,
      "loss": 6.1196,
      "mean_token_accuracy": 0.16185372918844224,
      "num_tokens": 26640983.0,
      "step": 10515
    },
    {
      "entropy": 6.2854310989379885,
      "epoch": 1.2140796306982111,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048642425272404016,
      "loss": 6.1315,
      "mean_token_accuracy": 0.16684163957834244,
      "num_tokens": 26653931.0,
      "step": 10520
    },
    {
      "entropy": 6.2561181545257565,
      "epoch": 1.2146566647432198,
      "grad_norm": 1.125,
      "learning_rate": 0.0004864101326104248,
      "loss": 6.1087,
      "mean_token_accuracy": 0.16679566949605942,
      "num_tokens": 26666316.0,
      "step": 10525
    },
    {
      "entropy": 6.294846725463867,
      "epoch": 1.2152336987882286,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004863960053861348,
      "loss": 6.1256,
      "mean_token_accuracy": 0.16777746826410295,
      "num_tokens": 26679818.0,
      "step": 10530
    },
    {
      "entropy": 6.284996414184571,
      "epoch": 1.215810732833237,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048638187105164507,
      "loss": 6.1108,
      "mean_token_accuracy": 0.17611703127622605,
      "num_tokens": 26693553.0,
      "step": 10535
    },
    {
      "entropy": 6.235351085662842,
      "epoch": 1.216387766878246,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004863677296074311,
      "loss": 6.1045,
      "mean_token_accuracy": 0.16603602170944215,
      "num_tokens": 26706037.0,
      "step": 10540
    },
    {
      "entropy": 6.3097106456756595,
      "epoch": 1.2169648009232545,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004863535810539686,
      "loss": 6.1411,
      "mean_token_accuracy": 0.16210662722587585,
      "num_tokens": 26719071.0,
      "step": 10545
    },
    {
      "entropy": 6.199039030075073,
      "epoch": 1.2175418349682632,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048633942539173325,
      "loss": 6.0356,
      "mean_token_accuracy": 0.1667512074112892,
      "num_tokens": 26731367.0,
      "step": 10550
    },
    {
      "entropy": 6.224197292327881,
      "epoch": 1.2181188690132718,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048632526262120144,
      "loss": 6.0754,
      "mean_token_accuracy": 0.1683653026819229,
      "num_tokens": 26744945.0,
      "step": 10555
    },
    {
      "entropy": 6.2178281307220455,
      "epoch": 1.2186959030582805,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004863110927428493,
      "loss": 6.1748,
      "mean_token_accuracy": 0.15920467376708985,
      "num_tokens": 26758237.0,
      "step": 10560
    },
    {
      "entropy": 6.401515626907349,
      "epoch": 1.219272937103289,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004862969157571536,
      "loss": 6.1549,
      "mean_token_accuracy": 0.16242912411689758,
      "num_tokens": 26770172.0,
      "step": 10565
    },
    {
      "entropy": 6.329786062240601,
      "epoch": 1.2198499711482977,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048628273166459105,
      "loss": 6.0834,
      "mean_token_accuracy": 0.16931937038898467,
      "num_tokens": 26783152.0,
      "step": 10570
    },
    {
      "entropy": 6.194137001037598,
      "epoch": 1.2204270051933064,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048626854046563876,
      "loss": 6.1034,
      "mean_token_accuracy": 0.1694098263978958,
      "num_tokens": 26795023.0,
      "step": 10575
    },
    {
      "entropy": 6.298760414123535,
      "epoch": 1.221004039238315,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048625434216077404,
      "loss": 6.1518,
      "mean_token_accuracy": 0.1617591693997383,
      "num_tokens": 26807715.0,
      "step": 10580
    },
    {
      "entropy": 6.288610363006592,
      "epoch": 1.2215810732833237,
      "grad_norm": 1.75,
      "learning_rate": 0.00048624013675047453,
      "loss": 6.1162,
      "mean_token_accuracy": 0.15889105349779128,
      "num_tokens": 26819181.0,
      "step": 10585
    },
    {
      "entropy": 6.240978765487671,
      "epoch": 1.2221581073283323,
      "grad_norm": 1.25,
      "learning_rate": 0.00048622592423521783,
      "loss": 6.0834,
      "mean_token_accuracy": 0.16744539737701417,
      "num_tokens": 26831039.0,
      "step": 10590
    },
    {
      "entropy": 6.176622295379639,
      "epoch": 1.222735141373341,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004862117046154822,
      "loss": 5.9952,
      "mean_token_accuracy": 0.16880216747522353,
      "num_tokens": 26843715.0,
      "step": 10595
    },
    {
      "entropy": 6.2198052406311035,
      "epoch": 1.2233121754183496,
      "grad_norm": 1.125,
      "learning_rate": 0.00048619747789174577,
      "loss": 5.9808,
      "mean_token_accuracy": 0.17575788795948027,
      "num_tokens": 26855856.0,
      "step": 10600
    },
    {
      "entropy": 6.227771759033203,
      "epoch": 1.2238892094633584,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004861832440644871,
      "loss": 6.1455,
      "mean_token_accuracy": 0.1645462304353714,
      "num_tokens": 26869368.0,
      "step": 10605
    },
    {
      "entropy": 6.134281778335572,
      "epoch": 1.224466243508367,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048616900313418486,
      "loss": 5.9456,
      "mean_token_accuracy": 0.17228889912366868,
      "num_tokens": 26882230.0,
      "step": 10610
    },
    {
      "entropy": 6.255297422409058,
      "epoch": 1.2250432775533757,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048615475510131815,
      "loss": 6.0865,
      "mean_token_accuracy": 0.16498906165361404,
      "num_tokens": 26896307.0,
      "step": 10615
    },
    {
      "entropy": 6.272714900970459,
      "epoch": 1.2256203115983844,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048614049996636614,
      "loss": 6.1495,
      "mean_token_accuracy": 0.16476012468338014,
      "num_tokens": 26908759.0,
      "step": 10620
    },
    {
      "entropy": 6.255472707748413,
      "epoch": 1.226197345643393,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048612623772980825,
      "loss": 6.0879,
      "mean_token_accuracy": 0.170181205868721,
      "num_tokens": 26922179.0,
      "step": 10625
    },
    {
      "entropy": 6.265120172500611,
      "epoch": 1.2267743796884016,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048611196839212426,
      "loss": 6.0777,
      "mean_token_accuracy": 0.1617940381169319,
      "num_tokens": 26934225.0,
      "step": 10630
    },
    {
      "entropy": 6.21250548362732,
      "epoch": 1.2273514137334103,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000486097691953794,
      "loss": 6.0747,
      "mean_token_accuracy": 0.16690510958433152,
      "num_tokens": 26946644.0,
      "step": 10635
    },
    {
      "entropy": 6.306379222869873,
      "epoch": 1.227928447778419,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048608340841529784,
      "loss": 6.1627,
      "mean_token_accuracy": 0.16227346509695054,
      "num_tokens": 26959504.0,
      "step": 10640
    },
    {
      "entropy": 6.3103962421417235,
      "epoch": 1.2285054818234276,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000486069117777116,
      "loss": 6.1125,
      "mean_token_accuracy": 0.16754785180091858,
      "num_tokens": 26972274.0,
      "step": 10645
    },
    {
      "entropy": 6.2478917121887205,
      "epoch": 1.2290825158684362,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004860548200397293,
      "loss": 6.1413,
      "mean_token_accuracy": 0.16638704985380173,
      "num_tokens": 26985800.0,
      "step": 10650
    },
    {
      "entropy": 6.298679828643799,
      "epoch": 1.2296595499134448,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048604051520361846,
      "loss": 6.1319,
      "mean_token_accuracy": 0.1631200835108757,
      "num_tokens": 26999058.0,
      "step": 10655
    },
    {
      "entropy": 6.293829154968262,
      "epoch": 1.2302365839584535,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048602620326926484,
      "loss": 6.1311,
      "mean_token_accuracy": 0.17051169276237488,
      "num_tokens": 27010142.0,
      "step": 10660
    },
    {
      "entropy": 6.214754104614258,
      "epoch": 1.2308136180034621,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048601188423714965,
      "loss": 6.0308,
      "mean_token_accuracy": 0.166061632335186,
      "num_tokens": 27021547.0,
      "step": 10665
    },
    {
      "entropy": 6.278275871276856,
      "epoch": 1.231390652048471,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004859975581077545,
      "loss": 6.1063,
      "mean_token_accuracy": 0.16576358377933503,
      "num_tokens": 27034817.0,
      "step": 10670
    },
    {
      "entropy": 6.226116514205932,
      "epoch": 1.2319676860934794,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004859832248815614,
      "loss": 6.0238,
      "mean_token_accuracy": 0.1711491972208023,
      "num_tokens": 27047567.0,
      "step": 10675
    },
    {
      "entropy": 6.25311689376831,
      "epoch": 1.2325447201384883,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004859688845590522,
      "loss": 6.1842,
      "mean_token_accuracy": 0.15874741822481156,
      "num_tokens": 27060862.0,
      "step": 10680
    },
    {
      "entropy": 6.229175615310669,
      "epoch": 1.233121754183497,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048595453714070944,
      "loss": 6.1071,
      "mean_token_accuracy": 0.1741679847240448,
      "num_tokens": 27073216.0,
      "step": 10685
    },
    {
      "entropy": 6.224675464630127,
      "epoch": 1.2336987882285055,
      "grad_norm": 1.125,
      "learning_rate": 0.0004859401826270156,
      "loss": 6.0765,
      "mean_token_accuracy": 0.16757579892873764,
      "num_tokens": 27085640.0,
      "step": 10690
    },
    {
      "entropy": 6.289561128616333,
      "epoch": 1.2342758222735142,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004859258210184536,
      "loss": 6.1214,
      "mean_token_accuracy": 0.16126213371753692,
      "num_tokens": 27097814.0,
      "step": 10695
    },
    {
      "entropy": 6.303145170211792,
      "epoch": 1.2348528563185228,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048591145231550623,
      "loss": 6.1511,
      "mean_token_accuracy": 0.1597980484366417,
      "num_tokens": 27110449.0,
      "step": 10700
    },
    {
      "entropy": 6.331226587295532,
      "epoch": 1.2354298903635315,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048589707651865697,
      "loss": 6.1578,
      "mean_token_accuracy": 0.16326843500137328,
      "num_tokens": 27123035.0,
      "step": 10705
    },
    {
      "entropy": 6.187960243225097,
      "epoch": 1.23600692440854,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004858826936283893,
      "loss": 6.0458,
      "mean_token_accuracy": 0.1702803671360016,
      "num_tokens": 27134794.0,
      "step": 10710
    },
    {
      "entropy": 6.2434814929962155,
      "epoch": 1.2365839584535487,
      "grad_norm": 1.109375,
      "learning_rate": 0.000485868303645187,
      "loss": 6.0907,
      "mean_token_accuracy": 0.16455400735139847,
      "num_tokens": 27147492.0,
      "step": 10715
    },
    {
      "entropy": 6.23840012550354,
      "epoch": 1.2371609924985574,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048585390656953397,
      "loss": 6.1228,
      "mean_token_accuracy": 0.16601805686950682,
      "num_tokens": 27159453.0,
      "step": 10720
    },
    {
      "entropy": 6.269198560714722,
      "epoch": 1.237738026543566,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004858395024019147,
      "loss": 6.0431,
      "mean_token_accuracy": 0.16868919581174852,
      "num_tokens": 27175021.0,
      "step": 10725
    },
    {
      "entropy": 6.238038873672485,
      "epoch": 1.2383150605885747,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004858250911428133,
      "loss": 6.0673,
      "mean_token_accuracy": 0.16851505935192107,
      "num_tokens": 27187612.0,
      "step": 10730
    },
    {
      "entropy": 6.193956470489502,
      "epoch": 1.2388920946335833,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004858106727927148,
      "loss": 6.0397,
      "mean_token_accuracy": 0.164214026927948,
      "num_tokens": 27201045.0,
      "step": 10735
    },
    {
      "entropy": 6.318602085113525,
      "epoch": 1.239469128678592,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000485796247352104,
      "loss": 6.1923,
      "mean_token_accuracy": 0.15878598988056183,
      "num_tokens": 27213934.0,
      "step": 10740
    },
    {
      "entropy": 6.264531946182251,
      "epoch": 1.2400461627236008,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004857818148214662,
      "loss": 6.0462,
      "mean_token_accuracy": 0.16933714598417282,
      "num_tokens": 27225695.0,
      "step": 10745
    },
    {
      "entropy": 6.2596455097198485,
      "epoch": 1.2406231967686094,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004857673752012866,
      "loss": 6.0864,
      "mean_token_accuracy": 0.16722105145454408,
      "num_tokens": 27238065.0,
      "step": 10750
    },
    {
      "entropy": 6.252855968475342,
      "epoch": 1.241200230813618,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048575292849205114,
      "loss": 6.0901,
      "mean_token_accuracy": 0.16022679209709167,
      "num_tokens": 27250334.0,
      "step": 10755
    },
    {
      "entropy": 6.291858863830567,
      "epoch": 1.2417772648586267,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004857384746942456,
      "loss": 6.0676,
      "mean_token_accuracy": 0.171857488155365,
      "num_tokens": 27262397.0,
      "step": 10760
    },
    {
      "entropy": 6.265959548950195,
      "epoch": 1.2423542989036354,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004857240138083562,
      "loss": 6.0716,
      "mean_token_accuracy": 0.16098155379295348,
      "num_tokens": 27274691.0,
      "step": 10765
    },
    {
      "entropy": 6.234134531021118,
      "epoch": 1.242931332948644,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004857095458348692,
      "loss": 6.1207,
      "mean_token_accuracy": 0.16318233907222748,
      "num_tokens": 27286970.0,
      "step": 10770
    },
    {
      "entropy": 6.165913343429565,
      "epoch": 1.2435083669936526,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048569507077427134,
      "loss": 5.9676,
      "mean_token_accuracy": 0.1725240245461464,
      "num_tokens": 27299788.0,
      "step": 10775
    },
    {
      "entropy": 6.251590585708618,
      "epoch": 1.2440854010386613,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004856805886270494,
      "loss": 5.9637,
      "mean_token_accuracy": 0.1734328970313072,
      "num_tokens": 27312040.0,
      "step": 10780
    },
    {
      "entropy": 6.259579992294311,
      "epoch": 1.24466243508367,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004856660993936905,
      "loss": 6.2689,
      "mean_token_accuracy": 0.16384754776954652,
      "num_tokens": 27325583.0,
      "step": 10785
    },
    {
      "entropy": 6.222048044204712,
      "epoch": 1.2452394691286786,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000485651603074682,
      "loss": 6.0892,
      "mean_token_accuracy": 0.16639239192008973,
      "num_tokens": 27337525.0,
      "step": 10790
    },
    {
      "entropy": 6.24515209197998,
      "epoch": 1.2458165031736872,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004856370996705115,
      "loss": 6.1217,
      "mean_token_accuracy": 0.1600983053445816,
      "num_tokens": 27351368.0,
      "step": 10795
    },
    {
      "entropy": 6.299312162399292,
      "epoch": 1.2463935372186958,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004856225891816667,
      "loss": 6.0826,
      "mean_token_accuracy": 0.16464124917984008,
      "num_tokens": 27364497.0,
      "step": 10800
    },
    {
      "entropy": 6.274144649505615,
      "epoch": 1.2469705712637045,
      "grad_norm": 1.375,
      "learning_rate": 0.0004856080716086358,
      "loss": 6.0713,
      "mean_token_accuracy": 0.16515886336565017,
      "num_tokens": 27377401.0,
      "step": 10805
    },
    {
      "entropy": 6.2394935131073,
      "epoch": 1.2475476053087133,
      "grad_norm": 1.390625,
      "learning_rate": 0.000485593546951907,
      "loss": 6.0496,
      "mean_token_accuracy": 0.1699673518538475,
      "num_tokens": 27389712.0,
      "step": 10810
    },
    {
      "entropy": 6.193137741088867,
      "epoch": 1.2481246393537218,
      "grad_norm": 1.375,
      "learning_rate": 0.0004855790152119688,
      "loss": 6.0764,
      "mean_token_accuracy": 0.1644223541021347,
      "num_tokens": 27403193.0,
      "step": 10815
    },
    {
      "entropy": 6.265658283233643,
      "epoch": 1.2487016733987306,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004855644763893102,
      "loss": 6.1173,
      "mean_token_accuracy": 0.16088301986455916,
      "num_tokens": 27415648.0,
      "step": 10820
    },
    {
      "entropy": 6.294152021408081,
      "epoch": 1.2492787074437393,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048554993048441987,
      "loss": 6.0886,
      "mean_token_accuracy": 0.16649938821792604,
      "num_tokens": 27428825.0,
      "step": 10825
    },
    {
      "entropy": 6.2872052669525145,
      "epoch": 1.249855741488748,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004855353774977873,
      "loss": 6.074,
      "mean_token_accuracy": 0.17564499825239183,
      "num_tokens": 27441867.0,
      "step": 10830
    },
    {
      "entropy": 6.212026071548462,
      "epoch": 1.2504327755337565,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048552081742990184,
      "loss": 6.0925,
      "mean_token_accuracy": 0.16412311941385269,
      "num_tokens": 27454688.0,
      "step": 10835
    },
    {
      "entropy": 6.265987920761108,
      "epoch": 1.2510098095787652,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048550625028125327,
      "loss": 6.199,
      "mean_token_accuracy": 0.15844877809286118,
      "num_tokens": 27466585.0,
      "step": 10840
    },
    {
      "entropy": 6.28060622215271,
      "epoch": 1.2515868436237738,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004854916760523315,
      "loss": 6.1239,
      "mean_token_accuracy": 0.16734313368797302,
      "num_tokens": 27479111.0,
      "step": 10845
    },
    {
      "entropy": 6.235991621017456,
      "epoch": 1.2521638776687825,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048547709474362684,
      "loss": 6.0345,
      "mean_token_accuracy": 0.1700179859995842,
      "num_tokens": 27491359.0,
      "step": 10850
    },
    {
      "entropy": 6.248804903030395,
      "epoch": 1.252740911713791,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004854625063556296,
      "loss": 6.0839,
      "mean_token_accuracy": 0.16641808450222015,
      "num_tokens": 27504609.0,
      "step": 10855
    },
    {
      "entropy": 6.267738389968872,
      "epoch": 1.2533179457587997,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004854479108888305,
      "loss": 6.1004,
      "mean_token_accuracy": 0.16285301446914674,
      "num_tokens": 27517270.0,
      "step": 10860
    },
    {
      "entropy": 6.246438789367676,
      "epoch": 1.2538949798038084,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048543330834372047,
      "loss": 6.0571,
      "mean_token_accuracy": 0.16470762640237807,
      "num_tokens": 27530019.0,
      "step": 10865
    },
    {
      "entropy": 6.257916975021362,
      "epoch": 1.254472013848817,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00048541869872079064,
      "loss": 6.1415,
      "mean_token_accuracy": 0.1631333440542221,
      "num_tokens": 27542091.0,
      "step": 10870
    },
    {
      "entropy": 6.308704900741577,
      "epoch": 1.2550490478938257,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004854040820205324,
      "loss": 6.1534,
      "mean_token_accuracy": 0.1636780947446823,
      "num_tokens": 27554326.0,
      "step": 10875
    },
    {
      "entropy": 6.208969926834106,
      "epoch": 1.2556260819388343,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004853894582434373,
      "loss": 6.1645,
      "mean_token_accuracy": 0.16645378321409227,
      "num_tokens": 27567242.0,
      "step": 10880
    },
    {
      "entropy": 6.221932363510132,
      "epoch": 1.2562031159838432,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004853748273899974,
      "loss": 6.0298,
      "mean_token_accuracy": 0.1725248634815216,
      "num_tokens": 27579595.0,
      "step": 10885
    },
    {
      "entropy": 6.288884401321411,
      "epoch": 1.2567801500288516,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004853601894607046,
      "loss": 6.0726,
      "mean_token_accuracy": 0.1685558781027794,
      "num_tokens": 27592694.0,
      "step": 10890
    },
    {
      "entropy": 6.213584613800049,
      "epoch": 1.2573571840738604,
      "grad_norm": 1.25,
      "learning_rate": 0.0004853455444560514,
      "loss": 5.9973,
      "mean_token_accuracy": 0.1766217514872551,
      "num_tokens": 27605532.0,
      "step": 10895
    },
    {
      "entropy": 6.178055477142334,
      "epoch": 1.257934218118869,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004853308923765302,
      "loss": 6.0757,
      "mean_token_accuracy": 0.16671376079320907,
      "num_tokens": 27618394.0,
      "step": 10900
    },
    {
      "entropy": 6.266275453567505,
      "epoch": 1.2585112521638777,
      "grad_norm": 1.09375,
      "learning_rate": 0.000485316233222634,
      "loss": 6.1017,
      "mean_token_accuracy": 0.16398294419050216,
      "num_tokens": 27631293.0,
      "step": 10905
    },
    {
      "entropy": 6.274036645889282,
      "epoch": 1.2590882862088864,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004853015669948557,
      "loss": 6.1181,
      "mean_token_accuracy": 0.16425407230854033,
      "num_tokens": 27643972.0,
      "step": 10910
    },
    {
      "entropy": 6.1195347785949705,
      "epoch": 1.259665320253895,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048528689369368863,
      "loss": 5.992,
      "mean_token_accuracy": 0.17337488383054733,
      "num_tokens": 27656521.0,
      "step": 10915
    },
    {
      "entropy": 6.256008577346802,
      "epoch": 1.2602423542989036,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004852722133196264,
      "loss": 6.0903,
      "mean_token_accuracy": 0.1645122140645981,
      "num_tokens": 27670108.0,
      "step": 10920
    },
    {
      "entropy": 6.144779920578003,
      "epoch": 1.2608193883439123,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004852575258731627,
      "loss": 6.0111,
      "mean_token_accuracy": 0.17569229900836944,
      "num_tokens": 27683528.0,
      "step": 10925
    },
    {
      "entropy": 6.1580735206604,
      "epoch": 1.261396422388921,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004852428313547916,
      "loss": 5.9548,
      "mean_token_accuracy": 0.17180353552103042,
      "num_tokens": 27696172.0,
      "step": 10930
    },
    {
      "entropy": 6.154382038116455,
      "epoch": 1.2619734564339296,
      "grad_norm": 1.125,
      "learning_rate": 0.00048522812976500726,
      "loss": 6.0442,
      "mean_token_accuracy": 0.17109397947788238,
      "num_tokens": 27709514.0,
      "step": 10935
    },
    {
      "entropy": 6.1956593036651615,
      "epoch": 1.2625504904789382,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048521342110430417,
      "loss": 6.0693,
      "mean_token_accuracy": 0.17368873357772827,
      "num_tokens": 27721743.0,
      "step": 10940
    },
    {
      "entropy": 6.255778646469116,
      "epoch": 1.2631275245239468,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048519870537317713,
      "loss": 6.0807,
      "mean_token_accuracy": 0.17230431735515594,
      "num_tokens": 27733802.0,
      "step": 10945
    },
    {
      "entropy": 6.2531044483184814,
      "epoch": 1.2637045585689557,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048518398257212103,
      "loss": 6.0957,
      "mean_token_accuracy": 0.17118960320949556,
      "num_tokens": 27746099.0,
      "step": 10950
    },
    {
      "entropy": 6.23198618888855,
      "epoch": 1.2642815926139641,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004851692527016311,
      "loss": 6.1196,
      "mean_token_accuracy": 0.1619269534945488,
      "num_tokens": 27758334.0,
      "step": 10955
    },
    {
      "entropy": 6.307512426376343,
      "epoch": 1.264858626658973,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004851545157622027,
      "loss": 6.0734,
      "mean_token_accuracy": 0.1659957319498062,
      "num_tokens": 27770544.0,
      "step": 10960
    },
    {
      "entropy": 6.233137416839599,
      "epoch": 1.2654356607039816,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004851397717543316,
      "loss": 6.0573,
      "mean_token_accuracy": 0.17768033742904663,
      "num_tokens": 27783328.0,
      "step": 10965
    },
    {
      "entropy": 6.157389736175537,
      "epoch": 1.2660126947489903,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004851250206785137,
      "loss": 6.0878,
      "mean_token_accuracy": 0.17061226665973664,
      "num_tokens": 27796771.0,
      "step": 10970
    },
    {
      "entropy": 6.235627031326294,
      "epoch": 1.266589728793999,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048511026253524503,
      "loss": 6.0255,
      "mean_token_accuracy": 0.17563619911670686,
      "num_tokens": 27809718.0,
      "step": 10975
    },
    {
      "entropy": 6.289537048339843,
      "epoch": 1.2671667628390075,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004850954973250221,
      "loss": 6.0933,
      "mean_token_accuracy": 0.16914141327142715,
      "num_tokens": 27822750.0,
      "step": 10980
    },
    {
      "entropy": 6.154406785964966,
      "epoch": 1.2677437968840162,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048508072504834144,
      "loss": 6.0777,
      "mean_token_accuracy": 0.16820984780788423,
      "num_tokens": 27834783.0,
      "step": 10985
    },
    {
      "entropy": 6.186893177032471,
      "epoch": 1.2683208309290248,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048506594570569997,
      "loss": 6.0829,
      "mean_token_accuracy": 0.1696375474333763,
      "num_tokens": 27848355.0,
      "step": 10990
    },
    {
      "entropy": 6.194720983505249,
      "epoch": 1.2688978649740335,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004850511592975947,
      "loss": 6.0666,
      "mean_token_accuracy": 0.17027020156383516,
      "num_tokens": 27859924.0,
      "step": 10995
    },
    {
      "entropy": 6.288698482513428,
      "epoch": 1.269474899019042,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004850363658245231,
      "loss": 6.0909,
      "mean_token_accuracy": 0.16614526957273484,
      "num_tokens": 27872186.0,
      "step": 11000
    },
    {
      "entropy": 6.270848894119263,
      "epoch": 1.2700519330640507,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004850215652869826,
      "loss": 6.0508,
      "mean_token_accuracy": 0.1698956534266472,
      "num_tokens": 27884512.0,
      "step": 11005
    },
    {
      "entropy": 6.246271562576294,
      "epoch": 1.2706289671090594,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004850067576854711,
      "loss": 6.1259,
      "mean_token_accuracy": 0.16594596952199936,
      "num_tokens": 27898665.0,
      "step": 11010
    },
    {
      "entropy": 6.2279764175415036,
      "epoch": 1.271206001154068,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004849919430204867,
      "loss": 6.0345,
      "mean_token_accuracy": 0.16853374540805816,
      "num_tokens": 27911198.0,
      "step": 11015
    },
    {
      "entropy": 6.204041242599487,
      "epoch": 1.2717830351990767,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004849771212925275,
      "loss": 6.0846,
      "mean_token_accuracy": 0.16857811361551284,
      "num_tokens": 27923677.0,
      "step": 11020
    },
    {
      "entropy": 6.204813671112061,
      "epoch": 1.2723600692440855,
      "grad_norm": 1.375,
      "learning_rate": 0.00048496229250209223,
      "loss": 6.024,
      "mean_token_accuracy": 0.17417444586753844,
      "num_tokens": 27936269.0,
      "step": 11025
    },
    {
      "entropy": 6.270748805999756,
      "epoch": 1.272937103289094,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004849474566496795,
      "loss": 6.1517,
      "mean_token_accuracy": 0.1663002535700798,
      "num_tokens": 27950463.0,
      "step": 11030
    },
    {
      "entropy": 6.226537895202637,
      "epoch": 1.2735141373341028,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004849326137357883,
      "loss": 6.0807,
      "mean_token_accuracy": 0.16937850713729857,
      "num_tokens": 27961948.0,
      "step": 11035
    },
    {
      "entropy": 6.233771944046021,
      "epoch": 1.2740911713791114,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000484917763760918,
      "loss": 6.0904,
      "mean_token_accuracy": 0.16448246985673903,
      "num_tokens": 27974075.0,
      "step": 11040
    },
    {
      "entropy": 6.280529594421386,
      "epoch": 1.27466820542412,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048490290672556784,
      "loss": 6.1424,
      "mean_token_accuracy": 0.16328521072864532,
      "num_tokens": 27987587.0,
      "step": 11045
    },
    {
      "entropy": 6.338178968429565,
      "epoch": 1.2752452394691287,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004848880426302378,
      "loss": 6.1054,
      "mean_token_accuracy": 0.1676922008395195,
      "num_tokens": 27999438.0,
      "step": 11050
    },
    {
      "entropy": 6.1836838722229,
      "epoch": 1.2758222735141374,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004848731714754277,
      "loss": 6.0117,
      "mean_token_accuracy": 0.17591148018836975,
      "num_tokens": 28012646.0,
      "step": 11055
    },
    {
      "entropy": 6.253022336959839,
      "epoch": 1.276399307559146,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004848582932616377,
      "loss": 6.1631,
      "mean_token_accuracy": 0.1612337663769722,
      "num_tokens": 28025133.0,
      "step": 11060
    },
    {
      "entropy": 6.317792749404907,
      "epoch": 1.2769763416041546,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004848434079893682,
      "loss": 6.1117,
      "mean_token_accuracy": 0.1634943038225174,
      "num_tokens": 28038062.0,
      "step": 11065
    },
    {
      "entropy": 6.301358222961426,
      "epoch": 1.2775533756491633,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004848285156591201,
      "loss": 6.1294,
      "mean_token_accuracy": 0.1632598713040352,
      "num_tokens": 28050617.0,
      "step": 11070
    },
    {
      "entropy": 6.258947372436523,
      "epoch": 1.278130409694172,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048481361627139384,
      "loss": 6.1812,
      "mean_token_accuracy": 0.1600431129336357,
      "num_tokens": 28063692.0,
      "step": 11075
    },
    {
      "entropy": 6.333370971679687,
      "epoch": 1.2787074437391805,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048479870982669096,
      "loss": 6.0661,
      "mean_token_accuracy": 0.17366596311330795,
      "num_tokens": 28075628.0,
      "step": 11080
    },
    {
      "entropy": 6.255078506469727,
      "epoch": 1.2792844777841892,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048478379632551266,
      "loss": 6.0732,
      "mean_token_accuracy": 0.1659583181142807,
      "num_tokens": 28087998.0,
      "step": 11085
    },
    {
      "entropy": 6.236647081375122,
      "epoch": 1.279861511829198,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004847688757683606,
      "loss": 6.1717,
      "mean_token_accuracy": 0.16491060853004455,
      "num_tokens": 28101260.0,
      "step": 11090
    },
    {
      "entropy": 6.305707597732544,
      "epoch": 1.2804385458742065,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004847539481557366,
      "loss": 6.1715,
      "mean_token_accuracy": 0.16244861632585525,
      "num_tokens": 28115077.0,
      "step": 11095
    },
    {
      "entropy": 6.2849588871002195,
      "epoch": 1.2810155799192153,
      "grad_norm": 1.375,
      "learning_rate": 0.00048473901348814277,
      "loss": 6.0716,
      "mean_token_accuracy": 0.16737069189548492,
      "num_tokens": 28126661.0,
      "step": 11100
    },
    {
      "entropy": 6.232383680343628,
      "epoch": 1.281592613964224,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004847240717660814,
      "loss": 6.1348,
      "mean_token_accuracy": 0.163059800863266,
      "num_tokens": 28140330.0,
      "step": 11105
    },
    {
      "entropy": 6.3044031143188475,
      "epoch": 1.2821696480092326,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048470912299005493,
      "loss": 6.0933,
      "mean_token_accuracy": 0.16603059470653533,
      "num_tokens": 28153411.0,
      "step": 11110
    },
    {
      "entropy": 6.161843729019165,
      "epoch": 1.2827466820542412,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048469416716056635,
      "loss": 6.0023,
      "mean_token_accuracy": 0.1748439610004425,
      "num_tokens": 28165812.0,
      "step": 11115
    },
    {
      "entropy": 6.191231822967529,
      "epoch": 1.28332371609925,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004846792042781187,
      "loss": 6.114,
      "mean_token_accuracy": 0.1679704174399376,
      "num_tokens": 28177702.0,
      "step": 11120
    },
    {
      "entropy": 6.2003367900848385,
      "epoch": 1.2839007501442585,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048466423434321513,
      "loss": 5.9956,
      "mean_token_accuracy": 0.16934535950422286,
      "num_tokens": 28190712.0,
      "step": 11125
    },
    {
      "entropy": 6.285148191452026,
      "epoch": 1.2844777841892672,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048464925735635907,
      "loss": 6.0775,
      "mean_token_accuracy": 0.16204079389572143,
      "num_tokens": 28204410.0,
      "step": 11130
    },
    {
      "entropy": 6.240182304382325,
      "epoch": 1.2850548182342758,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048463427331805445,
      "loss": 6.1393,
      "mean_token_accuracy": 0.15854208320379257,
      "num_tokens": 28217739.0,
      "step": 11135
    },
    {
      "entropy": 6.1983977317810055,
      "epoch": 1.2856318522792844,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004846192822288052,
      "loss": 6.0486,
      "mean_token_accuracy": 0.17073060125112532,
      "num_tokens": 28230335.0,
      "step": 11140
    },
    {
      "entropy": 6.237967395782471,
      "epoch": 1.286208886324293,
      "grad_norm": 1.0,
      "learning_rate": 0.0004846042840891155,
      "loss": 6.0933,
      "mean_token_accuracy": 0.16582798808813096,
      "num_tokens": 28243384.0,
      "step": 11145
    },
    {
      "entropy": 6.2428234100341795,
      "epoch": 1.2867859203693017,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004845892788994899,
      "loss": 6.0802,
      "mean_token_accuracy": 0.1655213177204132,
      "num_tokens": 28254963.0,
      "step": 11150
    },
    {
      "entropy": 6.253415632247925,
      "epoch": 1.2873629544143104,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004845742666604329,
      "loss": 6.0858,
      "mean_token_accuracy": 0.16508060991764067,
      "num_tokens": 28267320.0,
      "step": 11155
    },
    {
      "entropy": 6.211197996139527,
      "epoch": 1.287939988459319,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048455924737244953,
      "loss": 6.1112,
      "mean_token_accuracy": 0.16733122617006302,
      "num_tokens": 28279631.0,
      "step": 11160
    },
    {
      "entropy": 6.231957674026489,
      "epoch": 1.2885170225043279,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048454422103604505,
      "loss": 6.084,
      "mean_token_accuracy": 0.16551550179719926,
      "num_tokens": 28292356.0,
      "step": 11165
    },
    {
      "entropy": 6.204105710983276,
      "epoch": 1.2890940565493363,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004845291876517247,
      "loss": 6.143,
      "mean_token_accuracy": 0.16550988852977752,
      "num_tokens": 28306714.0,
      "step": 11170
    },
    {
      "entropy": 6.247619152069092,
      "epoch": 1.2896710905943451,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048451414721999424,
      "loss": 6.1067,
      "mean_token_accuracy": 0.16294003874063492,
      "num_tokens": 28319851.0,
      "step": 11175
    },
    {
      "entropy": 6.294659280776978,
      "epoch": 1.2902481246393538,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048449909974135954,
      "loss": 6.1091,
      "mean_token_accuracy": 0.1608608454465866,
      "num_tokens": 28333048.0,
      "step": 11180
    },
    {
      "entropy": 6.210321712493896,
      "epoch": 1.2908251586843624,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004844840452163267,
      "loss": 6.0441,
      "mean_token_accuracy": 0.1676930531859398,
      "num_tokens": 28345215.0,
      "step": 11185
    },
    {
      "entropy": 6.139756917953491,
      "epoch": 1.291402192729371,
      "grad_norm": 1.125,
      "learning_rate": 0.000484468983645402,
      "loss": 6.0031,
      "mean_token_accuracy": 0.17219747453927994,
      "num_tokens": 28356594.0,
      "step": 11190
    },
    {
      "entropy": 6.312803459167481,
      "epoch": 1.2919792267743797,
      "grad_norm": 1.375,
      "learning_rate": 0.00048445391502909213,
      "loss": 6.1194,
      "mean_token_accuracy": 0.15938821732997893,
      "num_tokens": 28368728.0,
      "step": 11195
    },
    {
      "entropy": 6.2894970893859865,
      "epoch": 1.2925562608193883,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048443883936790386,
      "loss": 6.1138,
      "mean_token_accuracy": 0.16270171999931335,
      "num_tokens": 28382290.0,
      "step": 11200
    },
    {
      "entropy": 6.21103949546814,
      "epoch": 1.293133294864397,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048442375666234427,
      "loss": 6.0893,
      "mean_token_accuracy": 0.16655988097190857,
      "num_tokens": 28394687.0,
      "step": 11205
    },
    {
      "entropy": 6.246458387374878,
      "epoch": 1.2937103289094056,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004844086669129206,
      "loss": 6.1251,
      "mean_token_accuracy": 0.16301217824220657,
      "num_tokens": 28407187.0,
      "step": 11210
    },
    {
      "entropy": 6.277497959136963,
      "epoch": 1.2942873629544143,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048439357012014045,
      "loss": 6.1405,
      "mean_token_accuracy": 0.16198444813489915,
      "num_tokens": 28419476.0,
      "step": 11215
    },
    {
      "entropy": 6.205084896087646,
      "epoch": 1.294864396999423,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004843784662845116,
      "loss": 6.0439,
      "mean_token_accuracy": 0.17078927904367447,
      "num_tokens": 28433952.0,
      "step": 11220
    },
    {
      "entropy": 6.296877050399781,
      "epoch": 1.2954414310444315,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000484363355406542,
      "loss": 6.1234,
      "mean_token_accuracy": 0.16698863357305527,
      "num_tokens": 28445687.0,
      "step": 11225
    },
    {
      "entropy": 6.196484756469727,
      "epoch": 1.2960184650894404,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048434823748674,
      "loss": 6.049,
      "mean_token_accuracy": 0.16045490354299546,
      "num_tokens": 28458883.0,
      "step": 11230
    },
    {
      "entropy": 6.162314939498901,
      "epoch": 1.2965954991344488,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048433311252561403,
      "loss": 6.0832,
      "mean_token_accuracy": 0.16035669296979904,
      "num_tokens": 28471850.0,
      "step": 11235
    },
    {
      "entropy": 6.26575345993042,
      "epoch": 1.2971725331794577,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004843179805236728,
      "loss": 6.1077,
      "mean_token_accuracy": 0.16764277815818787,
      "num_tokens": 28483640.0,
      "step": 11240
    },
    {
      "entropy": 6.341026830673218,
      "epoch": 1.297749567224466,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048430284148142516,
      "loss": 6.1251,
      "mean_token_accuracy": 0.16511788517236708,
      "num_tokens": 28497108.0,
      "step": 11245
    },
    {
      "entropy": 6.254047918319702,
      "epoch": 1.298326601269475,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004842876953993805,
      "loss": 6.0496,
      "mean_token_accuracy": 0.16369156390428544,
      "num_tokens": 28509071.0,
      "step": 11250
    },
    {
      "entropy": 6.224642133712768,
      "epoch": 1.2989036353144836,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004842725422780482,
      "loss": 6.0436,
      "mean_token_accuracy": 0.16543311327695848,
      "num_tokens": 28521822.0,
      "step": 11255
    },
    {
      "entropy": 6.291625547409057,
      "epoch": 1.2994806693594922,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004842573821179378,
      "loss": 6.2285,
      "mean_token_accuracy": 0.15752912014722825,
      "num_tokens": 28535809.0,
      "step": 11260
    },
    {
      "entropy": 6.1928857326507565,
      "epoch": 1.3000577034045009,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004842422149195593,
      "loss": 6.0288,
      "mean_token_accuracy": 0.16652102768421173,
      "num_tokens": 28549397.0,
      "step": 11265
    },
    {
      "entropy": 6.307521820068359,
      "epoch": 1.3006347374495095,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048422704068342294,
      "loss": 6.0521,
      "mean_token_accuracy": 0.17152201384305954,
      "num_tokens": 28562490.0,
      "step": 11270
    },
    {
      "entropy": 6.301024007797241,
      "epoch": 1.3012117714945182,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004842118594100389,
      "loss": 6.0872,
      "mean_token_accuracy": 0.16641155630350113,
      "num_tokens": 28575039.0,
      "step": 11275
    },
    {
      "entropy": 6.226265716552734,
      "epoch": 1.3017888055395268,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048419667109991793,
      "loss": 6.0202,
      "mean_token_accuracy": 0.16767802685499192,
      "num_tokens": 28587889.0,
      "step": 11280
    },
    {
      "entropy": 6.092101192474365,
      "epoch": 1.3023658395845354,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048418147575357085,
      "loss": 6.0393,
      "mean_token_accuracy": 0.17195614278316498,
      "num_tokens": 28600440.0,
      "step": 11285
    },
    {
      "entropy": 6.22184190750122,
      "epoch": 1.302942873629544,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004841662733715087,
      "loss": 6.0745,
      "mean_token_accuracy": 0.16288651823997496,
      "num_tokens": 28613763.0,
      "step": 11290
    },
    {
      "entropy": 6.29033670425415,
      "epoch": 1.3035199076745527,
      "grad_norm": 1.125,
      "learning_rate": 0.00048415106395424294,
      "loss": 6.0928,
      "mean_token_accuracy": 0.16747820675373076,
      "num_tokens": 28625768.0,
      "step": 11295
    },
    {
      "entropy": 6.256534767150879,
      "epoch": 1.3040969417195614,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048413584750228494,
      "loss": 6.0786,
      "mean_token_accuracy": 0.16239041537046434,
      "num_tokens": 28638207.0,
      "step": 11300
    },
    {
      "entropy": 6.281332302093506,
      "epoch": 1.3046739757645702,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048412062401614653,
      "loss": 6.161,
      "mean_token_accuracy": 0.16133101880550385,
      "num_tokens": 28650348.0,
      "step": 11305
    },
    {
      "entropy": 6.2839240550994875,
      "epoch": 1.3052510098095786,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048410539349634,
      "loss": 6.1081,
      "mean_token_accuracy": 0.1673157185316086,
      "num_tokens": 28663723.0,
      "step": 11310
    },
    {
      "entropy": 6.3154550075531,
      "epoch": 1.3058280438545875,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048409015594337725,
      "loss": 6.1775,
      "mean_token_accuracy": 0.16159482151269913,
      "num_tokens": 28676877.0,
      "step": 11315
    },
    {
      "entropy": 6.24700517654419,
      "epoch": 1.3064050778995961,
      "grad_norm": 1.09375,
      "learning_rate": 0.000484074911357771,
      "loss": 6.1189,
      "mean_token_accuracy": 0.15994147807359696,
      "num_tokens": 28689124.0,
      "step": 11320
    },
    {
      "entropy": 6.250334310531616,
      "epoch": 1.3069821119446048,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048405965974003404,
      "loss": 6.0469,
      "mean_token_accuracy": 0.17461151927709578,
      "num_tokens": 28701576.0,
      "step": 11325
    },
    {
      "entropy": 6.22020788192749,
      "epoch": 1.3075591459896134,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048404440109067927,
      "loss": 6.0905,
      "mean_token_accuracy": 0.1635117456316948,
      "num_tokens": 28715364.0,
      "step": 11330
    },
    {
      "entropy": 6.267953109741211,
      "epoch": 1.308136180034622,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004840291354102198,
      "loss": 6.0789,
      "mean_token_accuracy": 0.1635272666811943,
      "num_tokens": 28727390.0,
      "step": 11335
    },
    {
      "entropy": 6.236126375198364,
      "epoch": 1.3087132140796307,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004840138626991693,
      "loss": 6.1257,
      "mean_token_accuracy": 0.16292401254177094,
      "num_tokens": 28739310.0,
      "step": 11340
    },
    {
      "entropy": 6.248704862594605,
      "epoch": 1.3092902481246393,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004839985829580413,
      "loss": 6.0722,
      "mean_token_accuracy": 0.165024833381176,
      "num_tokens": 28750517.0,
      "step": 11345
    },
    {
      "entropy": 6.252824306488037,
      "epoch": 1.309867282169648,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048398329618734977,
      "loss": 6.0634,
      "mean_token_accuracy": 0.16562945693731307,
      "num_tokens": 28763195.0,
      "step": 11350
    },
    {
      "entropy": 6.2291654586792,
      "epoch": 1.3104443162146566,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004839680023876089,
      "loss": 6.1404,
      "mean_token_accuracy": 0.16846496909856795,
      "num_tokens": 28776697.0,
      "step": 11355
    },
    {
      "entropy": 6.165535306930542,
      "epoch": 1.3110213502596653,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004839527015593331,
      "loss": 6.0481,
      "mean_token_accuracy": 0.16787307411432267,
      "num_tokens": 28788885.0,
      "step": 11360
    },
    {
      "entropy": 6.238626956939697,
      "epoch": 1.311598384304674,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048393739370303684,
      "loss": 6.0744,
      "mean_token_accuracy": 0.17097601145505906,
      "num_tokens": 28800502.0,
      "step": 11365
    },
    {
      "entropy": 6.2409947395324705,
      "epoch": 1.3121754183496828,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004839220788192353,
      "loss": 6.0365,
      "mean_token_accuracy": 0.17041050493717194,
      "num_tokens": 28811985.0,
      "step": 11370
    },
    {
      "entropy": 6.212496376037597,
      "epoch": 1.3127524523946912,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048390675690844335,
      "loss": 6.1032,
      "mean_token_accuracy": 0.16937023103237153,
      "num_tokens": 28825492.0,
      "step": 11375
    },
    {
      "entropy": 6.251315307617188,
      "epoch": 1.3133294864397,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004838914279711764,
      "loss": 6.0856,
      "mean_token_accuracy": 0.16669962853193282,
      "num_tokens": 28838889.0,
      "step": 11380
    },
    {
      "entropy": 6.280895519256592,
      "epoch": 1.3139065204847085,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048387609200795003,
      "loss": 6.128,
      "mean_token_accuracy": 0.1591188907623291,
      "num_tokens": 28851341.0,
      "step": 11385
    },
    {
      "entropy": 6.238052129745483,
      "epoch": 1.3144835545297173,
      "grad_norm": 1.125,
      "learning_rate": 0.00048386074901928,
      "loss": 6.098,
      "mean_token_accuracy": 0.16667544841766357,
      "num_tokens": 28862861.0,
      "step": 11390
    },
    {
      "entropy": 6.2404868602752686,
      "epoch": 1.315060588574726,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004838453990056825,
      "loss": 6.1276,
      "mean_token_accuracy": 0.16566232591867447,
      "num_tokens": 28876261.0,
      "step": 11395
    },
    {
      "entropy": 6.22313814163208,
      "epoch": 1.3156376226197346,
      "grad_norm": 1.125,
      "learning_rate": 0.00048383004196767373,
      "loss": 6.0579,
      "mean_token_accuracy": 0.16965186446905137,
      "num_tokens": 28888710.0,
      "step": 11400
    },
    {
      "entropy": 6.184049415588379,
      "epoch": 1.3162146566647432,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004838146779057702,
      "loss": 6.0604,
      "mean_token_accuracy": 0.16784154772758483,
      "num_tokens": 28900606.0,
      "step": 11405
    },
    {
      "entropy": 6.2627331733703615,
      "epoch": 1.3167916907097519,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004837993068204887,
      "loss": 6.0432,
      "mean_token_accuracy": 0.17265450060367585,
      "num_tokens": 28912979.0,
      "step": 11410
    },
    {
      "entropy": 6.1196033477783205,
      "epoch": 1.3173687247547605,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048378392871234634,
      "loss": 6.0239,
      "mean_token_accuracy": 0.17114109694957733,
      "num_tokens": 28927597.0,
      "step": 11415
    },
    {
      "entropy": 6.214802360534668,
      "epoch": 1.3179457587997692,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004837685435818601,
      "loss": 6.0125,
      "mean_token_accuracy": 0.1738092914223671,
      "num_tokens": 28938914.0,
      "step": 11420
    },
    {
      "entropy": 6.272409915924072,
      "epoch": 1.3185227928447778,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004837531514295477,
      "loss": 6.1177,
      "mean_token_accuracy": 0.16232675611972808,
      "num_tokens": 28951347.0,
      "step": 11425
    },
    {
      "entropy": 6.24831371307373,
      "epoch": 1.3190998268897864,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004837377522559267,
      "loss": 6.0246,
      "mean_token_accuracy": 0.17304832488298416,
      "num_tokens": 28964904.0,
      "step": 11430
    },
    {
      "entropy": 6.163503217697143,
      "epoch": 1.319676860934795,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048372234606151507,
      "loss": 5.9008,
      "mean_token_accuracy": 0.1842797502875328,
      "num_tokens": 28977818.0,
      "step": 11435
    },
    {
      "entropy": 6.157026863098144,
      "epoch": 1.3202538949798037,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048370693284683106,
      "loss": 6.0718,
      "mean_token_accuracy": 0.16945477873086928,
      "num_tokens": 28991521.0,
      "step": 11440
    },
    {
      "entropy": 6.2621959209442135,
      "epoch": 1.3208309290248126,
      "grad_norm": 3.015625,
      "learning_rate": 0.00048369151261239303,
      "loss": 6.1272,
      "mean_token_accuracy": 0.1610644832253456,
      "num_tokens": 29003339.0,
      "step": 11445
    },
    {
      "entropy": 6.237879800796509,
      "epoch": 1.321407963069821,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004836760853587196,
      "loss": 6.1584,
      "mean_token_accuracy": 0.1674976348876953,
      "num_tokens": 29017037.0,
      "step": 11450
    },
    {
      "entropy": 6.293738317489624,
      "epoch": 1.3219849971148299,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048366065108632967,
      "loss": 6.0746,
      "mean_token_accuracy": 0.16420203298330308,
      "num_tokens": 29029496.0,
      "step": 11455
    },
    {
      "entropy": 6.201098537445068,
      "epoch": 1.3225620311598385,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048364520979574246,
      "loss": 5.9388,
      "mean_token_accuracy": 0.16896577030420304,
      "num_tokens": 29042344.0,
      "step": 11460
    },
    {
      "entropy": 6.230416107177734,
      "epoch": 1.3231390652048471,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048362976148747715,
      "loss": 6.0929,
      "mean_token_accuracy": 0.16275997906923295,
      "num_tokens": 29055518.0,
      "step": 11465
    },
    {
      "entropy": 6.207172536849976,
      "epoch": 1.3237160992498558,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004836143061620536,
      "loss": 6.0138,
      "mean_token_accuracy": 0.17673430293798448,
      "num_tokens": 29067109.0,
      "step": 11470
    },
    {
      "entropy": 6.126046514511108,
      "epoch": 1.3242931332948644,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004835988438199914,
      "loss": 5.9547,
      "mean_token_accuracy": 0.17737251222133638,
      "num_tokens": 29079766.0,
      "step": 11475
    },
    {
      "entropy": 6.263830137252808,
      "epoch": 1.324870167339873,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004835833744618107,
      "loss": 6.0984,
      "mean_token_accuracy": 0.16063766032457352,
      "num_tokens": 29092371.0,
      "step": 11480
    },
    {
      "entropy": 6.2960512161254885,
      "epoch": 1.3254472013848817,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004835678980880318,
      "loss": 6.127,
      "mean_token_accuracy": 0.16574866622686385,
      "num_tokens": 29105551.0,
      "step": 11485
    },
    {
      "entropy": 6.296307373046875,
      "epoch": 1.3260242354298903,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004835524146991753,
      "loss": 6.1313,
      "mean_token_accuracy": 0.16656022667884826,
      "num_tokens": 29118584.0,
      "step": 11490
    },
    {
      "entropy": 6.1876743793487545,
      "epoch": 1.326601269474899,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048353692429576185,
      "loss": 6.0818,
      "mean_token_accuracy": 0.17024472504854202,
      "num_tokens": 29131972.0,
      "step": 11495
    },
    {
      "entropy": 6.233011960983276,
      "epoch": 1.3271783035199076,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004835214268783126,
      "loss": 6.0478,
      "mean_token_accuracy": 0.16684432923793793,
      "num_tokens": 29143502.0,
      "step": 11500
    },
    {
      "entropy": 6.210533857345581,
      "epoch": 1.3277553375649163,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048350592244734866,
      "loss": 6.0508,
      "mean_token_accuracy": 0.1628810301423073,
      "num_tokens": 29156765.0,
      "step": 11505
    },
    {
      "entropy": 6.199675178527832,
      "epoch": 1.3283323716099251,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004834904110033917,
      "loss": 6.0634,
      "mean_token_accuracy": 0.1672512486577034,
      "num_tokens": 29168858.0,
      "step": 11510
    },
    {
      "entropy": 6.196543788909912,
      "epoch": 1.3289094056549335,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048347489254696327,
      "loss": 6.1005,
      "mean_token_accuracy": 0.17037901431322097,
      "num_tokens": 29182497.0,
      "step": 11515
    },
    {
      "entropy": 6.1987714767456055,
      "epoch": 1.3294864396999424,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004834593670785854,
      "loss": 6.0654,
      "mean_token_accuracy": 0.17931186109781266,
      "num_tokens": 29194659.0,
      "step": 11520
    },
    {
      "entropy": 6.250151300430298,
      "epoch": 1.3300634737449508,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048344383459878024,
      "loss": 6.164,
      "mean_token_accuracy": 0.16262299716472625,
      "num_tokens": 29208192.0,
      "step": 11525
    },
    {
      "entropy": 6.2489794254302975,
      "epoch": 1.3306405077899597,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048342829510807024,
      "loss": 6.058,
      "mean_token_accuracy": 0.16538500487804414,
      "num_tokens": 29220532.0,
      "step": 11530
    },
    {
      "entropy": 6.259033918380737,
      "epoch": 1.3312175418349683,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000483412748606978,
      "loss": 6.0797,
      "mean_token_accuracy": 0.16156879514455796,
      "num_tokens": 29233802.0,
      "step": 11535
    },
    {
      "entropy": 6.203406476974488,
      "epoch": 1.331794575879977,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004833971950960266,
      "loss": 5.9964,
      "mean_token_accuracy": 0.1645379677414894,
      "num_tokens": 29246099.0,
      "step": 11540
    },
    {
      "entropy": 6.22878885269165,
      "epoch": 1.3323716099249856,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004833816345757391,
      "loss": 6.0942,
      "mean_token_accuracy": 0.16495446264743804,
      "num_tokens": 29259546.0,
      "step": 11545
    },
    {
      "entropy": 6.25659441947937,
      "epoch": 1.3329486439699942,
      "grad_norm": 1.125,
      "learning_rate": 0.0004833660670466387,
      "loss": 6.0508,
      "mean_token_accuracy": 0.1738343209028244,
      "num_tokens": 29271980.0,
      "step": 11550
    },
    {
      "entropy": 6.236073017120361,
      "epoch": 1.3335256780150029,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004833504925092492,
      "loss": 5.9959,
      "mean_token_accuracy": 0.1659337803721428,
      "num_tokens": 29284679.0,
      "step": 11555
    },
    {
      "entropy": 6.222265958786011,
      "epoch": 1.3341027120600115,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004833349109640944,
      "loss": 6.0583,
      "mean_token_accuracy": 0.17521647959947587,
      "num_tokens": 29296446.0,
      "step": 11560
    },
    {
      "entropy": 6.217844152450562,
      "epoch": 1.3346797461050202,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004833193224116983,
      "loss": 6.1484,
      "mean_token_accuracy": 0.1543285295367241,
      "num_tokens": 29311036.0,
      "step": 11565
    },
    {
      "entropy": 6.236973571777344,
      "epoch": 1.3352567801500288,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048330372685258526,
      "loss": 6.0667,
      "mean_token_accuracy": 0.17181538045406342,
      "num_tokens": 29324908.0,
      "step": 11570
    },
    {
      "entropy": 6.213325691223145,
      "epoch": 1.3358338141950374,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004832881242872799,
      "loss": 6.0607,
      "mean_token_accuracy": 0.16678272187709808,
      "num_tokens": 29337755.0,
      "step": 11575
    },
    {
      "entropy": 6.181098318099975,
      "epoch": 1.336410848240046,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004832725147163069,
      "loss": 6.136,
      "mean_token_accuracy": 0.16465533822774886,
      "num_tokens": 29352148.0,
      "step": 11580
    },
    {
      "entropy": 6.301937055587769,
      "epoch": 1.336987882285055,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048325689814019134,
      "loss": 6.0627,
      "mean_token_accuracy": 0.16859597861766815,
      "num_tokens": 29365714.0,
      "step": 11585
    },
    {
      "entropy": 6.138181877136231,
      "epoch": 1.3375649163300634,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004832412745594585,
      "loss": 6.0257,
      "mean_token_accuracy": 0.1757267251610756,
      "num_tokens": 29377699.0,
      "step": 11590
    },
    {
      "entropy": 6.200165510177612,
      "epoch": 1.3381419503750722,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048322564397463376,
      "loss": 6.0384,
      "mean_token_accuracy": 0.16549893021583556,
      "num_tokens": 29390293.0,
      "step": 11595
    },
    {
      "entropy": 6.260862159729004,
      "epoch": 1.3387189844200809,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048321000638624296,
      "loss": 6.0228,
      "mean_token_accuracy": 0.17569999992847443,
      "num_tokens": 29403148.0,
      "step": 11600
    },
    {
      "entropy": 6.2256103515625,
      "epoch": 1.3392960184650895,
      "grad_norm": 1.171875,
      "learning_rate": 0.000483194361794812,
      "loss": 6.0458,
      "mean_token_accuracy": 0.17279241532087325,
      "num_tokens": 29416457.0,
      "step": 11605
    },
    {
      "entropy": 6.274567413330078,
      "epoch": 1.3398730525100981,
      "grad_norm": 1.234375,
      "learning_rate": 0.000483178710200867,
      "loss": 6.1945,
      "mean_token_accuracy": 0.1609697937965393,
      "num_tokens": 29429657.0,
      "step": 11610
    },
    {
      "entropy": 6.268039560317993,
      "epoch": 1.3404500865551068,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004831630516049346,
      "loss": 6.0217,
      "mean_token_accuracy": 0.17185084968805314,
      "num_tokens": 29442517.0,
      "step": 11615
    },
    {
      "entropy": 6.234312343597412,
      "epoch": 1.3410271206001154,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004831473860075414,
      "loss": 5.9946,
      "mean_token_accuracy": 0.17028348743915558,
      "num_tokens": 29455360.0,
      "step": 11620
    },
    {
      "entropy": 6.245578813552856,
      "epoch": 1.341604154645124,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048313171340921417,
      "loss": 6.109,
      "mean_token_accuracy": 0.16786335855722428,
      "num_tokens": 29469321.0,
      "step": 11625
    },
    {
      "entropy": 6.231622219085693,
      "epoch": 1.3421811886901327,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048311603381048025,
      "loss": 6.0481,
      "mean_token_accuracy": 0.17694897949695587,
      "num_tokens": 29481688.0,
      "step": 11630
    },
    {
      "entropy": 6.198208618164062,
      "epoch": 1.3427582227351413,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004831003472118668,
      "loss": 6.0717,
      "mean_token_accuracy": 0.17185677886009215,
      "num_tokens": 29493468.0,
      "step": 11635
    },
    {
      "entropy": 6.231797409057617,
      "epoch": 1.34333525678015,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004830846536139016,
      "loss": 6.1738,
      "mean_token_accuracy": 0.1629571169614792,
      "num_tokens": 29506254.0,
      "step": 11640
    },
    {
      "entropy": 6.302359628677368,
      "epoch": 1.3439122908251586,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004830689530171124,
      "loss": 6.1299,
      "mean_token_accuracy": 0.1607202872633934,
      "num_tokens": 29518838.0,
      "step": 11645
    },
    {
      "entropy": 6.234106826782226,
      "epoch": 1.3444893248701673,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004830532454220273,
      "loss": 6.0418,
      "mean_token_accuracy": 0.1678894951939583,
      "num_tokens": 29531252.0,
      "step": 11650
    },
    {
      "entropy": 6.303366994857788,
      "epoch": 1.345066358915176,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048303753082917474,
      "loss": 6.1505,
      "mean_token_accuracy": 0.16442754715681077,
      "num_tokens": 29545125.0,
      "step": 11655
    },
    {
      "entropy": 6.331900930404663,
      "epoch": 1.3456433929601848,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048302180923908306,
      "loss": 6.1275,
      "mean_token_accuracy": 0.1650748670101166,
      "num_tokens": 29558011.0,
      "step": 11660
    },
    {
      "entropy": 6.240604019165039,
      "epoch": 1.3462204270051932,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048300608065228126,
      "loss": 6.0589,
      "mean_token_accuracy": 0.16700272411108016,
      "num_tokens": 29570864.0,
      "step": 11665
    },
    {
      "entropy": 6.244296503067017,
      "epoch": 1.346797461050202,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048299034506929815,
      "loss": 6.0635,
      "mean_token_accuracy": 0.16397027373313905,
      "num_tokens": 29583098.0,
      "step": 11670
    },
    {
      "entropy": 6.149010133743286,
      "epoch": 1.3473744950952107,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048297460249066315,
      "loss": 5.9558,
      "mean_token_accuracy": 0.17619435340166092,
      "num_tokens": 29594980.0,
      "step": 11675
    },
    {
      "entropy": 6.228585624694825,
      "epoch": 1.3479515291402193,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004829588529169057,
      "loss": 6.0886,
      "mean_token_accuracy": 0.16212412863969802,
      "num_tokens": 29608424.0,
      "step": 11680
    },
    {
      "entropy": 6.230656337738037,
      "epoch": 1.348528563185228,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004829430963485555,
      "loss": 5.9936,
      "mean_token_accuracy": 0.1664886713027954,
      "num_tokens": 29621784.0,
      "step": 11685
    },
    {
      "entropy": 6.088874626159668,
      "epoch": 1.3491055972302366,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004829273327861426,
      "loss": 5.9525,
      "mean_token_accuracy": 0.17296147644519805,
      "num_tokens": 29633875.0,
      "step": 11690
    },
    {
      "entropy": 6.174909019470215,
      "epoch": 1.3496826312752452,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004829115622301971,
      "loss": 5.959,
      "mean_token_accuracy": 0.17539376020431519,
      "num_tokens": 29647510.0,
      "step": 11695
    },
    {
      "entropy": 6.295996952056885,
      "epoch": 1.3502596653202539,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00048289578468124956,
      "loss": 6.1111,
      "mean_token_accuracy": 0.1600963741540909,
      "num_tokens": 29659916.0,
      "step": 11700
    },
    {
      "entropy": 6.310726690292358,
      "epoch": 1.3508366993652625,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048288000013983046,
      "loss": 6.1249,
      "mean_token_accuracy": 0.16550098210573197,
      "num_tokens": 29673628.0,
      "step": 11705
    },
    {
      "entropy": 6.249656915664673,
      "epoch": 1.3514137334102712,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048286420860647086,
      "loss": 6.0912,
      "mean_token_accuracy": 0.17122483253479004,
      "num_tokens": 29685245.0,
      "step": 11710
    },
    {
      "entropy": 6.217113590240478,
      "epoch": 1.3519907674552798,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048284841008170185,
      "loss": 6.0166,
      "mean_token_accuracy": 0.17496787309646605,
      "num_tokens": 29698205.0,
      "step": 11715
    },
    {
      "entropy": 6.2207762718200685,
      "epoch": 1.3525678015002884,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048283260456605487,
      "loss": 6.1189,
      "mean_token_accuracy": 0.16303366422653198,
      "num_tokens": 29712060.0,
      "step": 11720
    },
    {
      "entropy": 6.279819965362549,
      "epoch": 1.3531448355452973,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004828167920600614,
      "loss": 6.072,
      "mean_token_accuracy": 0.16497474908828735,
      "num_tokens": 29724452.0,
      "step": 11725
    },
    {
      "entropy": 6.222843837738037,
      "epoch": 1.3537218695903057,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004828009725642534,
      "loss": 6.0019,
      "mean_token_accuracy": 0.16934843808412553,
      "num_tokens": 29735754.0,
      "step": 11730
    },
    {
      "entropy": 6.100900793075562,
      "epoch": 1.3542989036353146,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004827851460791628,
      "loss": 5.9861,
      "mean_token_accuracy": 0.1702702075242996,
      "num_tokens": 29747918.0,
      "step": 11735
    },
    {
      "entropy": 6.2715764999389645,
      "epoch": 1.3548759376803232,
      "grad_norm": 1.953125,
      "learning_rate": 0.00048276931260532213,
      "loss": 6.0628,
      "mean_token_accuracy": 0.17006913870573043,
      "num_tokens": 29760152.0,
      "step": 11740
    },
    {
      "entropy": 6.212006139755249,
      "epoch": 1.3554529717253319,
      "grad_norm": 1.125,
      "learning_rate": 0.0004827534721432639,
      "loss": 6.0448,
      "mean_token_accuracy": 0.16846529245376587,
      "num_tokens": 29771677.0,
      "step": 11745
    },
    {
      "entropy": 6.239493751525879,
      "epoch": 1.3560300057703405,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004827376246935207,
      "loss": 6.0628,
      "mean_token_accuracy": 0.16834279596805574,
      "num_tokens": 29784523.0,
      "step": 11750
    },
    {
      "entropy": 6.196507978439331,
      "epoch": 1.3566070398153491,
      "grad_norm": 1.25,
      "learning_rate": 0.0004827217702566257,
      "loss": 6.0052,
      "mean_token_accuracy": 0.17502202838659286,
      "num_tokens": 29798181.0,
      "step": 11755
    },
    {
      "entropy": 6.237059640884399,
      "epoch": 1.3571840738603578,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048270590883311217,
      "loss": 6.1809,
      "mean_token_accuracy": 0.16236244589090348,
      "num_tokens": 29810530.0,
      "step": 11760
    },
    {
      "entropy": 6.229019498825073,
      "epoch": 1.3577611079053664,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048269004042351363,
      "loss": 6.0883,
      "mean_token_accuracy": 0.17302400916814803,
      "num_tokens": 29823929.0,
      "step": 11765
    },
    {
      "entropy": 6.265957403182983,
      "epoch": 1.358338141950375,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004826741650283637,
      "loss": 6.0986,
      "mean_token_accuracy": 0.16292303800582886,
      "num_tokens": 29836405.0,
      "step": 11770
    },
    {
      "entropy": 6.271416139602661,
      "epoch": 1.3589151759953837,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004826582826481963,
      "loss": 6.0213,
      "mean_token_accuracy": 0.16648781299591064,
      "num_tokens": 29849392.0,
      "step": 11775
    },
    {
      "entropy": 6.245918893814087,
      "epoch": 1.3594922100403923,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004826423932835458,
      "loss": 6.0457,
      "mean_token_accuracy": 0.17648170590400697,
      "num_tokens": 29862431.0,
      "step": 11780
    },
    {
      "entropy": 6.188624525070191,
      "epoch": 1.360069244085401,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048262649693494653,
      "loss": 6.0578,
      "mean_token_accuracy": 0.18135675191879272,
      "num_tokens": 29875628.0,
      "step": 11785
    },
    {
      "entropy": 6.25065426826477,
      "epoch": 1.3606462781304096,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004826105936029332,
      "loss": 6.1259,
      "mean_token_accuracy": 0.16146169155836104,
      "num_tokens": 29888399.0,
      "step": 11790
    },
    {
      "entropy": 6.310590505599976,
      "epoch": 1.3612233121754183,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004825946832880406,
      "loss": 6.0815,
      "mean_token_accuracy": 0.16686558574438096,
      "num_tokens": 29900776.0,
      "step": 11795
    },
    {
      "entropy": 6.276925039291382,
      "epoch": 1.3618003462204271,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048257876599080404,
      "loss": 6.1009,
      "mean_token_accuracy": 0.16280327439308168,
      "num_tokens": 29913764.0,
      "step": 11800
    },
    {
      "entropy": 6.199784803390503,
      "epoch": 1.3623773802654355,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048256284171175874,
      "loss": 6.0408,
      "mean_token_accuracy": 0.16483252495527267,
      "num_tokens": 29925884.0,
      "step": 11805
    },
    {
      "entropy": 6.1182914733886715,
      "epoch": 1.3629544143104444,
      "grad_norm": 1.25,
      "learning_rate": 0.00048254691045144035,
      "loss": 5.9527,
      "mean_token_accuracy": 0.18289717584848403,
      "num_tokens": 29938151.0,
      "step": 11810
    },
    {
      "entropy": 6.157000780105591,
      "epoch": 1.363531448355453,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004825309722103848,
      "loss": 6.0278,
      "mean_token_accuracy": 0.16867677569389344,
      "num_tokens": 29950696.0,
      "step": 11815
    },
    {
      "entropy": 6.279379177093506,
      "epoch": 1.3641084824004617,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000482515026989128,
      "loss": 6.1157,
      "mean_token_accuracy": 0.16332835257053374,
      "num_tokens": 29963295.0,
      "step": 11820
    },
    {
      "entropy": 6.18568000793457,
      "epoch": 1.3646855164454703,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048249907478820634,
      "loss": 6.0304,
      "mean_token_accuracy": 0.172616046667099,
      "num_tokens": 29976709.0,
      "step": 11825
    },
    {
      "entropy": 6.170938777923584,
      "epoch": 1.365262550490479,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048248311560815637,
      "loss": 5.9894,
      "mean_token_accuracy": 0.1750449240207672,
      "num_tokens": 29989341.0,
      "step": 11830
    },
    {
      "entropy": 6.307317781448364,
      "epoch": 1.3658395845354876,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004824671494495149,
      "loss": 6.0696,
      "mean_token_accuracy": 0.1649801701307297,
      "num_tokens": 30001630.0,
      "step": 11835
    },
    {
      "entropy": 6.311050701141357,
      "epoch": 1.3664166185804962,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004824511763128189,
      "loss": 6.1232,
      "mean_token_accuracy": 0.1598815068602562,
      "num_tokens": 30014629.0,
      "step": 11840
    },
    {
      "entropy": 6.117914056777954,
      "epoch": 1.3669936526255049,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048243519619860567,
      "loss": 6.0118,
      "mean_token_accuracy": 0.17595477849245073,
      "num_tokens": 30025410.0,
      "step": 11845
    },
    {
      "entropy": 6.241130208969116,
      "epoch": 1.3675706866705135,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004824192091074126,
      "loss": 6.0642,
      "mean_token_accuracy": 0.1696405291557312,
      "num_tokens": 30036955.0,
      "step": 11850
    },
    {
      "entropy": 6.252166318893432,
      "epoch": 1.3681477207155222,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004824032150397775,
      "loss": 6.0776,
      "mean_token_accuracy": 0.16315291970968246,
      "num_tokens": 30049426.0,
      "step": 11855
    },
    {
      "entropy": 6.205201435089111,
      "epoch": 1.3687247547605308,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048238721399623824,
      "loss": 6.0972,
      "mean_token_accuracy": 0.17019951492547988,
      "num_tokens": 30061139.0,
      "step": 11860
    },
    {
      "entropy": 6.267063331604004,
      "epoch": 1.3693017888055397,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048237120597733316,
      "loss": 6.1209,
      "mean_token_accuracy": 0.16265926510095596,
      "num_tokens": 30074307.0,
      "step": 11865
    },
    {
      "entropy": 6.23607382774353,
      "epoch": 1.369878822850548,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004823551909836005,
      "loss": 6.0027,
      "mean_token_accuracy": 0.16685540676116944,
      "num_tokens": 30086610.0,
      "step": 11870
    },
    {
      "entropy": 6.186155652999878,
      "epoch": 1.370455856895557,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048233916901557896,
      "loss": 6.1283,
      "mean_token_accuracy": 0.16397479325532913,
      "num_tokens": 30098971.0,
      "step": 11875
    },
    {
      "entropy": 6.300810241699219,
      "epoch": 1.3710328909405656,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048232314007380753,
      "loss": 6.0662,
      "mean_token_accuracy": 0.16778755336999893,
      "num_tokens": 30110521.0,
      "step": 11880
    },
    {
      "entropy": 6.16118049621582,
      "epoch": 1.3716099249855742,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048230710415882524,
      "loss": 6.0009,
      "mean_token_accuracy": 0.17631883025169373,
      "num_tokens": 30123687.0,
      "step": 11885
    },
    {
      "entropy": 6.14832763671875,
      "epoch": 1.3721869590305829,
      "grad_norm": 1.25,
      "learning_rate": 0.00048229106127117144,
      "loss": 6.0117,
      "mean_token_accuracy": 0.17309411913156508,
      "num_tokens": 30137396.0,
      "step": 11890
    },
    {
      "entropy": 6.279986047744751,
      "epoch": 1.3727639930755915,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004822750114113858,
      "loss": 6.0965,
      "mean_token_accuracy": 0.1673493891954422,
      "num_tokens": 30149791.0,
      "step": 11895
    },
    {
      "entropy": 6.142151927947998,
      "epoch": 1.3733410271206001,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004822589545800081,
      "loss": 5.9294,
      "mean_token_accuracy": 0.18450151830911637,
      "num_tokens": 30162829.0,
      "step": 11900
    },
    {
      "entropy": 6.196474933624268,
      "epoch": 1.3739180611656088,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004822428907775784,
      "loss": 5.9995,
      "mean_token_accuracy": 0.171402445435524,
      "num_tokens": 30174476.0,
      "step": 11905
    },
    {
      "entropy": 6.160983276367188,
      "epoch": 1.3744950952106174,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048222682000463704,
      "loss": 6.0031,
      "mean_token_accuracy": 0.17317581176757812,
      "num_tokens": 30186573.0,
      "step": 11910
    },
    {
      "entropy": 6.211895561218261,
      "epoch": 1.375072129255626,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004822107422617245,
      "loss": 6.0373,
      "mean_token_accuracy": 0.1723754361271858,
      "num_tokens": 30198482.0,
      "step": 11915
    },
    {
      "entropy": 6.237060070037842,
      "epoch": 1.3756491633006347,
      "grad_norm": 2.03125,
      "learning_rate": 0.00048219465754938156,
      "loss": 6.0426,
      "mean_token_accuracy": 0.1679088890552521,
      "num_tokens": 30211206.0,
      "step": 11920
    },
    {
      "entropy": 6.215777826309204,
      "epoch": 1.3762261973456433,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048217856586814926,
      "loss": 6.0865,
      "mean_token_accuracy": 0.15811911523342131,
      "num_tokens": 30224250.0,
      "step": 11925
    },
    {
      "entropy": 6.222831296920776,
      "epoch": 1.376803231390652,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048216246721856875,
      "loss": 6.0667,
      "mean_token_accuracy": 0.17145148515701295,
      "num_tokens": 30236212.0,
      "step": 11930
    },
    {
      "entropy": 6.297378635406494,
      "epoch": 1.3773802654356606,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048214636160118164,
      "loss": 6.1921,
      "mean_token_accuracy": 0.15789121240377427,
      "num_tokens": 30250681.0,
      "step": 11935
    },
    {
      "entropy": 6.297422170639038,
      "epoch": 1.3779572994806695,
      "grad_norm": 1.0,
      "learning_rate": 0.0004821302490165295,
      "loss": 6.1846,
      "mean_token_accuracy": 0.1560791626572609,
      "num_tokens": 30264659.0,
      "step": 11940
    },
    {
      "entropy": 6.237554216384888,
      "epoch": 1.378534333525678,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004821141294651544,
      "loss": 6.0301,
      "mean_token_accuracy": 0.16998827010393142,
      "num_tokens": 30278821.0,
      "step": 11945
    },
    {
      "entropy": 6.212662839889527,
      "epoch": 1.3791113675706868,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048209800294759836,
      "loss": 6.0885,
      "mean_token_accuracy": 0.16028614789247514,
      "num_tokens": 30290502.0,
      "step": 11950
    },
    {
      "entropy": 6.207283306121826,
      "epoch": 1.3796884016156954,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004820818694644039,
      "loss": 6.0535,
      "mean_token_accuracy": 0.15980068743228912,
      "num_tokens": 30304499.0,
      "step": 11955
    },
    {
      "entropy": 6.2244867324829105,
      "epoch": 1.380265435660704,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048206572901611364,
      "loss": 6.0425,
      "mean_token_accuracy": 0.17032324969768525,
      "num_tokens": 30315435.0,
      "step": 11960
    },
    {
      "entropy": 6.219475507736206,
      "epoch": 1.3808424697057127,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048204958160327034,
      "loss": 6.0696,
      "mean_token_accuracy": 0.1650005578994751,
      "num_tokens": 30327502.0,
      "step": 11965
    },
    {
      "entropy": 6.172317361831665,
      "epoch": 1.3814195037507213,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048203342722641726,
      "loss": 6.0583,
      "mean_token_accuracy": 0.17196011394262314,
      "num_tokens": 30341193.0,
      "step": 11970
    },
    {
      "entropy": 6.210407543182373,
      "epoch": 1.38199653779573,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048201726588609776,
      "loss": 6.0536,
      "mean_token_accuracy": 0.16616689264774323,
      "num_tokens": 30352899.0,
      "step": 11975
    },
    {
      "entropy": 6.249279975891113,
      "epoch": 1.3825735718407386,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048200109758285534,
      "loss": 6.0864,
      "mean_token_accuracy": 0.1677591934800148,
      "num_tokens": 30366672.0,
      "step": 11980
    },
    {
      "entropy": 6.25421986579895,
      "epoch": 1.3831506058857472,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004819849223172337,
      "loss": 6.0722,
      "mean_token_accuracy": 0.16757689118385316,
      "num_tokens": 30379124.0,
      "step": 11985
    },
    {
      "entropy": 6.148027038574218,
      "epoch": 1.3837276399307559,
      "grad_norm": 1.25,
      "learning_rate": 0.00048196874008977716,
      "loss": 5.9755,
      "mean_token_accuracy": 0.175503271818161,
      "num_tokens": 30391974.0,
      "step": 11990
    },
    {
      "entropy": 6.166814088821411,
      "epoch": 1.3843046739757645,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004819525509010298,
      "loss": 6.0096,
      "mean_token_accuracy": 0.16837238818407058,
      "num_tokens": 30404711.0,
      "step": 11995
    },
    {
      "entropy": 6.1980644226074215,
      "epoch": 1.3848817080207732,
      "grad_norm": 1.25,
      "learning_rate": 0.0004819363547515361,
      "loss": 6.0224,
      "mean_token_accuracy": 0.17253003865480424,
      "num_tokens": 30418500.0,
      "step": 12000
    },
    {
      "epoch": 1.3848817080207732,
      "eval_entropy": 6.130081328201403,
      "eval_loss": 6.174646854400635,
      "eval_mean_token_accuracy": 0.16762067316555998,
      "eval_num_tokens": 30418500.0,
      "eval_runtime": 23.2153,
      "eval_samples_per_second": 1211.957,
      "eval_steps_per_second": 151.495,
      "step": 12000
    },
    {
      "entropy": 6.26961760520935,
      "epoch": 1.385458742065782,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000481920151641841,
      "loss": 6.12,
      "mean_token_accuracy": 0.16312455385923386,
      "num_tokens": 30430851.0,
      "step": 12005
    },
    {
      "entropy": 6.1342285633087155,
      "epoch": 1.3860357761107904,
      "grad_norm": 1.375,
      "learning_rate": 0.00048190394157248935,
      "loss": 5.947,
      "mean_token_accuracy": 0.18462190628051758,
      "num_tokens": 30442885.0,
      "step": 12010
    },
    {
      "entropy": 6.18363356590271,
      "epoch": 1.3866128101557993,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004818877245440264,
      "loss": 6.1051,
      "mean_token_accuracy": 0.16310730427503586,
      "num_tokens": 30456145.0,
      "step": 12015
    },
    {
      "entropy": 6.276082277297974,
      "epoch": 1.3871898442008077,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048187150055699763,
      "loss": 6.1029,
      "mean_token_accuracy": 0.16550741642713546,
      "num_tokens": 30468222.0,
      "step": 12020
    },
    {
      "entropy": 6.23896369934082,
      "epoch": 1.3877668782458166,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004818552696119487,
      "loss": 6.0495,
      "mean_token_accuracy": 0.16325829178094864,
      "num_tokens": 30481713.0,
      "step": 12025
    },
    {
      "entropy": 6.050390720367432,
      "epoch": 1.3883439122908252,
      "grad_norm": 1.375,
      "learning_rate": 0.0004818390317094255,
      "loss": 5.8708,
      "mean_token_accuracy": 0.17589485198259353,
      "num_tokens": 30495622.0,
      "step": 12030
    },
    {
      "entropy": 6.199893283843994,
      "epoch": 1.3889209463358339,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004818227868499742,
      "loss": 6.0787,
      "mean_token_accuracy": 0.1669609546661377,
      "num_tokens": 30507959.0,
      "step": 12035
    },
    {
      "entropy": 6.20874433517456,
      "epoch": 1.3894979803808425,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004818065350341412,
      "loss": 6.1128,
      "mean_token_accuracy": 0.16586913615465165,
      "num_tokens": 30520814.0,
      "step": 12040
    },
    {
      "entropy": 6.274769258499146,
      "epoch": 1.3900750144258511,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048179027626247325,
      "loss": 6.0931,
      "mean_token_accuracy": 0.16860166937112808,
      "num_tokens": 30533761.0,
      "step": 12045
    },
    {
      "entropy": 6.211928796768189,
      "epoch": 1.3906520484708598,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004817740105355169,
      "loss": 6.0524,
      "mean_token_accuracy": 0.16639767736196517,
      "num_tokens": 30546131.0,
      "step": 12050
    },
    {
      "entropy": 6.199551630020141,
      "epoch": 1.3912290825158684,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048175773785381947,
      "loss": 6.0139,
      "mean_token_accuracy": 0.17093193680047988,
      "num_tokens": 30559867.0,
      "step": 12055
    },
    {
      "entropy": 6.231350994110107,
      "epoch": 1.391806116560877,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048174145821792833,
      "loss": 5.9966,
      "mean_token_accuracy": 0.1762056052684784,
      "num_tokens": 30572089.0,
      "step": 12060
    },
    {
      "entropy": 6.169072675704956,
      "epoch": 1.3923831506058857,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004817251716283908,
      "loss": 6.0549,
      "mean_token_accuracy": 0.17105615735054017,
      "num_tokens": 30585579.0,
      "step": 12065
    },
    {
      "entropy": 6.165319204330444,
      "epoch": 1.3929601846508943,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004817088780857548,
      "loss": 6.088,
      "mean_token_accuracy": 0.17060659527778627,
      "num_tokens": 30597560.0,
      "step": 12070
    },
    {
      "entropy": 6.202260303497314,
      "epoch": 1.393537218695903,
      "grad_norm": 1.125,
      "learning_rate": 0.00048169257759056845,
      "loss": 5.9845,
      "mean_token_accuracy": 0.17014180421829223,
      "num_tokens": 30609323.0,
      "step": 12075
    },
    {
      "entropy": 6.174258518218994,
      "epoch": 1.3941142527409118,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048167627014337994,
      "loss": 6.0788,
      "mean_token_accuracy": 0.16650353521108627,
      "num_tokens": 30621437.0,
      "step": 12080
    },
    {
      "entropy": 6.199364089965821,
      "epoch": 1.3946912867859202,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048165995574473764,
      "loss": 6.0435,
      "mean_token_accuracy": 0.17507200837135314,
      "num_tokens": 30634656.0,
      "step": 12085
    },
    {
      "entropy": 6.244093608856201,
      "epoch": 1.395268320830929,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048164363439519043,
      "loss": 6.0211,
      "mean_token_accuracy": 0.1768357127904892,
      "num_tokens": 30645456.0,
      "step": 12090
    },
    {
      "entropy": 6.105206060409546,
      "epoch": 1.3958453548759377,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004816273060952873,
      "loss": 5.995,
      "mean_token_accuracy": 0.17482708692550658,
      "num_tokens": 30658093.0,
      "step": 12095
    },
    {
      "entropy": 6.210802364349365,
      "epoch": 1.3964223889209464,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048161097084557726,
      "loss": 6.0489,
      "mean_token_accuracy": 0.1678304672241211,
      "num_tokens": 30670301.0,
      "step": 12100
    },
    {
      "entropy": 6.241651916503907,
      "epoch": 1.396999422965955,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048159462864660993,
      "loss": 6.0158,
      "mean_token_accuracy": 0.16986437886953354,
      "num_tokens": 30682512.0,
      "step": 12105
    },
    {
      "entropy": 6.161880397796631,
      "epoch": 1.3975764570109637,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004815782794989348,
      "loss": 5.9373,
      "mean_token_accuracy": 0.1770050570368767,
      "num_tokens": 30693677.0,
      "step": 12110
    },
    {
      "entropy": 6.1591959476470945,
      "epoch": 1.3981534910559723,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004815619234031019,
      "loss": 5.9824,
      "mean_token_accuracy": 0.16919902116060256,
      "num_tokens": 30706994.0,
      "step": 12115
    },
    {
      "entropy": 6.170262384414673,
      "epoch": 1.398730525100981,
      "grad_norm": 1.125,
      "learning_rate": 0.0004815455603596613,
      "loss": 5.9568,
      "mean_token_accuracy": 0.17787652164697648,
      "num_tokens": 30721164.0,
      "step": 12120
    },
    {
      "entropy": 6.128379249572754,
      "epoch": 1.3993075591459896,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004815291903691634,
      "loss": 5.9265,
      "mean_token_accuracy": 0.17501710653305053,
      "num_tokens": 30734337.0,
      "step": 12125
    },
    {
      "entropy": 6.205810260772705,
      "epoch": 1.3998845931909982,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048151281343215873,
      "loss": 6.0982,
      "mean_token_accuracy": 0.16452783793210984,
      "num_tokens": 30747565.0,
      "step": 12130
    },
    {
      "entropy": 6.227529764175415,
      "epoch": 1.4004616272360069,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004814964295491982,
      "loss": 6.0146,
      "mean_token_accuracy": 0.17472368627786636,
      "num_tokens": 30759285.0,
      "step": 12135
    },
    {
      "entropy": 6.257336521148682,
      "epoch": 1.4010386612810155,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048148003872083286,
      "loss": 6.0926,
      "mean_token_accuracy": 0.17082367241382598,
      "num_tokens": 30772390.0,
      "step": 12140
    },
    {
      "entropy": 6.199289703369141,
      "epoch": 1.4016156953260244,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048146364094761384,
      "loss": 6.0112,
      "mean_token_accuracy": 0.17535965293645858,
      "num_tokens": 30784845.0,
      "step": 12145
    },
    {
      "entropy": 6.26460452079773,
      "epoch": 1.4021927293710328,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048144723623009297,
      "loss": 6.0406,
      "mean_token_accuracy": 0.16838831156492234,
      "num_tokens": 30796705.0,
      "step": 12150
    },
    {
      "entropy": 6.168288040161133,
      "epoch": 1.4027697634160416,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004814308245688218,
      "loss": 6.0528,
      "mean_token_accuracy": 0.17289142906665803,
      "num_tokens": 30809143.0,
      "step": 12155
    },
    {
      "entropy": 6.200293922424317,
      "epoch": 1.40334679746105,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00048141440596435235,
      "loss": 6.078,
      "mean_token_accuracy": 0.16614389568567275,
      "num_tokens": 30821519.0,
      "step": 12160
    },
    {
      "entropy": 6.135768270492553,
      "epoch": 1.403923831506059,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004813979804172369,
      "loss": 5.9038,
      "mean_token_accuracy": 0.17931211292743682,
      "num_tokens": 30835310.0,
      "step": 12165
    },
    {
      "entropy": 6.160478115081787,
      "epoch": 1.4045008655510676,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048138154792802795,
      "loss": 6.0008,
      "mean_token_accuracy": 0.17310880422592162,
      "num_tokens": 30847966.0,
      "step": 12170
    },
    {
      "entropy": 6.298379230499267,
      "epoch": 1.4050778995960762,
      "grad_norm": 1.375,
      "learning_rate": 0.0004813651084972781,
      "loss": 6.074,
      "mean_token_accuracy": 0.16076691895723344,
      "num_tokens": 30860732.0,
      "step": 12175
    },
    {
      "entropy": 6.262832975387573,
      "epoch": 1.4056549336410848,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048134866212554036,
      "loss": 6.1201,
      "mean_token_accuracy": 0.16459095180034639,
      "num_tokens": 30874846.0,
      "step": 12180
    },
    {
      "entropy": 6.241823434829712,
      "epoch": 1.4062319676860935,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004813322088133679,
      "loss": 5.9553,
      "mean_token_accuracy": 0.17984518706798552,
      "num_tokens": 30888470.0,
      "step": 12185
    },
    {
      "entropy": 6.2655829906463625,
      "epoch": 1.4068090017311021,
      "grad_norm": 1.609375,
      "learning_rate": 0.00048131574856131407,
      "loss": 6.1085,
      "mean_token_accuracy": 0.1606815055012703,
      "num_tokens": 30901124.0,
      "step": 12190
    },
    {
      "entropy": 6.192427158355713,
      "epoch": 1.4073860357761108,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004812992813699324,
      "loss": 6.0871,
      "mean_token_accuracy": 0.16555003225803375,
      "num_tokens": 30912533.0,
      "step": 12195
    },
    {
      "entropy": 6.175000190734863,
      "epoch": 1.4079630698211194,
      "grad_norm": 1.34375,
      "learning_rate": 0.000481282807239777,
      "loss": 6.0335,
      "mean_token_accuracy": 0.1687975063920021,
      "num_tokens": 30925070.0,
      "step": 12200
    },
    {
      "entropy": 6.276103782653808,
      "epoch": 1.408540103866128,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004812663261714019,
      "loss": 6.0745,
      "mean_token_accuracy": 0.16618521511554718,
      "num_tokens": 30938305.0,
      "step": 12205
    },
    {
      "entropy": 6.260602331161499,
      "epoch": 1.4091171379111367,
      "grad_norm": 1.25,
      "learning_rate": 0.0004812498381653613,
      "loss": 6.1118,
      "mean_token_accuracy": 0.16490792632102966,
      "num_tokens": 30951020.0,
      "step": 12210
    },
    {
      "entropy": 6.211295890808105,
      "epoch": 1.4096941719561453,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004812333432222098,
      "loss": 6.0058,
      "mean_token_accuracy": 0.17003893554210664,
      "num_tokens": 30963712.0,
      "step": 12215
    },
    {
      "entropy": 6.231636953353882,
      "epoch": 1.4102712060011542,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004812168413425023,
      "loss": 5.9924,
      "mean_token_accuracy": 0.1743488922715187,
      "num_tokens": 30975859.0,
      "step": 12220
    },
    {
      "entropy": 6.2659214496612545,
      "epoch": 1.4108482400461626,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048120033252679374,
      "loss": 6.099,
      "mean_token_accuracy": 0.16174319833517076,
      "num_tokens": 30989100.0,
      "step": 12225
    },
    {
      "entropy": 6.160903644561768,
      "epoch": 1.4114252740911715,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048118381677563946,
      "loss": 6.0793,
      "mean_token_accuracy": 0.1699076697230339,
      "num_tokens": 31001921.0,
      "step": 12230
    },
    {
      "entropy": 6.259254026412964,
      "epoch": 1.41200230813618,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004811672940895949,
      "loss": 6.1084,
      "mean_token_accuracy": 0.16021304577589035,
      "num_tokens": 31015642.0,
      "step": 12235
    },
    {
      "entropy": 6.229254627227784,
      "epoch": 1.4125793421811887,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004811507644692158,
      "loss": 6.0349,
      "mean_token_accuracy": 0.16906412243843078,
      "num_tokens": 31028964.0,
      "step": 12240
    },
    {
      "entropy": 6.246864748001099,
      "epoch": 1.4131563762261974,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004811342279150581,
      "loss": 6.1212,
      "mean_token_accuracy": 0.16789718270301818,
      "num_tokens": 31040567.0,
      "step": 12245
    },
    {
      "entropy": 6.279138469696045,
      "epoch": 1.413733410271206,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004811176844276781,
      "loss": 6.12,
      "mean_token_accuracy": 0.1620114877820015,
      "num_tokens": 31053005.0,
      "step": 12250
    },
    {
      "entropy": 6.23945722579956,
      "epoch": 1.4143104443162147,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004811011340076322,
      "loss": 6.0565,
      "mean_token_accuracy": 0.16643643975257874,
      "num_tokens": 31065971.0,
      "step": 12255
    },
    {
      "entropy": 6.203128957748413,
      "epoch": 1.4148874783612233,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048108457665547695,
      "loss": 5.9832,
      "mean_token_accuracy": 0.17818353623151778,
      "num_tokens": 31078762.0,
      "step": 12260
    },
    {
      "entropy": 6.288204860687256,
      "epoch": 1.415464512406232,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004810680123717694,
      "loss": 5.9904,
      "mean_token_accuracy": 0.17016339004039766,
      "num_tokens": 31091462.0,
      "step": 12265
    },
    {
      "entropy": 6.2541498184204105,
      "epoch": 1.4160415464512406,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004810514411570666,
      "loss": 6.0483,
      "mean_token_accuracy": 0.16790302097797394,
      "num_tokens": 31104610.0,
      "step": 12270
    },
    {
      "entropy": 6.134290552139282,
      "epoch": 1.4166185804962492,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004810348630119259,
      "loss": 6.0691,
      "mean_token_accuracy": 0.1710997626185417,
      "num_tokens": 31117125.0,
      "step": 12275
    },
    {
      "entropy": 6.208537483215332,
      "epoch": 1.4171956145412579,
      "grad_norm": 1.5,
      "learning_rate": 0.00048101827793690493,
      "loss": 6.0413,
      "mean_token_accuracy": 0.1682871013879776,
      "num_tokens": 31129703.0,
      "step": 12280
    },
    {
      "entropy": 6.237831783294678,
      "epoch": 1.4177726485862667,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004810016859325615,
      "loss": 6.0435,
      "mean_token_accuracy": 0.16955197006464004,
      "num_tokens": 31142127.0,
      "step": 12285
    },
    {
      "entropy": 6.191223239898681,
      "epoch": 1.4183496826312751,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004809850869994537,
      "loss": 6.0401,
      "mean_token_accuracy": 0.17489763647317885,
      "num_tokens": 31154665.0,
      "step": 12290
    },
    {
      "entropy": 6.212182521820068,
      "epoch": 1.418926716676284,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004809684811381398,
      "loss": 6.0716,
      "mean_token_accuracy": 0.17062939554452897,
      "num_tokens": 31167004.0,
      "step": 12295
    },
    {
      "entropy": 6.24400577545166,
      "epoch": 1.4195037507212924,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004809518683491785,
      "loss": 6.0522,
      "mean_token_accuracy": 0.16716936826705933,
      "num_tokens": 31178979.0,
      "step": 12300
    },
    {
      "entropy": 6.217967081069946,
      "epoch": 1.4200807847663013,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048093524863312823,
      "loss": 6.0832,
      "mean_token_accuracy": 0.16065378189086915,
      "num_tokens": 31190877.0,
      "step": 12305
    },
    {
      "entropy": 6.281795692443848,
      "epoch": 1.42065781881131,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004809186219905482,
      "loss": 6.0732,
      "mean_token_accuracy": 0.16857749819755555,
      "num_tokens": 31202988.0,
      "step": 12310
    },
    {
      "entropy": 6.273308229446411,
      "epoch": 1.4212348528563186,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004809019884219976,
      "loss": 6.0715,
      "mean_token_accuracy": 0.16564688831567764,
      "num_tokens": 31215436.0,
      "step": 12315
    },
    {
      "entropy": 6.277575540542602,
      "epoch": 1.4218118869013272,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048088534792803595,
      "loss": 6.0571,
      "mean_token_accuracy": 0.15880540907382965,
      "num_tokens": 31228260.0,
      "step": 12320
    },
    {
      "entropy": 6.2636778354644775,
      "epoch": 1.4223889209463358,
      "grad_norm": 1.609375,
      "learning_rate": 0.00048086870050922286,
      "loss": 6.1187,
      "mean_token_accuracy": 0.15972764492034913,
      "num_tokens": 31242496.0,
      "step": 12325
    },
    {
      "entropy": 6.252569580078125,
      "epoch": 1.4229659549913445,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048085204616611833,
      "loss": 6.0646,
      "mean_token_accuracy": 0.16664598882198334,
      "num_tokens": 31254375.0,
      "step": 12330
    },
    {
      "entropy": 6.251256799697876,
      "epoch": 1.4235429890363531,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00048083538489928246,
      "loss": 6.1353,
      "mean_token_accuracy": 0.1585749350488186,
      "num_tokens": 31266590.0,
      "step": 12335
    },
    {
      "entropy": 6.33405613899231,
      "epoch": 1.4241200230813618,
      "grad_norm": 1.25,
      "learning_rate": 0.0004808187167092757,
      "loss": 6.1531,
      "mean_token_accuracy": 0.16426093876361847,
      "num_tokens": 31278928.0,
      "step": 12340
    },
    {
      "entropy": 6.156169414520264,
      "epoch": 1.4246970571263704,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004808020415966586,
      "loss": 5.9282,
      "mean_token_accuracy": 0.17937763780355453,
      "num_tokens": 31292144.0,
      "step": 12345
    },
    {
      "entropy": 6.199422168731689,
      "epoch": 1.425274091171379,
      "grad_norm": 1.0625,
      "learning_rate": 0.000480785359561992,
      "loss": 6.0114,
      "mean_token_accuracy": 0.17004047334194183,
      "num_tokens": 31303690.0,
      "step": 12350
    },
    {
      "entropy": 6.222023630142212,
      "epoch": 1.4258511252163877,
      "grad_norm": 1.125,
      "learning_rate": 0.00048076867060583704,
      "loss": 6.0305,
      "mean_token_accuracy": 0.17170640230178832,
      "num_tokens": 31315433.0,
      "step": 12355
    },
    {
      "entropy": 6.2131434917449955,
      "epoch": 1.4264281592613965,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004807519747287551,
      "loss": 6.008,
      "mean_token_accuracy": 0.17607229501008986,
      "num_tokens": 31327483.0,
      "step": 12360
    },
    {
      "entropy": 6.236727237701416,
      "epoch": 1.427005193306405,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004807352719313078,
      "loss": 6.0976,
      "mean_token_accuracy": 0.1667151555418968,
      "num_tokens": 31339167.0,
      "step": 12365
    },
    {
      "entropy": 6.169782257080078,
      "epoch": 1.4275822273514138,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004807185622140567,
      "loss": 5.9775,
      "mean_token_accuracy": 0.17382654547691345,
      "num_tokens": 31352476.0,
      "step": 12370
    },
    {
      "entropy": 6.165315961837768,
      "epoch": 1.4281592613964225,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048070184557756396,
      "loss": 5.9939,
      "mean_token_accuracy": 0.1793465092778206,
      "num_tokens": 31364286.0,
      "step": 12375
    },
    {
      "entropy": 6.235998678207397,
      "epoch": 1.428736295441431,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048068512202239177,
      "loss": 6.0262,
      "mean_token_accuracy": 0.16516128331422805,
      "num_tokens": 31378141.0,
      "step": 12380
    },
    {
      "entropy": 6.201778364181519,
      "epoch": 1.4293133294864397,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004806683915491028,
      "loss": 6.014,
      "mean_token_accuracy": 0.18075743317604065,
      "num_tokens": 31390738.0,
      "step": 12385
    },
    {
      "entropy": 6.121563863754273,
      "epoch": 1.4298903635314484,
      "grad_norm": 1.25,
      "learning_rate": 0.0004806516541582596,
      "loss": 6.0209,
      "mean_token_accuracy": 0.16548589169979094,
      "num_tokens": 31402289.0,
      "step": 12390
    },
    {
      "entropy": 6.2844654560089115,
      "epoch": 1.430467397576457,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048063490985042506,
      "loss": 6.094,
      "mean_token_accuracy": 0.17097145617008208,
      "num_tokens": 31416029.0,
      "step": 12395
    },
    {
      "entropy": 6.172171497344971,
      "epoch": 1.4310444316214657,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004806181586261626,
      "loss": 6.0406,
      "mean_token_accuracy": 0.17326876968145372,
      "num_tokens": 31428980.0,
      "step": 12400
    },
    {
      "entropy": 6.1689659595489506,
      "epoch": 1.4316214656664743,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048060140048603544,
      "loss": 5.995,
      "mean_token_accuracy": 0.17331080436706542,
      "num_tokens": 31441424.0,
      "step": 12405
    },
    {
      "entropy": 6.278450632095337,
      "epoch": 1.432198499711483,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004805846354306073,
      "loss": 6.0271,
      "mean_token_accuracy": 0.16783675849437713,
      "num_tokens": 31453968.0,
      "step": 12410
    },
    {
      "entropy": 6.236226367950439,
      "epoch": 1.4327755337564916,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048056786346044214,
      "loss": 6.1026,
      "mean_token_accuracy": 0.16841660737991332,
      "num_tokens": 31466385.0,
      "step": 12415
    },
    {
      "entropy": 6.230169773101807,
      "epoch": 1.4333525678015002,
      "grad_norm": 1.703125,
      "learning_rate": 0.00048055108457610395,
      "loss": 6.0753,
      "mean_token_accuracy": 0.16451437175273895,
      "num_tokens": 31478432.0,
      "step": 12420
    },
    {
      "entropy": 6.265069961547852,
      "epoch": 1.4339296018465089,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004805342987781571,
      "loss": 6.089,
      "mean_token_accuracy": 0.16451953053474427,
      "num_tokens": 31490435.0,
      "step": 12425
    },
    {
      "entropy": 6.258297395706177,
      "epoch": 1.4345066358915175,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004805175060671663,
      "loss": 6.0209,
      "mean_token_accuracy": 0.17181486636400223,
      "num_tokens": 31502573.0,
      "step": 12430
    },
    {
      "entropy": 6.232821226119995,
      "epoch": 1.4350836699365264,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004805007064436962,
      "loss": 6.0992,
      "mean_token_accuracy": 0.1664665937423706,
      "num_tokens": 31516071.0,
      "step": 12435
    },
    {
      "entropy": 6.2183116436004635,
      "epoch": 1.4356607039815348,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004804838999083119,
      "loss": 6.0272,
      "mean_token_accuracy": 0.16762907207012176,
      "num_tokens": 31528895.0,
      "step": 12440
    },
    {
      "entropy": 6.321974229812622,
      "epoch": 1.4362377380265436,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004804670864615787,
      "loss": 6.1382,
      "mean_token_accuracy": 0.16287727206945418,
      "num_tokens": 31541759.0,
      "step": 12445
    },
    {
      "entropy": 6.2230544090271,
      "epoch": 1.4368147720715523,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048045026610406223,
      "loss": 6.0735,
      "mean_token_accuracy": 0.1635092705488205,
      "num_tokens": 31554249.0,
      "step": 12450
    },
    {
      "entropy": 6.299511241912842,
      "epoch": 1.437391806116561,
      "grad_norm": 1.3125,
      "learning_rate": 0.000480433438836328,
      "loss": 6.1403,
      "mean_token_accuracy": 0.1611890897154808,
      "num_tokens": 31567673.0,
      "step": 12455
    },
    {
      "entropy": 6.196793127059936,
      "epoch": 1.4379688401615696,
      "grad_norm": 1.375,
      "learning_rate": 0.00048041660465894206,
      "loss": 5.9811,
      "mean_token_accuracy": 0.1772790238261223,
      "num_tokens": 31581209.0,
      "step": 12460
    },
    {
      "entropy": 6.173445844650269,
      "epoch": 1.4385458742065782,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004803997635724707,
      "loss": 6.0657,
      "mean_token_accuracy": 0.16836874485015868,
      "num_tokens": 31593873.0,
      "step": 12465
    },
    {
      "entropy": 6.195061254501343,
      "epoch": 1.4391229082515868,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004803829155774804,
      "loss": 6.004,
      "mean_token_accuracy": 0.16877640932798385,
      "num_tokens": 31605942.0,
      "step": 12470
    },
    {
      "entropy": 6.269618129730224,
      "epoch": 1.4396999422965955,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004803660606745377,
      "loss": 6.0531,
      "mean_token_accuracy": 0.16527083814144133,
      "num_tokens": 31617648.0,
      "step": 12475
    },
    {
      "entropy": 6.189599990844727,
      "epoch": 1.4402769763416041,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048034919886420953,
      "loss": 6.0318,
      "mean_token_accuracy": 0.16477131694555283,
      "num_tokens": 31630774.0,
      "step": 12480
    },
    {
      "entropy": 6.232012701034546,
      "epoch": 1.4408540103866128,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048033233014706304,
      "loss": 6.0758,
      "mean_token_accuracy": 0.16045521944761276,
      "num_tokens": 31642888.0,
      "step": 12485
    },
    {
      "entropy": 6.18221001625061,
      "epoch": 1.4414310444316214,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048031545452366565,
      "loss": 5.9954,
      "mean_token_accuracy": 0.16896130442619323,
      "num_tokens": 31655180.0,
      "step": 12490
    },
    {
      "entropy": 6.156050157546997,
      "epoch": 1.44200807847663,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048029857199458493,
      "loss": 6.0548,
      "mean_token_accuracy": 0.16745578944683076,
      "num_tokens": 31667452.0,
      "step": 12495
    },
    {
      "entropy": 6.263525581359863,
      "epoch": 1.442585112521639,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048028168256038873,
      "loss": 6.0686,
      "mean_token_accuracy": 0.16707100868225097,
      "num_tokens": 31680995.0,
      "step": 12500
    },
    {
      "entropy": 6.251485967636109,
      "epoch": 1.4431621465666473,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048026478622164513,
      "loss": 6.1006,
      "mean_token_accuracy": 0.1677337571978569,
      "num_tokens": 31693604.0,
      "step": 12505
    },
    {
      "entropy": 6.235770225524902,
      "epoch": 1.4437391806116562,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048024788297892234,
      "loss": 6.0776,
      "mean_token_accuracy": 0.16443003416061402,
      "num_tokens": 31706370.0,
      "step": 12510
    },
    {
      "entropy": 6.20050082206726,
      "epoch": 1.4443162146566648,
      "grad_norm": 1.328125,
      "learning_rate": 0.000480230972832789,
      "loss": 6.0133,
      "mean_token_accuracy": 0.16382628977298735,
      "num_tokens": 31719265.0,
      "step": 12515
    },
    {
      "entropy": 6.140182876586914,
      "epoch": 1.4448932487016735,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048021405578381384,
      "loss": 5.9437,
      "mean_token_accuracy": 0.1786150723695755,
      "num_tokens": 31732587.0,
      "step": 12520
    },
    {
      "entropy": 6.187860774993896,
      "epoch": 1.445470282746682,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004801971318325659,
      "loss": 6.024,
      "mean_token_accuracy": 0.16767466068267822,
      "num_tokens": 31744664.0,
      "step": 12525
    },
    {
      "entropy": 6.220370292663574,
      "epoch": 1.4460473167916907,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004801802009796142,
      "loss": 6.034,
      "mean_token_accuracy": 0.1767251029610634,
      "num_tokens": 31757814.0,
      "step": 12530
    },
    {
      "entropy": 6.151736640930176,
      "epoch": 1.4466243508366994,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004801632632255285,
      "loss": 5.992,
      "mean_token_accuracy": 0.16925470232963563,
      "num_tokens": 31770575.0,
      "step": 12535
    },
    {
      "entropy": 6.159832000732422,
      "epoch": 1.447201384881708,
      "grad_norm": 1.125,
      "learning_rate": 0.0004801463185708783,
      "loss": 6.0673,
      "mean_token_accuracy": 0.1688586503267288,
      "num_tokens": 31782746.0,
      "step": 12540
    },
    {
      "entropy": 6.217080736160279,
      "epoch": 1.4477784189267167,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048012936701623363,
      "loss": 5.9628,
      "mean_token_accuracy": 0.17586362808942796,
      "num_tokens": 31794459.0,
      "step": 12545
    },
    {
      "entropy": 6.166052055358887,
      "epoch": 1.4483554529717253,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048011240856216456,
      "loss": 5.955,
      "mean_token_accuracy": 0.17674486786127092,
      "num_tokens": 31806205.0,
      "step": 12550
    },
    {
      "entropy": 6.180267667770385,
      "epoch": 1.448932487016734,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048009544320924154,
      "loss": 6.0713,
      "mean_token_accuracy": 0.16638890355825425,
      "num_tokens": 31818590.0,
      "step": 12555
    },
    {
      "entropy": 6.249054431915283,
      "epoch": 1.4495095210617426,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004800784709580351,
      "loss": 5.998,
      "mean_token_accuracy": 0.16679723411798478,
      "num_tokens": 31830702.0,
      "step": 12560
    },
    {
      "entropy": 6.180447435379028,
      "epoch": 1.4500865551067512,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004800614918091161,
      "loss": 5.9788,
      "mean_token_accuracy": 0.16819451749324799,
      "num_tokens": 31844326.0,
      "step": 12565
    },
    {
      "entropy": 6.153980112075805,
      "epoch": 1.4506635891517599,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004800445057630558,
      "loss": 6.0447,
      "mean_token_accuracy": 0.16177617609500886,
      "num_tokens": 31857153.0,
      "step": 12570
    },
    {
      "entropy": 6.207056760787964,
      "epoch": 1.4512406231967687,
      "grad_norm": 1.375,
      "learning_rate": 0.0004800275128204254,
      "loss": 6.0195,
      "mean_token_accuracy": 0.17025156915187836,
      "num_tokens": 31870589.0,
      "step": 12575
    },
    {
      "entropy": 6.201283025741577,
      "epoch": 1.4518176572417771,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048001051298179637,
      "loss": 5.9766,
      "mean_token_accuracy": 0.17282721251249314,
      "num_tokens": 31882867.0,
      "step": 12580
    },
    {
      "entropy": 6.170648384094238,
      "epoch": 1.452394691286786,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004799935062477407,
      "loss": 6.0353,
      "mean_token_accuracy": 0.16624790281057358,
      "num_tokens": 31895582.0,
      "step": 12585
    },
    {
      "entropy": 6.224724197387696,
      "epoch": 1.4529717253317946,
      "grad_norm": 1.125,
      "learning_rate": 0.00047997649261883013,
      "loss": 6.0144,
      "mean_token_accuracy": 0.1730797305703163,
      "num_tokens": 31907873.0,
      "step": 12590
    },
    {
      "entropy": 6.20830078125,
      "epoch": 1.4535487593768033,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004799594720956371,
      "loss": 5.9238,
      "mean_token_accuracy": 0.17136916816234588,
      "num_tokens": 31920459.0,
      "step": 12595
    },
    {
      "entropy": 6.223221397399902,
      "epoch": 1.454125793421812,
      "grad_norm": 1.546875,
      "learning_rate": 0.00047994244467873407,
      "loss": 6.0713,
      "mean_token_accuracy": 0.16101902574300767,
      "num_tokens": 31934707.0,
      "step": 12600
    },
    {
      "entropy": 6.267828035354614,
      "epoch": 1.4547028274668206,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047992541036869364,
      "loss": 6.1386,
      "mean_token_accuracy": 0.16569984704256058,
      "num_tokens": 31946891.0,
      "step": 12605
    },
    {
      "entropy": 6.2044243812561035,
      "epoch": 1.4552798615118292,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004799083691660889,
      "loss": 6.0132,
      "mean_token_accuracy": 0.16808016300201417,
      "num_tokens": 31960300.0,
      "step": 12610
    },
    {
      "entropy": 6.223467636108398,
      "epoch": 1.4558568955568378,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004798913210714929,
      "loss": 6.0587,
      "mean_token_accuracy": 0.1708383560180664,
      "num_tokens": 31973565.0,
      "step": 12615
    },
    {
      "entropy": 6.187826442718506,
      "epoch": 1.4564339296018465,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00047987426608547915,
      "loss": 6.0622,
      "mean_token_accuracy": 0.16502688229084014,
      "num_tokens": 31986256.0,
      "step": 12620
    },
    {
      "entropy": 6.190675592422485,
      "epoch": 1.4570109636468551,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004798572042086212,
      "loss": 6.0026,
      "mean_token_accuracy": 0.1740684226155281,
      "num_tokens": 31999034.0,
      "step": 12625
    },
    {
      "entropy": 6.256427335739136,
      "epoch": 1.4575879976918638,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047984013544149286,
      "loss": 6.0713,
      "mean_token_accuracy": 0.16957840770483018,
      "num_tokens": 32010213.0,
      "step": 12630
    },
    {
      "entropy": 6.310705614089966,
      "epoch": 1.4581650317368724,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047982305978466836,
      "loss": 6.1049,
      "mean_token_accuracy": 0.1676050156354904,
      "num_tokens": 32022516.0,
      "step": 12635
    },
    {
      "entropy": 6.234405612945556,
      "epoch": 1.4587420657818813,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00047980597723872205,
      "loss": 6.038,
      "mean_token_accuracy": 0.16381986886262895,
      "num_tokens": 32035644.0,
      "step": 12640
    },
    {
      "entropy": 6.248470306396484,
      "epoch": 1.4593190998268897,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004797888878042283,
      "loss": 6.1387,
      "mean_token_accuracy": 0.16827376037836075,
      "num_tokens": 32049547.0,
      "step": 12645
    },
    {
      "entropy": 6.168174886703492,
      "epoch": 1.4598961338718985,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047977179148176217,
      "loss": 6.0066,
      "mean_token_accuracy": 0.17118684947490692,
      "num_tokens": 32062071.0,
      "step": 12650
    },
    {
      "entropy": 6.257451057434082,
      "epoch": 1.4604731679169072,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004797546882718985,
      "loss": 6.0822,
      "mean_token_accuracy": 0.15844431668519973,
      "num_tokens": 32073964.0,
      "step": 12655
    },
    {
      "entropy": 6.285771656036377,
      "epoch": 1.4610502019619158,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047973757817521256,
      "loss": 6.066,
      "mean_token_accuracy": 0.16742073595523835,
      "num_tokens": 32087151.0,
      "step": 12660
    },
    {
      "entropy": 6.252929735183716,
      "epoch": 1.4616272360069245,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004797204611922799,
      "loss": 6.0519,
      "mean_token_accuracy": 0.16763749867677688,
      "num_tokens": 32099072.0,
      "step": 12665
    },
    {
      "entropy": 6.148719930648804,
      "epoch": 1.462204270051933,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00047970333732367626,
      "loss": 5.9992,
      "mean_token_accuracy": 0.17236820012331008,
      "num_tokens": 32112182.0,
      "step": 12670
    },
    {
      "entropy": 6.0970946788787845,
      "epoch": 1.4627813040969417,
      "grad_norm": 1.125,
      "learning_rate": 0.00047968620656997754,
      "loss": 5.8585,
      "mean_token_accuracy": 0.18248848766088485,
      "num_tokens": 32124067.0,
      "step": 12675
    },
    {
      "entropy": 6.20596981048584,
      "epoch": 1.4633583381419504,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047966906893175994,
      "loss": 6.0389,
      "mean_token_accuracy": 0.17081668376922607,
      "num_tokens": 32136983.0,
      "step": 12680
    },
    {
      "entropy": 6.227654266357422,
      "epoch": 1.463935372186959,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004796519244095998,
      "loss": 6.0323,
      "mean_token_accuracy": 0.17243611961603164,
      "num_tokens": 32150103.0,
      "step": 12685
    },
    {
      "entropy": 6.2273177146911625,
      "epoch": 1.4645124062319677,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047963477300407394,
      "loss": 6.0742,
      "mean_token_accuracy": 0.15987755805253984,
      "num_tokens": 32162647.0,
      "step": 12690
    },
    {
      "entropy": 6.163004064559937,
      "epoch": 1.4650894402769763,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047961761471575903,
      "loss": 5.9495,
      "mean_token_accuracy": 0.1758941113948822,
      "num_tokens": 32174707.0,
      "step": 12695
    },
    {
      "entropy": 6.203647708892822,
      "epoch": 1.465666474321985,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047960044954523237,
      "loss": 6.0127,
      "mean_token_accuracy": 0.177766090631485,
      "num_tokens": 32187784.0,
      "step": 12700
    },
    {
      "entropy": 6.289816617965698,
      "epoch": 1.4662435083669936,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047958327749307117,
      "loss": 6.124,
      "mean_token_accuracy": 0.1687891349196434,
      "num_tokens": 32200300.0,
      "step": 12705
    },
    {
      "entropy": 6.229695558547974,
      "epoch": 1.4668205424120022,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047956609855985305,
      "loss": 6.0079,
      "mean_token_accuracy": 0.16917745172977447,
      "num_tokens": 32212836.0,
      "step": 12710
    },
    {
      "entropy": 6.059126234054565,
      "epoch": 1.467397576457011,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004795489127461559,
      "loss": 5.8665,
      "mean_token_accuracy": 0.18514809012413025,
      "num_tokens": 32226230.0,
      "step": 12715
    },
    {
      "entropy": 6.251426696777344,
      "epoch": 1.4679746105020195,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047953172005255756,
      "loss": 6.1202,
      "mean_token_accuracy": 0.16973395943641661,
      "num_tokens": 32238994.0,
      "step": 12720
    },
    {
      "entropy": 6.228357458114624,
      "epoch": 1.4685516445470284,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004795145204796365,
      "loss": 6.0625,
      "mean_token_accuracy": 0.16718969196081163,
      "num_tokens": 32250983.0,
      "step": 12725
    },
    {
      "entropy": 6.205570363998413,
      "epoch": 1.469128678592037,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004794973140279711,
      "loss": 6.0655,
      "mean_token_accuracy": 0.16379925906658171,
      "num_tokens": 32263839.0,
      "step": 12730
    },
    {
      "entropy": 6.263099098205567,
      "epoch": 1.4697057126370456,
      "grad_norm": 1.53125,
      "learning_rate": 0.00047948010069814005,
      "loss": 6.0353,
      "mean_token_accuracy": 0.1677084296941757,
      "num_tokens": 32275539.0,
      "step": 12735
    },
    {
      "entropy": 6.248122215270996,
      "epoch": 1.4702827466820543,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004794628804907225,
      "loss": 6.0268,
      "mean_token_accuracy": 0.17067530304193496,
      "num_tokens": 32287421.0,
      "step": 12740
    },
    {
      "entropy": 6.203702020645141,
      "epoch": 1.470859780727063,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047944565340629755,
      "loss": 6.0885,
      "mean_token_accuracy": 0.16034181118011476,
      "num_tokens": 32300246.0,
      "step": 12745
    },
    {
      "entropy": 6.257831287384033,
      "epoch": 1.4714368147720716,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00047942841944544453,
      "loss": 6.0364,
      "mean_token_accuracy": 0.17367035746574402,
      "num_tokens": 32311957.0,
      "step": 12750
    },
    {
      "entropy": 6.141442966461182,
      "epoch": 1.4720138488170802,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004794111786087431,
      "loss": 6.0169,
      "mean_token_accuracy": 0.17246225476264954,
      "num_tokens": 32324710.0,
      "step": 12755
    },
    {
      "entropy": 6.196126508712768,
      "epoch": 1.4725908828620888,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004793939308967733,
      "loss": 6.0684,
      "mean_token_accuracy": 0.17600734680891036,
      "num_tokens": 32337252.0,
      "step": 12760
    },
    {
      "entropy": 6.325767946243286,
      "epoch": 1.4731679169070975,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004793766763101152,
      "loss": 6.0435,
      "mean_token_accuracy": 0.16541136652231217,
      "num_tokens": 32348957.0,
      "step": 12765
    },
    {
      "entropy": 6.218016242980957,
      "epoch": 1.4737449509521061,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047935941484934905,
      "loss": 6.0755,
      "mean_token_accuracy": 0.16323485523462294,
      "num_tokens": 32362395.0,
      "step": 12770
    },
    {
      "entropy": 6.174684286117554,
      "epoch": 1.4743219849971148,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047934214651505547,
      "loss": 5.9884,
      "mean_token_accuracy": 0.17242209017276763,
      "num_tokens": 32375962.0,
      "step": 12775
    },
    {
      "entropy": 6.242296886444092,
      "epoch": 1.4748990190421236,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047932487130781533,
      "loss": 6.0858,
      "mean_token_accuracy": 0.16375706046819688,
      "num_tokens": 32387457.0,
      "step": 12780
    },
    {
      "entropy": 6.239092254638672,
      "epoch": 1.475476053087132,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004793075892282097,
      "loss": 5.9666,
      "mean_token_accuracy": 0.17163785845041274,
      "num_tokens": 32400263.0,
      "step": 12785
    },
    {
      "entropy": 6.178677940368653,
      "epoch": 1.476053087132141,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004792903002768197,
      "loss": 6.0529,
      "mean_token_accuracy": 0.16690157502889633,
      "num_tokens": 32412552.0,
      "step": 12790
    },
    {
      "entropy": 6.2029482364654545,
      "epoch": 1.4766301211771495,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047927300445422687,
      "loss": 6.0983,
      "mean_token_accuracy": 0.1674714744091034,
      "num_tokens": 32424922.0,
      "step": 12795
    },
    {
      "entropy": 6.303258085250855,
      "epoch": 1.4772071552221582,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004792557017610131,
      "loss": 6.0814,
      "mean_token_accuracy": 0.16482014954090118,
      "num_tokens": 32438164.0,
      "step": 12800
    },
    {
      "entropy": 6.238474273681641,
      "epoch": 1.4777841892671668,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047923839219776027,
      "loss": 5.9766,
      "mean_token_accuracy": 0.17610716074705124,
      "num_tokens": 32450718.0,
      "step": 12805
    },
    {
      "entropy": 6.154720878601074,
      "epoch": 1.4783612233121755,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004792210757650506,
      "loss": 5.9718,
      "mean_token_accuracy": 0.17466458082199096,
      "num_tokens": 32464562.0,
      "step": 12810
    },
    {
      "entropy": 6.222097587585449,
      "epoch": 1.478938257357184,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047920375246346636,
      "loss": 6.0915,
      "mean_token_accuracy": 0.16450839936733247,
      "num_tokens": 32476707.0,
      "step": 12815
    },
    {
      "entropy": 6.243588161468506,
      "epoch": 1.4795152914021927,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047918642229359044,
      "loss": 6.0835,
      "mean_token_accuracy": 0.1676979646086693,
      "num_tokens": 32489979.0,
      "step": 12820
    },
    {
      "entropy": 6.252044725418091,
      "epoch": 1.4800923254472014,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004791690852560056,
      "loss": 6.1104,
      "mean_token_accuracy": 0.16395504176616668,
      "num_tokens": 32502479.0,
      "step": 12825
    },
    {
      "entropy": 6.313782691955566,
      "epoch": 1.48066935949221,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000479151741351295,
      "loss": 6.0487,
      "mean_token_accuracy": 0.16525107026100158,
      "num_tokens": 32515075.0,
      "step": 12830
    },
    {
      "entropy": 6.184175205230713,
      "epoch": 1.4812463935372187,
      "grad_norm": 1.09375,
      "learning_rate": 0.000479134390580042,
      "loss": 6.0794,
      "mean_token_accuracy": 0.16342978179454803,
      "num_tokens": 32527647.0,
      "step": 12835
    },
    {
      "entropy": 6.251873016357422,
      "epoch": 1.4818234275822273,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047911703294283015,
      "loss": 6.0617,
      "mean_token_accuracy": 0.16858035922050477,
      "num_tokens": 32539040.0,
      "step": 12840
    },
    {
      "entropy": 6.258333015441894,
      "epoch": 1.482400461627236,
      "grad_norm": 3.328125,
      "learning_rate": 0.00047909966844024334,
      "loss": 6.0546,
      "mean_token_accuracy": 0.16797869950532912,
      "num_tokens": 32551657.0,
      "step": 12845
    },
    {
      "entropy": 6.27937421798706,
      "epoch": 1.4829774956722446,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047908229707286547,
      "loss": 6.1693,
      "mean_token_accuracy": 0.1627509891986847,
      "num_tokens": 32564796.0,
      "step": 12850
    },
    {
      "entropy": 6.287845134735107,
      "epoch": 1.4835545297172534,
      "grad_norm": 1.125,
      "learning_rate": 0.000479064918841281,
      "loss": 6.0788,
      "mean_token_accuracy": 0.16349496692419052,
      "num_tokens": 32576808.0,
      "step": 12855
    },
    {
      "entropy": 6.2645570755004885,
      "epoch": 1.4841315637622619,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047904753374607427,
      "loss": 6.091,
      "mean_token_accuracy": 0.1660340204834938,
      "num_tokens": 32590855.0,
      "step": 12860
    },
    {
      "entropy": 6.282743215560913,
      "epoch": 1.4847085978072707,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047903014178783015,
      "loss": 6.1,
      "mean_token_accuracy": 0.16721449494361879,
      "num_tokens": 32603505.0,
      "step": 12865
    },
    {
      "entropy": 6.172833395004273,
      "epoch": 1.4852856318522794,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004790127429671335,
      "loss": 6.0275,
      "mean_token_accuracy": 0.17583226710557937,
      "num_tokens": 32616366.0,
      "step": 12870
    },
    {
      "entropy": 6.214118194580078,
      "epoch": 1.485862665897288,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004789953372845697,
      "loss": 6.1381,
      "mean_token_accuracy": 0.16323215067386626,
      "num_tokens": 32628087.0,
      "step": 12875
    },
    {
      "entropy": 6.275833559036255,
      "epoch": 1.4864396999422966,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000478977924740724,
      "loss": 6.1036,
      "mean_token_accuracy": 0.16198642998933793,
      "num_tokens": 32642048.0,
      "step": 12880
    },
    {
      "entropy": 6.260259914398193,
      "epoch": 1.4870167339873053,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004789605053361821,
      "loss": 6.0053,
      "mean_token_accuracy": 0.1714352548122406,
      "num_tokens": 32653353.0,
      "step": 12885
    },
    {
      "entropy": 6.19688835144043,
      "epoch": 1.487593768032314,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004789430790715299,
      "loss": 6.1233,
      "mean_token_accuracy": 0.1691578984260559,
      "num_tokens": 32666733.0,
      "step": 12890
    },
    {
      "entropy": 6.2053159236907955,
      "epoch": 1.4881708020773226,
      "grad_norm": 4.40625,
      "learning_rate": 0.00047892564594735355,
      "loss": 5.9912,
      "mean_token_accuracy": 0.1703313559293747,
      "num_tokens": 32678666.0,
      "step": 12895
    },
    {
      "entropy": 6.268673515319824,
      "epoch": 1.4887478361223312,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047890820596423943,
      "loss": 6.1156,
      "mean_token_accuracy": 0.16318060904741288,
      "num_tokens": 32691622.0,
      "step": 12900
    },
    {
      "entropy": 6.287637186050415,
      "epoch": 1.4893248701673398,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000478890759122774,
      "loss": 6.1736,
      "mean_token_accuracy": 0.16167111992835997,
      "num_tokens": 32704469.0,
      "step": 12905
    },
    {
      "entropy": 6.2565327167510985,
      "epoch": 1.4899019042123485,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004788733054235443,
      "loss": 5.9846,
      "mean_token_accuracy": 0.17693448364734649,
      "num_tokens": 32717119.0,
      "step": 12910
    },
    {
      "entropy": 6.094601202011108,
      "epoch": 1.4904789382573571,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047885584486713717,
      "loss": 5.8885,
      "mean_token_accuracy": 0.18691188842058182,
      "num_tokens": 32729687.0,
      "step": 12915
    },
    {
      "entropy": 6.24426794052124,
      "epoch": 1.491055972302366,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004788383774541399,
      "loss": 6.0592,
      "mean_token_accuracy": 0.16774754375219345,
      "num_tokens": 32741348.0,
      "step": 12920
    },
    {
      "entropy": 6.298770761489868,
      "epoch": 1.4916330063473744,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004788209031851402,
      "loss": 6.1267,
      "mean_token_accuracy": 0.1644484966993332,
      "num_tokens": 32754665.0,
      "step": 12925
    },
    {
      "entropy": 6.189231300354004,
      "epoch": 1.4922100403923833,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004788034220607256,
      "loss": 5.9993,
      "mean_token_accuracy": 0.16881134063005448,
      "num_tokens": 32767294.0,
      "step": 12930
    },
    {
      "entropy": 6.202424573898315,
      "epoch": 1.4927870744373917,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047878593408148405,
      "loss": 6.0244,
      "mean_token_accuracy": 0.16732936799526216,
      "num_tokens": 32779612.0,
      "step": 12935
    },
    {
      "entropy": 6.162698984146118,
      "epoch": 1.4933641084824005,
      "grad_norm": 2.453125,
      "learning_rate": 0.00047876843924800393,
      "loss": 6.0379,
      "mean_token_accuracy": 0.17055032253265381,
      "num_tokens": 32791826.0,
      "step": 12940
    },
    {
      "entropy": 6.105025720596314,
      "epoch": 1.4939411425274092,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004787509375608735,
      "loss": 5.9191,
      "mean_token_accuracy": 0.18344857394695283,
      "num_tokens": 32805044.0,
      "step": 12945
    },
    {
      "entropy": 6.188521528244019,
      "epoch": 1.4945181765724178,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047873342902068157,
      "loss": 6.0238,
      "mean_token_accuracy": 0.17168861329555513,
      "num_tokens": 32816797.0,
      "step": 12950
    },
    {
      "entropy": 6.17679796218872,
      "epoch": 1.4950952106174265,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047871591362801694,
      "loss": 5.9462,
      "mean_token_accuracy": 0.1767504930496216,
      "num_tokens": 32828670.0,
      "step": 12955
    },
    {
      "entropy": 6.278076314926148,
      "epoch": 1.495672244662435,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004786983913834687,
      "loss": 6.1291,
      "mean_token_accuracy": 0.15962619930505753,
      "num_tokens": 32841362.0,
      "step": 12960
    },
    {
      "entropy": 6.187237119674682,
      "epoch": 1.4962492787074437,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047868086228762633,
      "loss": 6.0104,
      "mean_token_accuracy": 0.1698865681886673,
      "num_tokens": 32853974.0,
      "step": 12965
    },
    {
      "entropy": 6.196225309371949,
      "epoch": 1.4968263127524524,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047866332634107926,
      "loss": 5.985,
      "mean_token_accuracy": 0.1731300249695778,
      "num_tokens": 32866683.0,
      "step": 12970
    },
    {
      "entropy": 6.245260000228882,
      "epoch": 1.497403346797461,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00047864578354441743,
      "loss": 6.0933,
      "mean_token_accuracy": 0.17037495821714402,
      "num_tokens": 32879294.0,
      "step": 12975
    },
    {
      "entropy": 6.0903685092926025,
      "epoch": 1.4979803808424696,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004786282338982308,
      "loss": 5.9372,
      "mean_token_accuracy": 0.1752368614077568,
      "num_tokens": 32891763.0,
      "step": 12980
    },
    {
      "entropy": 6.238949489593506,
      "epoch": 1.4985574148874783,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004786106774031096,
      "loss": 6.0856,
      "mean_token_accuracy": 0.16507013887166977,
      "num_tokens": 32903813.0,
      "step": 12985
    },
    {
      "entropy": 6.262663412094116,
      "epoch": 1.499134448932487,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004785931140596445,
      "loss": 6.0504,
      "mean_token_accuracy": 0.1720858633518219,
      "num_tokens": 32916847.0,
      "step": 12990
    },
    {
      "entropy": 6.17646279335022,
      "epoch": 1.4997114829774958,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047857554386842606,
      "loss": 5.9208,
      "mean_token_accuracy": 0.17650634348392485,
      "num_tokens": 32929727.0,
      "step": 12995
    },
    {
      "entropy": 6.183577013015747,
      "epoch": 1.5002885170225042,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047855796683004534,
      "loss": 6.0456,
      "mean_token_accuracy": 0.16841252148151398,
      "num_tokens": 32942140.0,
      "step": 13000
    },
    {
      "entropy": 6.250958776473999,
      "epoch": 1.500865551067513,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047854038294509356,
      "loss": 6.0447,
      "mean_token_accuracy": 0.16866240799427032,
      "num_tokens": 32954622.0,
      "step": 13005
    },
    {
      "entropy": 6.2002490043640135,
      "epoch": 1.5014425851125215,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004785227922141621,
      "loss": 6.0358,
      "mean_token_accuracy": 0.17986562699079514,
      "num_tokens": 32968118.0,
      "step": 13010
    },
    {
      "entropy": 6.220753574371338,
      "epoch": 1.5020196191575304,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047850519463784263,
      "loss": 6.0819,
      "mean_token_accuracy": 0.16689637303352356,
      "num_tokens": 32981275.0,
      "step": 13015
    },
    {
      "entropy": 6.318162441253662,
      "epoch": 1.502596653202539,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047848759021672693,
      "loss": 6.0884,
      "mean_token_accuracy": 0.1607356071472168,
      "num_tokens": 32995583.0,
      "step": 13020
    },
    {
      "entropy": 6.276556873321534,
      "epoch": 1.5031736872475476,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004784699789514073,
      "loss": 6.0415,
      "mean_token_accuracy": 0.16727252155542374,
      "num_tokens": 33008533.0,
      "step": 13025
    },
    {
      "entropy": 6.250124263763428,
      "epoch": 1.5037507212925563,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000478452360842476,
      "loss": 6.1488,
      "mean_token_accuracy": 0.16414920687675477,
      "num_tokens": 33023507.0,
      "step": 13030
    },
    {
      "entropy": 6.208919715881348,
      "epoch": 1.504327755337565,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047843473589052557,
      "loss": 5.9648,
      "mean_token_accuracy": 0.1750081568956375,
      "num_tokens": 33035655.0,
      "step": 13035
    },
    {
      "entropy": 6.22468113899231,
      "epoch": 1.5049047893825735,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047841710409614893,
      "loss": 6.067,
      "mean_token_accuracy": 0.16704362034797668,
      "num_tokens": 33047537.0,
      "step": 13040
    },
    {
      "entropy": 6.235771703720093,
      "epoch": 1.5054818234275822,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004783994654599389,
      "loss": 6.0288,
      "mean_token_accuracy": 0.17447531074285508,
      "num_tokens": 33060171.0,
      "step": 13045
    },
    {
      "entropy": 6.272533988952636,
      "epoch": 1.5060588574725908,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047838181998248897,
      "loss": 6.1016,
      "mean_token_accuracy": 0.16903458535671234,
      "num_tokens": 33073444.0,
      "step": 13050
    },
    {
      "entropy": 6.191557312011719,
      "epoch": 1.5066358915175995,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004783641676643925,
      "loss": 6.0988,
      "mean_token_accuracy": 0.16360597908496857,
      "num_tokens": 33086618.0,
      "step": 13055
    },
    {
      "entropy": 6.217271757125855,
      "epoch": 1.5072129255626083,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047834650850624334,
      "loss": 6.0054,
      "mean_token_accuracy": 0.1640032172203064,
      "num_tokens": 33098434.0,
      "step": 13060
    },
    {
      "entropy": 6.245991516113281,
      "epoch": 1.5077899596076167,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004783288425086353,
      "loss": 6.0293,
      "mean_token_accuracy": 0.16235798150300979,
      "num_tokens": 33110253.0,
      "step": 13065
    },
    {
      "entropy": 6.289529609680176,
      "epoch": 1.5083669936526256,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004783111696721627,
      "loss": 6.0696,
      "mean_token_accuracy": 0.17126161605119705,
      "num_tokens": 33122426.0,
      "step": 13070
    },
    {
      "entropy": 6.232216930389404,
      "epoch": 1.508944027697634,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004782934899974199,
      "loss": 5.9524,
      "mean_token_accuracy": 0.17151970267295838,
      "num_tokens": 33136609.0,
      "step": 13075
    },
    {
      "entropy": 6.194222927093506,
      "epoch": 1.5095210617426429,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047827580348500147,
      "loss": 6.1558,
      "mean_token_accuracy": 0.15753475129604338,
      "num_tokens": 33150351.0,
      "step": 13080
    },
    {
      "entropy": 6.254740476608276,
      "epoch": 1.5100980957876513,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047825811013550246,
      "loss": 6.0219,
      "mean_token_accuracy": 0.17257740944623948,
      "num_tokens": 33163256.0,
      "step": 13085
    },
    {
      "entropy": 6.2486409664154055,
      "epoch": 1.5106751298326602,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047824040994951786,
      "loss": 6.1033,
      "mean_token_accuracy": 0.16230327039957046,
      "num_tokens": 33177356.0,
      "step": 13090
    },
    {
      "entropy": 6.15275239944458,
      "epoch": 1.5112521638776688,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004782227029276429,
      "loss": 5.9589,
      "mean_token_accuracy": 0.17624272853136064,
      "num_tokens": 33189827.0,
      "step": 13095
    },
    {
      "entropy": 6.145482778549194,
      "epoch": 1.5118291979226774,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047820498907047345,
      "loss": 5.9043,
      "mean_token_accuracy": 0.17681120783090593,
      "num_tokens": 33202889.0,
      "step": 13100
    },
    {
      "entropy": 6.171259498596191,
      "epoch": 1.512406231967686,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000478187268378605,
      "loss": 5.9516,
      "mean_token_accuracy": 0.1757282242178917,
      "num_tokens": 33214973.0,
      "step": 13105
    },
    {
      "entropy": 6.252649402618408,
      "epoch": 1.5129832660126947,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047816954085263375,
      "loss": 6.0386,
      "mean_token_accuracy": 0.16758010238409043,
      "num_tokens": 33227515.0,
      "step": 13110
    },
    {
      "entropy": 6.283329772949219,
      "epoch": 1.5135603000577034,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004781518064931559,
      "loss": 6.0435,
      "mean_token_accuracy": 0.17407640665769578,
      "num_tokens": 33240858.0,
      "step": 13115
    },
    {
      "entropy": 6.174835824966431,
      "epoch": 1.514137334102712,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000478134065300768,
      "loss": 5.9649,
      "mean_token_accuracy": 0.17254628390073776,
      "num_tokens": 33253234.0,
      "step": 13120
    },
    {
      "entropy": 6.2249397277832035,
      "epoch": 1.5147143681477209,
      "grad_norm": 4.9375,
      "learning_rate": 0.0004781163172760667,
      "loss": 6.0497,
      "mean_token_accuracy": 0.1639400452375412,
      "num_tokens": 33265567.0,
      "step": 13125
    },
    {
      "entropy": 6.258707761764526,
      "epoch": 1.5152914021927293,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047809856241964893,
      "loss": 6.0603,
      "mean_token_accuracy": 0.16811907142400742,
      "num_tokens": 33278327.0,
      "step": 13130
    },
    {
      "entropy": 6.238284301757813,
      "epoch": 1.5158684362377381,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004780808007321119,
      "loss": 6.0422,
      "mean_token_accuracy": 0.1670468345284462,
      "num_tokens": 33291432.0,
      "step": 13135
    },
    {
      "entropy": 6.171168947219849,
      "epoch": 1.5164454702827466,
      "grad_norm": 1.125,
      "learning_rate": 0.0004780630322140531,
      "loss": 6.0781,
      "mean_token_accuracy": 0.1694598987698555,
      "num_tokens": 33304613.0,
      "step": 13140
    },
    {
      "entropy": 6.139973115921021,
      "epoch": 1.5170225043277554,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047804525686607,
      "loss": 5.9546,
      "mean_token_accuracy": 0.17407444417476653,
      "num_tokens": 33317258.0,
      "step": 13145
    },
    {
      "entropy": 6.160979747772217,
      "epoch": 1.5175995383727638,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004780274746887606,
      "loss": 6.001,
      "mean_token_accuracy": 0.1723225325345993,
      "num_tokens": 33329925.0,
      "step": 13150
    },
    {
      "entropy": 6.2055412292480465,
      "epoch": 1.5181765724177727,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047800968568272284,
      "loss": 5.976,
      "mean_token_accuracy": 0.17469632625579834,
      "num_tokens": 33343383.0,
      "step": 13155
    },
    {
      "entropy": 6.272555637359619,
      "epoch": 1.5187536064627813,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004779918898485551,
      "loss": 6.1139,
      "mean_token_accuracy": 0.156356580555439,
      "num_tokens": 33355960.0,
      "step": 13160
    },
    {
      "entropy": 6.25799789428711,
      "epoch": 1.51933064050779,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047797408718685614,
      "loss": 6.0538,
      "mean_token_accuracy": 0.16744242012500762,
      "num_tokens": 33368130.0,
      "step": 13165
    },
    {
      "entropy": 6.249115228652954,
      "epoch": 1.5199076745527986,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047795627769822447,
      "loss": 6.0865,
      "mean_token_accuracy": 0.16606530845165252,
      "num_tokens": 33381548.0,
      "step": 13170
    },
    {
      "entropy": 6.17447600364685,
      "epoch": 1.5204847085978073,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047793846138325925,
      "loss": 5.9831,
      "mean_token_accuracy": 0.1779807686805725,
      "num_tokens": 33393866.0,
      "step": 13175
    },
    {
      "entropy": 6.175324773788452,
      "epoch": 1.521061742642816,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004779206382425598,
      "loss": 6.0392,
      "mean_token_accuracy": 0.17075416445732117,
      "num_tokens": 33405683.0,
      "step": 13180
    },
    {
      "entropy": 6.190429258346557,
      "epoch": 1.5216387766878245,
      "grad_norm": 1.375,
      "learning_rate": 0.0004779028082767253,
      "loss": 5.9915,
      "mean_token_accuracy": 0.1777090087532997,
      "num_tokens": 33417049.0,
      "step": 13185
    },
    {
      "entropy": 6.170104217529297,
      "epoch": 1.5222158107328332,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004778849714863557,
      "loss": 5.9798,
      "mean_token_accuracy": 0.1774522691965103,
      "num_tokens": 33430106.0,
      "step": 13190
    },
    {
      "entropy": 6.1647742748260494,
      "epoch": 1.5227928447778418,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004778671278720508,
      "loss": 6.0163,
      "mean_token_accuracy": 0.17538946717977524,
      "num_tokens": 33443337.0,
      "step": 13195
    },
    {
      "entropy": 6.199500036239624,
      "epoch": 1.5233698788228507,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004778492774344109,
      "loss": 6.0306,
      "mean_token_accuracy": 0.17373048812150954,
      "num_tokens": 33458198.0,
      "step": 13200
    },
    {
      "entropy": 6.163569736480713,
      "epoch": 1.523946912867859,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004778314201740363,
      "loss": 5.8371,
      "mean_token_accuracy": 0.18836427181959153,
      "num_tokens": 33470169.0,
      "step": 13205
    },
    {
      "entropy": 6.2491371631622314,
      "epoch": 1.524523946912868,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047781355609152764,
      "loss": 6.094,
      "mean_token_accuracy": 0.17114727795124055,
      "num_tokens": 33482568.0,
      "step": 13210
    },
    {
      "entropy": 6.1838196277618405,
      "epoch": 1.5251009809578764,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004777956851874858,
      "loss": 5.9848,
      "mean_token_accuracy": 0.1765847846865654,
      "num_tokens": 33494092.0,
      "step": 13215
    },
    {
      "entropy": 6.268073511123657,
      "epoch": 1.5256780150028852,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047777780746251176,
      "loss": 6.0372,
      "mean_token_accuracy": 0.16972329318523408,
      "num_tokens": 33506961.0,
      "step": 13220
    },
    {
      "entropy": 6.195849514007568,
      "epoch": 1.5262550490478937,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047775992291720687,
      "loss": 6.048,
      "mean_token_accuracy": 0.16660842597484588,
      "num_tokens": 33519741.0,
      "step": 13225
    },
    {
      "entropy": 6.24710521697998,
      "epoch": 1.5268320830929025,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004777420315521728,
      "loss": 6.1004,
      "mean_token_accuracy": 0.1596265822649002,
      "num_tokens": 33532965.0,
      "step": 13230
    },
    {
      "entropy": 6.2416140079498295,
      "epoch": 1.5274091171379112,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004777241333680111,
      "loss": 5.9474,
      "mean_token_accuracy": 0.17829536646604538,
      "num_tokens": 33545979.0,
      "step": 13235
    },
    {
      "entropy": 6.149839210510254,
      "epoch": 1.5279861511829198,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004777062283653239,
      "loss": 6.0558,
      "mean_token_accuracy": 0.16851796358823776,
      "num_tokens": 33558148.0,
      "step": 13240
    },
    {
      "entropy": 6.254639530181885,
      "epoch": 1.5285631852279284,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00047768831654471333,
      "loss": 6.0036,
      "mean_token_accuracy": 0.18214876502752303,
      "num_tokens": 33570532.0,
      "step": 13245
    },
    {
      "entropy": 6.218555021286011,
      "epoch": 1.529140219272937,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047767039790678204,
      "loss": 6.006,
      "mean_token_accuracy": 0.16906740367412568,
      "num_tokens": 33582902.0,
      "step": 13250
    },
    {
      "entropy": 6.107049989700317,
      "epoch": 1.5297172533179457,
      "grad_norm": 1.125,
      "learning_rate": 0.00047765247245213255,
      "loss": 5.9355,
      "mean_token_accuracy": 0.17145386338233948,
      "num_tokens": 33594765.0,
      "step": 13255
    },
    {
      "entropy": 6.176938819885254,
      "epoch": 1.5302942873629544,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004776345401813678,
      "loss": 6.0318,
      "mean_token_accuracy": 0.16628430038690567,
      "num_tokens": 33606999.0,
      "step": 13260
    },
    {
      "entropy": 6.253328609466553,
      "epoch": 1.5308713214079632,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047761660109509095,
      "loss": 6.0652,
      "mean_token_accuracy": 0.16756111532449722,
      "num_tokens": 33619331.0,
      "step": 13265
    },
    {
      "entropy": 6.205666875839233,
      "epoch": 1.5314483554529716,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004775986551939054,
      "loss": 6.0261,
      "mean_token_accuracy": 0.1675693854689598,
      "num_tokens": 33632009.0,
      "step": 13270
    },
    {
      "entropy": 6.16805329322815,
      "epoch": 1.5320253894979805,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00047758070247841465,
      "loss": 5.9971,
      "mean_token_accuracy": 0.17036920189857482,
      "num_tokens": 33644468.0,
      "step": 13275
    },
    {
      "entropy": 6.152377557754517,
      "epoch": 1.532602423542989,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047756274294922266,
      "loss": 5.9018,
      "mean_token_accuracy": 0.18587636798620225,
      "num_tokens": 33657067.0,
      "step": 13280
    },
    {
      "entropy": 6.186591386795044,
      "epoch": 1.5331794575879978,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004775447766069334,
      "loss": 5.9996,
      "mean_token_accuracy": 0.17956343591213225,
      "num_tokens": 33669397.0,
      "step": 13285
    },
    {
      "entropy": 6.190193748474121,
      "epoch": 1.5337564916330062,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00047752680345215115,
      "loss": 5.9898,
      "mean_token_accuracy": 0.17159158885478973,
      "num_tokens": 33680997.0,
      "step": 13290
    },
    {
      "entropy": 6.177112102508545,
      "epoch": 1.534333525678015,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004775088234854805,
      "loss": 6.0001,
      "mean_token_accuracy": 0.17198223620653152,
      "num_tokens": 33693407.0,
      "step": 13295
    },
    {
      "entropy": 6.150848770141602,
      "epoch": 1.5349105597230237,
      "grad_norm": 1.25,
      "learning_rate": 0.0004774908367075262,
      "loss": 5.9355,
      "mean_token_accuracy": 0.18339731842279433,
      "num_tokens": 33705051.0,
      "step": 13300
    },
    {
      "entropy": 6.239871215820313,
      "epoch": 1.5354875937680323,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004774728431188931,
      "loss": 6.0723,
      "mean_token_accuracy": 0.16637013107538223,
      "num_tokens": 33717289.0,
      "step": 13305
    },
    {
      "entropy": 6.170883417129517,
      "epoch": 1.536064627813041,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047745484272018664,
      "loss": 5.9643,
      "mean_token_accuracy": 0.17800285518169404,
      "num_tokens": 33728494.0,
      "step": 13310
    },
    {
      "entropy": 6.151128721237183,
      "epoch": 1.5366416618580496,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004774368355120119,
      "loss": 6.071,
      "mean_token_accuracy": 0.17322352528572083,
      "num_tokens": 33742065.0,
      "step": 13315
    },
    {
      "entropy": 6.180449676513672,
      "epoch": 1.5372186959030583,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000477418821494975,
      "loss": 5.9615,
      "mean_token_accuracy": 0.1715274080634117,
      "num_tokens": 33754312.0,
      "step": 13320
    },
    {
      "entropy": 6.154176712036133,
      "epoch": 1.537795729948067,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004774008006696814,
      "loss": 6.0332,
      "mean_token_accuracy": 0.17376970648765563,
      "num_tokens": 33766938.0,
      "step": 13325
    },
    {
      "entropy": 6.204604053497315,
      "epoch": 1.5383727639930755,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004773827730367375,
      "loss": 6.0209,
      "mean_token_accuracy": 0.17332495599985123,
      "num_tokens": 33778494.0,
      "step": 13330
    },
    {
      "entropy": 6.12020206451416,
      "epoch": 1.5389497980380842,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047736473859674955,
      "loss": 5.9219,
      "mean_token_accuracy": 0.18146518617868423,
      "num_tokens": 33789932.0,
      "step": 13335
    },
    {
      "entropy": 6.07808141708374,
      "epoch": 1.539526832083093,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047734669735032404,
      "loss": 6.0606,
      "mean_token_accuracy": 0.1685696855187416,
      "num_tokens": 33803552.0,
      "step": 13340
    },
    {
      "entropy": 6.158413696289062,
      "epoch": 1.5401038661281015,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047732864929806796,
      "loss": 5.9623,
      "mean_token_accuracy": 0.17478156089782715,
      "num_tokens": 33816108.0,
      "step": 13345
    },
    {
      "entropy": 6.315565395355224,
      "epoch": 1.5406809001731103,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004773105944405883,
      "loss": 6.1066,
      "mean_token_accuracy": 0.16569467037916183,
      "num_tokens": 33829004.0,
      "step": 13350
    },
    {
      "entropy": 6.206098556518555,
      "epoch": 1.5412579342181187,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047729253277849226,
      "loss": 6.0286,
      "mean_token_accuracy": 0.1723840579390526,
      "num_tokens": 33840889.0,
      "step": 13355
    },
    {
      "entropy": 6.201770257949829,
      "epoch": 1.5418349682631276,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047727446431238734,
      "loss": 6.0248,
      "mean_token_accuracy": 0.1696290612220764,
      "num_tokens": 33855443.0,
      "step": 13360
    },
    {
      "entropy": 6.199795484542847,
      "epoch": 1.542412002308136,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004772563890428813,
      "loss": 6.0422,
      "mean_token_accuracy": 0.1666147977113724,
      "num_tokens": 33867365.0,
      "step": 13365
    },
    {
      "entropy": 6.1983802318573,
      "epoch": 1.5429890363531449,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004772383069705821,
      "loss": 5.9866,
      "mean_token_accuracy": 0.1712106317281723,
      "num_tokens": 33880438.0,
      "step": 13370
    },
    {
      "entropy": 6.207140350341797,
      "epoch": 1.5435660703981535,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004772202180960978,
      "loss": 6.0348,
      "mean_token_accuracy": 0.16188852936029435,
      "num_tokens": 33894812.0,
      "step": 13375
    },
    {
      "entropy": 6.227739763259888,
      "epoch": 1.5441431044431622,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047720212242003703,
      "loss": 5.9897,
      "mean_token_accuracy": 0.16861477345228196,
      "num_tokens": 33907619.0,
      "step": 13380
    },
    {
      "entropy": 6.207125806808472,
      "epoch": 1.5447201384881708,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047718401994300825,
      "loss": 6.0359,
      "mean_token_accuracy": 0.16943563371896744,
      "num_tokens": 33921109.0,
      "step": 13385
    },
    {
      "entropy": 6.219919586181641,
      "epoch": 1.5452971725331794,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047716591066562043,
      "loss": 6.0259,
      "mean_token_accuracy": 0.1624724119901657,
      "num_tokens": 33932874.0,
      "step": 13390
    },
    {
      "entropy": 6.236322116851807,
      "epoch": 1.545874206578188,
      "grad_norm": 2.453125,
      "learning_rate": 0.00047714779458848255,
      "loss": 6.0462,
      "mean_token_accuracy": 0.17164248824119568,
      "num_tokens": 33945291.0,
      "step": 13395
    },
    {
      "entropy": 6.113810873031616,
      "epoch": 1.5464512406231967,
      "grad_norm": 1.125,
      "learning_rate": 0.0004771296717122041,
      "loss": 5.8972,
      "mean_token_accuracy": 0.1820332556962967,
      "num_tokens": 33958180.0,
      "step": 13400
    },
    {
      "entropy": 6.303949308395386,
      "epoch": 1.5470282746682056,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004771115420373945,
      "loss": 6.1725,
      "mean_token_accuracy": 0.14768412113189697,
      "num_tokens": 33971428.0,
      "step": 13405
    },
    {
      "entropy": 6.305074167251587,
      "epoch": 1.547605308713214,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047709340556466366,
      "loss": 5.9882,
      "mean_token_accuracy": 0.17419161647558212,
      "num_tokens": 33984464.0,
      "step": 13410
    },
    {
      "entropy": 6.1262242794036865,
      "epoch": 1.5481823427582229,
      "grad_norm": 1.25,
      "learning_rate": 0.00047707526229462144,
      "loss": 5.9658,
      "mean_token_accuracy": 0.17201746702194215,
      "num_tokens": 33997290.0,
      "step": 13415
    },
    {
      "entropy": 6.178758716583252,
      "epoch": 1.5487593768032313,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047705711222787816,
      "loss": 6.0381,
      "mean_token_accuracy": 0.16968253999948502,
      "num_tokens": 34009249.0,
      "step": 13420
    },
    {
      "entropy": 6.218361806869507,
      "epoch": 1.5493364108482401,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047703895536504423,
      "loss": 5.9807,
      "mean_token_accuracy": 0.1707553192973137,
      "num_tokens": 34022583.0,
      "step": 13425
    },
    {
      "entropy": 6.298037958145142,
      "epoch": 1.5499134448932486,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004770207917067305,
      "loss": 6.028,
      "mean_token_accuracy": 0.17416543662548065,
      "num_tokens": 34036987.0,
      "step": 13430
    },
    {
      "entropy": 6.185347652435302,
      "epoch": 1.5504904789382574,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047700262125354765,
      "loss": 6.0874,
      "mean_token_accuracy": 0.166138257086277,
      "num_tokens": 34050385.0,
      "step": 13435
    },
    {
      "entropy": 6.200943183898926,
      "epoch": 1.551067512983266,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047698444400610707,
      "loss": 6.0193,
      "mean_token_accuracy": 0.1743549033999443,
      "num_tokens": 34064006.0,
      "step": 13440
    },
    {
      "entropy": 6.2587357521057125,
      "epoch": 1.5516445470282747,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047696625996502,
      "loss": 6.0526,
      "mean_token_accuracy": 0.17369845062494277,
      "num_tokens": 34075994.0,
      "step": 13445
    },
    {
      "entropy": 6.264259099960327,
      "epoch": 1.5522215810732833,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047694806913089815,
      "loss": 6.0658,
      "mean_token_accuracy": 0.16074010878801345,
      "num_tokens": 34087430.0,
      "step": 13450
    },
    {
      "entropy": 6.225534248352051,
      "epoch": 1.552798615118292,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004769298715043533,
      "loss": 6.0512,
      "mean_token_accuracy": 0.16836937218904496,
      "num_tokens": 34099951.0,
      "step": 13455
    },
    {
      "entropy": 6.179991149902344,
      "epoch": 1.5533756491633006,
      "grad_norm": 1.375,
      "learning_rate": 0.0004769116670859975,
      "loss": 5.9755,
      "mean_token_accuracy": 0.17226701825857163,
      "num_tokens": 34112048.0,
      "step": 13460
    },
    {
      "entropy": 6.286797761917114,
      "epoch": 1.5539526832083093,
      "grad_norm": 1.125,
      "learning_rate": 0.00047689345587644314,
      "loss": 6.1223,
      "mean_token_accuracy": 0.1711254119873047,
      "num_tokens": 34124820.0,
      "step": 13465
    },
    {
      "entropy": 6.12939600944519,
      "epoch": 1.554529717253318,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047687523787630256,
      "loss": 5.9357,
      "mean_token_accuracy": 0.17159577906131745,
      "num_tokens": 34137792.0,
      "step": 13470
    },
    {
      "entropy": 6.296533584594727,
      "epoch": 1.5551067512983265,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004768570130861887,
      "loss": 6.0662,
      "mean_token_accuracy": 0.1744540736079216,
      "num_tokens": 34149709.0,
      "step": 13475
    },
    {
      "entropy": 6.278332757949829,
      "epoch": 1.5556837853433354,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004768387815067144,
      "loss": 6.1931,
      "mean_token_accuracy": 0.1630217105150223,
      "num_tokens": 34163958.0,
      "step": 13480
    },
    {
      "entropy": 6.200136423110962,
      "epoch": 1.5562608193883438,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047682054313849304,
      "loss": 6.0071,
      "mean_token_accuracy": 0.16974304467439652,
      "num_tokens": 34176349.0,
      "step": 13485
    },
    {
      "entropy": 6.281996488571167,
      "epoch": 1.5568378534333527,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004768022979821379,
      "loss": 6.0757,
      "mean_token_accuracy": 0.16058100759983063,
      "num_tokens": 34188000.0,
      "step": 13490
    },
    {
      "entropy": 6.218197822570801,
      "epoch": 1.557414887478361,
      "grad_norm": 1.125,
      "learning_rate": 0.0004767840460382628,
      "loss": 5.9944,
      "mean_token_accuracy": 0.16840729117393494,
      "num_tokens": 34199675.0,
      "step": 13495
    },
    {
      "entropy": 6.211877775192261,
      "epoch": 1.55799192152337,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004767657873074815,
      "loss": 5.9897,
      "mean_token_accuracy": 0.16813745051622392,
      "num_tokens": 34212443.0,
      "step": 13500
    },
    {
      "entropy": 6.244054889678955,
      "epoch": 1.5585689555683784,
      "grad_norm": 1.125,
      "learning_rate": 0.0004767475217904081,
      "loss": 6.0731,
      "mean_token_accuracy": 0.16443396210670472,
      "num_tokens": 34224664.0,
      "step": 13505
    },
    {
      "entropy": 6.0536208152771,
      "epoch": 1.5591459896133872,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047672924948765705,
      "loss": 5.8937,
      "mean_token_accuracy": 0.17990826964378356,
      "num_tokens": 34236418.0,
      "step": 13510
    },
    {
      "entropy": 6.2100402355194095,
      "epoch": 1.5597230236583959,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047671097039984293,
      "loss": 5.9783,
      "mean_token_accuracy": 0.17179360389709472,
      "num_tokens": 34249508.0,
      "step": 13515
    },
    {
      "entropy": 6.245606851577759,
      "epoch": 1.5603000577034045,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047669268452758053,
      "loss": 6.064,
      "mean_token_accuracy": 0.16874691247940063,
      "num_tokens": 34263077.0,
      "step": 13520
    },
    {
      "entropy": 6.232436418533325,
      "epoch": 1.5608770917484132,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047667439187148476,
      "loss": 5.9257,
      "mean_token_accuracy": 0.18215187638998032,
      "num_tokens": 34276026.0,
      "step": 13525
    },
    {
      "entropy": 6.210673475265503,
      "epoch": 1.5614541257934218,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004766560924321711,
      "loss": 6.0883,
      "mean_token_accuracy": 0.16359824538230897,
      "num_tokens": 34288398.0,
      "step": 13530
    },
    {
      "entropy": 6.2507829666137695,
      "epoch": 1.5620311598384304,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047663778621025496,
      "loss": 6.0355,
      "mean_token_accuracy": 0.1632818415760994,
      "num_tokens": 34301064.0,
      "step": 13535
    },
    {
      "entropy": 6.259791660308838,
      "epoch": 1.562608193883439,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004766194732063519,
      "loss": 6.0301,
      "mean_token_accuracy": 0.17318796068429948,
      "num_tokens": 34312959.0,
      "step": 13540
    },
    {
      "entropy": 6.140162944793701,
      "epoch": 1.563185227928448,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047660115342107814,
      "loss": 5.976,
      "mean_token_accuracy": 0.17263364046812057,
      "num_tokens": 34325851.0,
      "step": 13545
    },
    {
      "entropy": 6.235706233978272,
      "epoch": 1.5637622619734564,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047658282685504973,
      "loss": 5.9657,
      "mean_token_accuracy": 0.17633154541254042,
      "num_tokens": 34338193.0,
      "step": 13550
    },
    {
      "entropy": 6.200155258178711,
      "epoch": 1.5643392960184652,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000476564493508883,
      "loss": 6.0821,
      "mean_token_accuracy": 0.16353416591882705,
      "num_tokens": 34350876.0,
      "step": 13555
    },
    {
      "entropy": 6.119752645492554,
      "epoch": 1.5649163300634736,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047654615338319466,
      "loss": 5.9345,
      "mean_token_accuracy": 0.179782697558403,
      "num_tokens": 34363614.0,
      "step": 13560
    },
    {
      "entropy": 6.147815179824829,
      "epoch": 1.5654933641084825,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004765278064786016,
      "loss": 6.0299,
      "mean_token_accuracy": 0.1766850009560585,
      "num_tokens": 34376238.0,
      "step": 13565
    },
    {
      "entropy": 6.258609104156494,
      "epoch": 1.566070398153491,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047650945279572085,
      "loss": 6.0318,
      "mean_token_accuracy": 0.17146103382110595,
      "num_tokens": 34389141.0,
      "step": 13570
    },
    {
      "entropy": 6.161709833145141,
      "epoch": 1.5666474321984998,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004764910923351698,
      "loss": 5.949,
      "mean_token_accuracy": 0.17622106671333312,
      "num_tokens": 34400844.0,
      "step": 13575
    },
    {
      "entropy": 6.089479303359985,
      "epoch": 1.5672244662435084,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047647272509756584,
      "loss": 5.904,
      "mean_token_accuracy": 0.1743240922689438,
      "num_tokens": 34413558.0,
      "step": 13580
    },
    {
      "entropy": 6.188379240036011,
      "epoch": 1.567801500288517,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004764543510835269,
      "loss": 6.0013,
      "mean_token_accuracy": 0.17119977474212647,
      "num_tokens": 34425189.0,
      "step": 13585
    },
    {
      "entropy": 6.202872467041016,
      "epoch": 1.5683785343335257,
      "grad_norm": 1.1875,
      "learning_rate": 0.000476435970293671,
      "loss": 5.9713,
      "mean_token_accuracy": 0.18041349053382874,
      "num_tokens": 34438649.0,
      "step": 13590
    },
    {
      "entropy": 6.148017311096192,
      "epoch": 1.5689555683785343,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047641758272861626,
      "loss": 5.9477,
      "mean_token_accuracy": 0.17446087896823884,
      "num_tokens": 34451159.0,
      "step": 13595
    },
    {
      "entropy": 6.24504828453064,
      "epoch": 1.569532602423543,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004763991883889811,
      "loss": 6.0301,
      "mean_token_accuracy": 0.161186982691288,
      "num_tokens": 34462821.0,
      "step": 13600
    },
    {
      "entropy": 6.203308486938477,
      "epoch": 1.5701096364685516,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004763807872753843,
      "loss": 6.0273,
      "mean_token_accuracy": 0.16369927674531937,
      "num_tokens": 34474845.0,
      "step": 13605
    },
    {
      "entropy": 6.187725496292114,
      "epoch": 1.5706866705135603,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047636237938844484,
      "loss": 6.0737,
      "mean_token_accuracy": 0.16586184054613112,
      "num_tokens": 34488501.0,
      "step": 13610
    },
    {
      "entropy": 6.219933319091797,
      "epoch": 1.571263704558569,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004763439647287817,
      "loss": 6.0144,
      "mean_token_accuracy": 0.1675948068499565,
      "num_tokens": 34500760.0,
      "step": 13615
    },
    {
      "entropy": 6.17669563293457,
      "epoch": 1.5718407386035778,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004763255432970144,
      "loss": 5.9826,
      "mean_token_accuracy": 0.17187965661287308,
      "num_tokens": 34513857.0,
      "step": 13620
    },
    {
      "entropy": 6.233660793304443,
      "epoch": 1.5724177726485862,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047630711509376235,
      "loss": 5.9721,
      "mean_token_accuracy": 0.1741819441318512,
      "num_tokens": 34526091.0,
      "step": 13625
    },
    {
      "entropy": 6.253244638442993,
      "epoch": 1.572994806693595,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004762886801196455,
      "loss": 6.0445,
      "mean_token_accuracy": 0.16377142667770386,
      "num_tokens": 34539353.0,
      "step": 13630
    },
    {
      "entropy": 6.1121604442596436,
      "epoch": 1.5735718407386035,
      "grad_norm": 1.125,
      "learning_rate": 0.00047627023837528386,
      "loss": 5.9163,
      "mean_token_accuracy": 0.18128162622451782,
      "num_tokens": 34551633.0,
      "step": 13635
    },
    {
      "entropy": 6.231601762771606,
      "epoch": 1.5741488747836123,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047625178986129775,
      "loss": 6.0444,
      "mean_token_accuracy": 0.1698276624083519,
      "num_tokens": 34563108.0,
      "step": 13640
    },
    {
      "entropy": 6.252461910247803,
      "epoch": 1.5747259088286207,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004762333345783078,
      "loss": 6.1065,
      "mean_token_accuracy": 0.16844332367181777,
      "num_tokens": 34576014.0,
      "step": 13645
    },
    {
      "entropy": 6.268393230438233,
      "epoch": 1.5753029428736296,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047621487252693436,
      "loss": 6.0668,
      "mean_token_accuracy": 0.16375408619642257,
      "num_tokens": 34590025.0,
      "step": 13650
    },
    {
      "entropy": 6.217789030075073,
      "epoch": 1.5758799769186382,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047619640370779876,
      "loss": 5.9819,
      "mean_token_accuracy": 0.17273401468992233,
      "num_tokens": 34601627.0,
      "step": 13655
    },
    {
      "entropy": 6.13266806602478,
      "epoch": 1.5764570109636469,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047617792812152207,
      "loss": 5.9482,
      "mean_token_accuracy": 0.17711748629808427,
      "num_tokens": 34613712.0,
      "step": 13660
    },
    {
      "entropy": 6.196488666534424,
      "epoch": 1.5770340450086555,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004761594457687256,
      "loss": 6.0935,
      "mean_token_accuracy": 0.16308836191892623,
      "num_tokens": 34625861.0,
      "step": 13665
    },
    {
      "entropy": 6.30399603843689,
      "epoch": 1.5776110790536642,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004761409566500313,
      "loss": 6.1279,
      "mean_token_accuracy": 0.16192880719900132,
      "num_tokens": 34638457.0,
      "step": 13670
    },
    {
      "entropy": 6.200469160079956,
      "epoch": 1.5781881130986728,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047612246076606066,
      "loss": 6.0177,
      "mean_token_accuracy": 0.16582965850830078,
      "num_tokens": 34652037.0,
      "step": 13675
    },
    {
      "entropy": 6.172366142272949,
      "epoch": 1.5787651471436814,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047610395811743594,
      "loss": 6.01,
      "mean_token_accuracy": 0.17226552367210388,
      "num_tokens": 34664902.0,
      "step": 13680
    },
    {
      "entropy": 6.2235805034637455,
      "epoch": 1.5793421811886903,
      "grad_norm": 1.25,
      "learning_rate": 0.00047608544870477956,
      "loss": 6.0267,
      "mean_token_accuracy": 0.16638857871294022,
      "num_tokens": 34676772.0,
      "step": 13685
    },
    {
      "entropy": 6.17095251083374,
      "epoch": 1.5799192152336987,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047606693252871395,
      "loss": 6.0232,
      "mean_token_accuracy": 0.17537558376789092,
      "num_tokens": 34690013.0,
      "step": 13690
    },
    {
      "entropy": 6.188656568527222,
      "epoch": 1.5804962492787076,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000476048409589862,
      "loss": 5.9907,
      "mean_token_accuracy": 0.1694439619779587,
      "num_tokens": 34702855.0,
      "step": 13695
    },
    {
      "entropy": 6.257997512817383,
      "epoch": 1.581073283323716,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004760298798888466,
      "loss": 6.067,
      "mean_token_accuracy": 0.1749182552099228,
      "num_tokens": 34716926.0,
      "step": 13700
    },
    {
      "entropy": 6.2890034198760985,
      "epoch": 1.5816503173687249,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004760113434262911,
      "loss": 6.0493,
      "mean_token_accuracy": 0.16408467590808867,
      "num_tokens": 34729528.0,
      "step": 13705
    },
    {
      "entropy": 6.245816993713379,
      "epoch": 1.5822273514137333,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047599280020281894,
      "loss": 6.104,
      "mean_token_accuracy": 0.16557103395462036,
      "num_tokens": 34742016.0,
      "step": 13710
    },
    {
      "entropy": 6.229723024368286,
      "epoch": 1.5828043854587421,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047597425021905364,
      "loss": 5.9833,
      "mean_token_accuracy": 0.17329696267843248,
      "num_tokens": 34754670.0,
      "step": 13715
    },
    {
      "entropy": 6.180801916122436,
      "epoch": 1.5833814195037508,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004759556934756194,
      "loss": 5.9484,
      "mean_token_accuracy": 0.17539605647325515,
      "num_tokens": 34767411.0,
      "step": 13720
    },
    {
      "entropy": 6.170360469818116,
      "epoch": 1.5839584535487594,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047593712997314017,
      "loss": 5.9756,
      "mean_token_accuracy": 0.1778625562787056,
      "num_tokens": 34780498.0,
      "step": 13725
    },
    {
      "entropy": 6.229995632171631,
      "epoch": 1.584535487593768,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047591855971224035,
      "loss": 6.0193,
      "mean_token_accuracy": 0.1733413740992546,
      "num_tokens": 34792918.0,
      "step": 13730
    },
    {
      "entropy": 6.1417567253112795,
      "epoch": 1.5851125216387767,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004758999826935446,
      "loss": 5.9451,
      "mean_token_accuracy": 0.16881987899541856,
      "num_tokens": 34805631.0,
      "step": 13735
    },
    {
      "entropy": 6.211400079727173,
      "epoch": 1.5856895556837853,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004758813989176778,
      "loss": 6.1203,
      "mean_token_accuracy": 0.16123454123735428,
      "num_tokens": 34818790.0,
      "step": 13740
    },
    {
      "entropy": 6.212410640716553,
      "epoch": 1.586266589728794,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047586280838526483,
      "loss": 5.9331,
      "mean_token_accuracy": 0.17979222387075425,
      "num_tokens": 34831701.0,
      "step": 13745
    },
    {
      "entropy": 6.188838148117066,
      "epoch": 1.5868436237738026,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004758442110969312,
      "loss": 6.0022,
      "mean_token_accuracy": 0.17371300458908082,
      "num_tokens": 34843628.0,
      "step": 13750
    },
    {
      "entropy": 6.168396520614624,
      "epoch": 1.5874206578188113,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004758256070533022,
      "loss": 5.9972,
      "mean_token_accuracy": 0.16509783416986465,
      "num_tokens": 34856327.0,
      "step": 13755
    },
    {
      "entropy": 6.199730062484742,
      "epoch": 1.5879976918638201,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004758069962550037,
      "loss": 5.9883,
      "mean_token_accuracy": 0.1696273148059845,
      "num_tokens": 34869973.0,
      "step": 13760
    },
    {
      "entropy": 6.134802627563476,
      "epoch": 1.5885747259088285,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047578837870266156,
      "loss": 5.9792,
      "mean_token_accuracy": 0.1832378551363945,
      "num_tokens": 34884633.0,
      "step": 13765
    },
    {
      "entropy": 6.270034980773926,
      "epoch": 1.5891517599538374,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047576975439690206,
      "loss": 6.0087,
      "mean_token_accuracy": 0.17153942286968232,
      "num_tokens": 34897364.0,
      "step": 13770
    },
    {
      "entropy": 6.106852769851685,
      "epoch": 1.5897287939988458,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047575112333835164,
      "loss": 5.9111,
      "mean_token_accuracy": 0.17637831419706346,
      "num_tokens": 34909789.0,
      "step": 13775
    },
    {
      "entropy": 6.147295999526977,
      "epoch": 1.5903058280438547,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00047573248552763684,
      "loss": 6.0119,
      "mean_token_accuracy": 0.16346476078033448,
      "num_tokens": 34923127.0,
      "step": 13780
    },
    {
      "entropy": 6.234229612350464,
      "epoch": 1.590882862088863,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047571384096538474,
      "loss": 6.0602,
      "mean_token_accuracy": 0.16443841606378556,
      "num_tokens": 34935710.0,
      "step": 13785
    },
    {
      "entropy": 6.112004280090332,
      "epoch": 1.591459896133872,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004756951896522222,
      "loss": 5.8937,
      "mean_token_accuracy": 0.17286065965890884,
      "num_tokens": 34947566.0,
      "step": 13790
    },
    {
      "entropy": 6.198678493499756,
      "epoch": 1.5920369301788806,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004756765315887766,
      "loss": 5.9939,
      "mean_token_accuracy": 0.1717917203903198,
      "num_tokens": 34960008.0,
      "step": 13795
    },
    {
      "entropy": 6.1810697555542,
      "epoch": 1.5926139642238892,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004756578667756756,
      "loss": 5.9756,
      "mean_token_accuracy": 0.1702072277665138,
      "num_tokens": 34972299.0,
      "step": 13800
    },
    {
      "entropy": 6.22912073135376,
      "epoch": 1.5931909982688979,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004756391952135469,
      "loss": 6.0262,
      "mean_token_accuracy": 0.1676354467868805,
      "num_tokens": 34984252.0,
      "step": 13805
    },
    {
      "entropy": 6.226155042648315,
      "epoch": 1.5937680323139065,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047562051690301855,
      "loss": 6.018,
      "mean_token_accuracy": 0.17211539745330812,
      "num_tokens": 34997711.0,
      "step": 13810
    },
    {
      "entropy": 6.21802134513855,
      "epoch": 1.5943450663589152,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00047560183184471873,
      "loss": 5.9703,
      "mean_token_accuracy": 0.17578355073928834,
      "num_tokens": 35010440.0,
      "step": 13815
    },
    {
      "entropy": 6.17285189628601,
      "epoch": 1.5949221004039238,
      "grad_norm": 1.015625,
      "learning_rate": 0.000475583140039276,
      "loss": 5.9367,
      "mean_token_accuracy": 0.17385512590408325,
      "num_tokens": 35022493.0,
      "step": 13820
    },
    {
      "entropy": 6.153857088088989,
      "epoch": 1.5954991344489324,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047556444148731897,
      "loss": 5.9884,
      "mean_token_accuracy": 0.17339059263467788,
      "num_tokens": 35035756.0,
      "step": 13825
    },
    {
      "entropy": 6.233834266662598,
      "epoch": 1.596076168493941,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004755457361894766,
      "loss": 5.9658,
      "mean_token_accuracy": 0.17424436360597612,
      "num_tokens": 35046946.0,
      "step": 13830
    },
    {
      "entropy": 6.20965986251831,
      "epoch": 1.59665320253895,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047552702414637793,
      "loss": 5.9982,
      "mean_token_accuracy": 0.15816480964422225,
      "num_tokens": 35060803.0,
      "step": 13835
    },
    {
      "entropy": 6.1471521854400635,
      "epoch": 1.5972302365839584,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047550830535865245,
      "loss": 5.9345,
      "mean_token_accuracy": 0.17486619651317598,
      "num_tokens": 35072834.0,
      "step": 13840
    },
    {
      "entropy": 6.27711706161499,
      "epoch": 1.5978072706289672,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047548957982692974,
      "loss": 6.0823,
      "mean_token_accuracy": 0.16915625631809234,
      "num_tokens": 35084872.0,
      "step": 13845
    },
    {
      "entropy": 6.273464202880859,
      "epoch": 1.5983843046739756,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004754708475518396,
      "loss": 6.1025,
      "mean_token_accuracy": 0.16550000309944152,
      "num_tokens": 35098070.0,
      "step": 13850
    },
    {
      "entropy": 6.269813442230225,
      "epoch": 1.5989613387189845,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047545210853401214,
      "loss": 6.1077,
      "mean_token_accuracy": 0.16680012941360473,
      "num_tokens": 35111614.0,
      "step": 13855
    },
    {
      "entropy": 6.245347881317139,
      "epoch": 1.599538372763993,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047543336277407753,
      "loss": 6.0794,
      "mean_token_accuracy": 0.1649049252271652,
      "num_tokens": 35124061.0,
      "step": 13860
    },
    {
      "entropy": 6.222730016708374,
      "epoch": 1.6001154068090018,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047541461027266624,
      "loss": 5.9426,
      "mean_token_accuracy": 0.1679711401462555,
      "num_tokens": 35135853.0,
      "step": 13865
    },
    {
      "entropy": 6.217486095428467,
      "epoch": 1.6006924408540104,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004753958510304091,
      "loss": 5.9812,
      "mean_token_accuracy": 0.16499261409044266,
      "num_tokens": 35147811.0,
      "step": 13870
    },
    {
      "entropy": 6.222936582565308,
      "epoch": 1.601269474899019,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047537708504793714,
      "loss": 6.032,
      "mean_token_accuracy": 0.17667135298252107,
      "num_tokens": 35161738.0,
      "step": 13875
    },
    {
      "entropy": 6.122700881958008,
      "epoch": 1.6018465089440277,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047535831232588146,
      "loss": 5.9746,
      "mean_token_accuracy": 0.17978452444076537,
      "num_tokens": 35174777.0,
      "step": 13880
    },
    {
      "entropy": 6.2753301620483395,
      "epoch": 1.6024235429890363,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047533953286487345,
      "loss": 6.09,
      "mean_token_accuracy": 0.1655922144651413,
      "num_tokens": 35186814.0,
      "step": 13885
    },
    {
      "entropy": 6.21658935546875,
      "epoch": 1.603000577034045,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004753207466655447,
      "loss": 5.9834,
      "mean_token_accuracy": 0.16464858651161193,
      "num_tokens": 35199229.0,
      "step": 13890
    },
    {
      "entropy": 6.1342321872711185,
      "epoch": 1.6035776110790536,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004753019537285273,
      "loss": 5.977,
      "mean_token_accuracy": 0.1721646308898926,
      "num_tokens": 35212110.0,
      "step": 13895
    },
    {
      "entropy": 6.273949480056762,
      "epoch": 1.6041546451240625,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047528315405445296,
      "loss": 6.1017,
      "mean_token_accuracy": 0.16062945425510405,
      "num_tokens": 35226203.0,
      "step": 13900
    },
    {
      "entropy": 6.185816669464112,
      "epoch": 1.604731679169071,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047526434764395435,
      "loss": 5.9841,
      "mean_token_accuracy": 0.17051070481538771,
      "num_tokens": 35238537.0,
      "step": 13905
    },
    {
      "entropy": 6.164809131622315,
      "epoch": 1.6053087132140798,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047524553449766386,
      "loss": 5.9889,
      "mean_token_accuracy": 0.17340071350336075,
      "num_tokens": 35251444.0,
      "step": 13910
    },
    {
      "entropy": 6.234498310089111,
      "epoch": 1.6058857472590882,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047522671461621433,
      "loss": 5.9981,
      "mean_token_accuracy": 0.17286839187145234,
      "num_tokens": 35264159.0,
      "step": 13915
    },
    {
      "entropy": 6.2039961338043215,
      "epoch": 1.606462781304097,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004752078880002386,
      "loss": 6.0205,
      "mean_token_accuracy": 0.1634536862373352,
      "num_tokens": 35277475.0,
      "step": 13920
    },
    {
      "entropy": 6.149963998794556,
      "epoch": 1.6070398153491054,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047518905465037005,
      "loss": 5.897,
      "mean_token_accuracy": 0.18117384612560272,
      "num_tokens": 35290739.0,
      "step": 13925
    },
    {
      "entropy": 6.097462797164917,
      "epoch": 1.6076168493941143,
      "grad_norm": 1.0,
      "learning_rate": 0.00047517021456724214,
      "loss": 5.9955,
      "mean_token_accuracy": 0.16663464903831482,
      "num_tokens": 35303053.0,
      "step": 13930
    },
    {
      "entropy": 6.219984436035157,
      "epoch": 1.608193883439123,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047515136775148843,
      "loss": 5.9946,
      "mean_token_accuracy": 0.17749239951372148,
      "num_tokens": 35316105.0,
      "step": 13935
    },
    {
      "entropy": 6.250874614715576,
      "epoch": 1.6087709174841316,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004751325142037429,
      "loss": 5.9819,
      "mean_token_accuracy": 0.17225621342658998,
      "num_tokens": 35329144.0,
      "step": 13940
    },
    {
      "entropy": 6.081976699829101,
      "epoch": 1.6093479515291402,
      "grad_norm": 1.125,
      "learning_rate": 0.00047511365392463974,
      "loss": 5.8603,
      "mean_token_accuracy": 0.17702547162771226,
      "num_tokens": 35342447.0,
      "step": 13945
    },
    {
      "entropy": 6.223639249801636,
      "epoch": 1.6099249855741489,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047509478691481317,
      "loss": 6.0141,
      "mean_token_accuracy": 0.17055661231279373,
      "num_tokens": 35354073.0,
      "step": 13950
    },
    {
      "entropy": 6.209834480285645,
      "epoch": 1.6105020196191575,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047507591317489783,
      "loss": 5.9455,
      "mean_token_accuracy": 0.17948899567127227,
      "num_tokens": 35367286.0,
      "step": 13955
    },
    {
      "entropy": 6.202068853378296,
      "epoch": 1.6110790536641661,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004750570327055286,
      "loss": 6.0185,
      "mean_token_accuracy": 0.17567377984523774,
      "num_tokens": 35379262.0,
      "step": 13960
    },
    {
      "entropy": 6.250647068023682,
      "epoch": 1.6116560877091748,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047503814550734034,
      "loss": 5.9602,
      "mean_token_accuracy": 0.17177603542804717,
      "num_tokens": 35392423.0,
      "step": 13965
    },
    {
      "entropy": 6.239200305938721,
      "epoch": 1.6122331217541834,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004750192515809685,
      "loss": 6.1148,
      "mean_token_accuracy": 0.1647377207875252,
      "num_tokens": 35405007.0,
      "step": 13970
    },
    {
      "entropy": 6.116919231414795,
      "epoch": 1.6128101557991923,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047500035092704843,
      "loss": 5.9394,
      "mean_token_accuracy": 0.18122438639402388,
      "num_tokens": 35418181.0,
      "step": 13975
    },
    {
      "entropy": 6.227433824539185,
      "epoch": 1.6133871898442007,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004749814435462159,
      "loss": 5.937,
      "mean_token_accuracy": 0.17713406085968017,
      "num_tokens": 35432495.0,
      "step": 13980
    },
    {
      "entropy": 6.177450037002563,
      "epoch": 1.6139642238892096,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004749625294391069,
      "loss": 5.9578,
      "mean_token_accuracy": 0.17766643911600113,
      "num_tokens": 35446327.0,
      "step": 13985
    },
    {
      "entropy": 6.1295312404632565,
      "epoch": 1.614541257934218,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047494360860635755,
      "loss": 5.9939,
      "mean_token_accuracy": 0.17990471720695494,
      "num_tokens": 35459964.0,
      "step": 13990
    },
    {
      "entropy": 6.1378742218017575,
      "epoch": 1.6151182919792268,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004749246810486042,
      "loss": 5.9441,
      "mean_token_accuracy": 0.17341957092285157,
      "num_tokens": 35471704.0,
      "step": 13995
    },
    {
      "entropy": 6.201165771484375,
      "epoch": 1.6156953260242353,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004749057467664836,
      "loss": 6.0392,
      "mean_token_accuracy": 0.16977255195379257,
      "num_tokens": 35483772.0,
      "step": 14000
    },
    {
      "entropy": 6.219329500198365,
      "epoch": 1.6162723600692441,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047488680576063247,
      "loss": 5.9674,
      "mean_token_accuracy": 0.18306666761636733,
      "num_tokens": 35497114.0,
      "step": 14005
    },
    {
      "entropy": 6.182672452926636,
      "epoch": 1.6168493941142528,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004748678580316878,
      "loss": 6.0344,
      "mean_token_accuracy": 0.1639961078763008,
      "num_tokens": 35509988.0,
      "step": 14010
    },
    {
      "entropy": 6.203671598434449,
      "epoch": 1.6174264281592614,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00047484890358028714,
      "loss": 6.0005,
      "mean_token_accuracy": 0.1811726301908493,
      "num_tokens": 35522613.0,
      "step": 14015
    },
    {
      "entropy": 6.225028419494629,
      "epoch": 1.61800346220427,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004748299424070678,
      "loss": 5.9717,
      "mean_token_accuracy": 0.17378117144107819,
      "num_tokens": 35534618.0,
      "step": 14020
    },
    {
      "entropy": 6.1102910995483395,
      "epoch": 1.6185804962492787,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004748109745126677,
      "loss": 5.9805,
      "mean_token_accuracy": 0.17011769562959672,
      "num_tokens": 35547461.0,
      "step": 14025
    },
    {
      "entropy": 6.215476846694946,
      "epoch": 1.6191575302942873,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047479199989772464,
      "loss": 5.9613,
      "mean_token_accuracy": 0.17455623298883438,
      "num_tokens": 35560026.0,
      "step": 14030
    },
    {
      "entropy": 6.2470067024230955,
      "epoch": 1.619734564339296,
      "grad_norm": 1.125,
      "learning_rate": 0.000474773018562877,
      "loss": 5.9353,
      "mean_token_accuracy": 0.17226501256227494,
      "num_tokens": 35571700.0,
      "step": 14035
    },
    {
      "entropy": 6.1181329727172855,
      "epoch": 1.6203115983843048,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047475403050876297,
      "loss": 6.0229,
      "mean_token_accuracy": 0.16893351823091507,
      "num_tokens": 35585366.0,
      "step": 14040
    },
    {
      "entropy": 6.275210952758789,
      "epoch": 1.6208886324293132,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004747350357360214,
      "loss": 6.0651,
      "mean_token_accuracy": 0.16842600256204604,
      "num_tokens": 35597809.0,
      "step": 14045
    },
    {
      "entropy": 6.232709646224976,
      "epoch": 1.621465666474322,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004747160342452912,
      "loss": 6.0185,
      "mean_token_accuracy": 0.16945921182632445,
      "num_tokens": 35610410.0,
      "step": 14050
    },
    {
      "entropy": 6.15349268913269,
      "epoch": 1.6220427005193305,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047469702603721134,
      "loss": 5.9905,
      "mean_token_accuracy": 0.1757126569747925,
      "num_tokens": 35624247.0,
      "step": 14055
    },
    {
      "entropy": 6.157156324386596,
      "epoch": 1.6226197345643394,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004746780111124212,
      "loss": 5.9136,
      "mean_token_accuracy": 0.17381737977266312,
      "num_tokens": 35636658.0,
      "step": 14060
    },
    {
      "entropy": 6.306695508956909,
      "epoch": 1.6231967686093478,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047465898947156033,
      "loss": 6.0599,
      "mean_token_accuracy": 0.16951913982629777,
      "num_tokens": 35649179.0,
      "step": 14065
    },
    {
      "entropy": 6.188014650344849,
      "epoch": 1.6237738026543567,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047463996111526854,
      "loss": 6.0151,
      "mean_token_accuracy": 0.1704736664891243,
      "num_tokens": 35661709.0,
      "step": 14070
    },
    {
      "entropy": 6.198279190063476,
      "epoch": 1.6243508366993653,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047462092604418576,
      "loss": 5.914,
      "mean_token_accuracy": 0.1795765846967697,
      "num_tokens": 35673337.0,
      "step": 14075
    },
    {
      "entropy": 6.218537473678589,
      "epoch": 1.624927870744374,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047460188425895236,
      "loss": 6.1273,
      "mean_token_accuracy": 0.16802418380975723,
      "num_tokens": 35688726.0,
      "step": 14080
    },
    {
      "entropy": 6.2344849586486815,
      "epoch": 1.6255049047893826,
      "grad_norm": 1.5625,
      "learning_rate": 0.00047458283576020874,
      "loss": 6.0636,
      "mean_token_accuracy": 0.16212450563907624,
      "num_tokens": 35702514.0,
      "step": 14085
    },
    {
      "entropy": 6.2704758644104,
      "epoch": 1.6260819388343912,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047456378054859554,
      "loss": 6.0777,
      "mean_token_accuracy": 0.16370144933462144,
      "num_tokens": 35715153.0,
      "step": 14090
    },
    {
      "entropy": 6.183321905136109,
      "epoch": 1.6266589728793999,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047454471862475375,
      "loss": 6.0609,
      "mean_token_accuracy": 0.17756673991680144,
      "num_tokens": 35725893.0,
      "step": 14095
    },
    {
      "entropy": 6.122617483139038,
      "epoch": 1.6272360069244085,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047452564998932446,
      "loss": 5.9024,
      "mean_token_accuracy": 0.18277804404497147,
      "num_tokens": 35738308.0,
      "step": 14100
    },
    {
      "entropy": 6.238106966018677,
      "epoch": 1.6278130409694171,
      "grad_norm": 1.078125,
      "learning_rate": 0.000474506574642949,
      "loss": 6.0482,
      "mean_token_accuracy": 0.1667554870247841,
      "num_tokens": 35750538.0,
      "step": 14105
    },
    {
      "entropy": 6.267595911026001,
      "epoch": 1.6283900750144258,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000474487492586269,
      "loss": 6.079,
      "mean_token_accuracy": 0.1659761905670166,
      "num_tokens": 35763974.0,
      "step": 14110
    },
    {
      "entropy": 6.168857479095459,
      "epoch": 1.6289671090594346,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004744684038199263,
      "loss": 5.965,
      "mean_token_accuracy": 0.18053883463144302,
      "num_tokens": 35775967.0,
      "step": 14115
    },
    {
      "entropy": 6.20915207862854,
      "epoch": 1.629544143104443,
      "grad_norm": 1.125,
      "learning_rate": 0.00047444930834456293,
      "loss": 5.9152,
      "mean_token_accuracy": 0.17555994540452957,
      "num_tokens": 35788231.0,
      "step": 14120
    },
    {
      "entropy": 6.218688297271728,
      "epoch": 1.630121177149452,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004744302061608212,
      "loss": 6.0266,
      "mean_token_accuracy": 0.1702222391963005,
      "num_tokens": 35801752.0,
      "step": 14125
    },
    {
      "entropy": 6.190368413925171,
      "epoch": 1.6306982111944603,
      "grad_norm": 1.0,
      "learning_rate": 0.00047441109726934345,
      "loss": 5.9871,
      "mean_token_accuracy": 0.16974416077136995,
      "num_tokens": 35814849.0,
      "step": 14130
    },
    {
      "entropy": 6.179901075363159,
      "epoch": 1.6312752452394692,
      "grad_norm": 1.125,
      "learning_rate": 0.00047439198167077256,
      "loss": 5.9815,
      "mean_token_accuracy": 0.17249080240726472,
      "num_tokens": 35827839.0,
      "step": 14135
    },
    {
      "entropy": 6.163280200958252,
      "epoch": 1.6318522792844776,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004743728593657514,
      "loss": 5.9407,
      "mean_token_accuracy": 0.17850152999162675,
      "num_tokens": 35839683.0,
      "step": 14140
    },
    {
      "entropy": 6.158884811401367,
      "epoch": 1.6324293133294865,
      "grad_norm": 1.25,
      "learning_rate": 0.0004743537303549231,
      "loss": 5.9229,
      "mean_token_accuracy": 0.18032356947660447,
      "num_tokens": 35853523.0,
      "step": 14145
    },
    {
      "entropy": 6.155524110794067,
      "epoch": 1.6330063473744951,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004743345946389311,
      "loss": 6.0173,
      "mean_token_accuracy": 0.17117015272378922,
      "num_tokens": 35865189.0,
      "step": 14150
    },
    {
      "entropy": 6.2249363422393795,
      "epoch": 1.6335833814195038,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047431545221841907,
      "loss": 6.0221,
      "mean_token_accuracy": 0.17346051037311555,
      "num_tokens": 35877352.0,
      "step": 14155
    },
    {
      "entropy": 6.216606521606446,
      "epoch": 1.6341604154645124,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047429630309403086,
      "loss": 6.014,
      "mean_token_accuracy": 0.1640690267086029,
      "num_tokens": 35890030.0,
      "step": 14160
    },
    {
      "entropy": 6.190902853012085,
      "epoch": 1.634737449509521,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047427714726641044,
      "loss": 5.9066,
      "mean_token_accuracy": 0.17736952155828475,
      "num_tokens": 35901822.0,
      "step": 14165
    },
    {
      "entropy": 6.1285459995269775,
      "epoch": 1.6353144835545297,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047425798473620215,
      "loss": 5.9568,
      "mean_token_accuracy": 0.17292165160179138,
      "num_tokens": 35914553.0,
      "step": 14170
    },
    {
      "entropy": 6.191268396377564,
      "epoch": 1.6358915175995383,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004742388155040505,
      "loss": 6.0447,
      "mean_token_accuracy": 0.17499910145998002,
      "num_tokens": 35928536.0,
      "step": 14175
    },
    {
      "entropy": 6.25783405303955,
      "epoch": 1.6364685516445472,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047421963957060026,
      "loss": 6.0363,
      "mean_token_accuracy": 0.16667375415563584,
      "num_tokens": 35941860.0,
      "step": 14180
    },
    {
      "entropy": 6.1880651950836185,
      "epoch": 1.6370455856895556,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004742004569364964,
      "loss": 6.0194,
      "mean_token_accuracy": 0.17363101541996,
      "num_tokens": 35957126.0,
      "step": 14185
    },
    {
      "entropy": 6.205398941040039,
      "epoch": 1.6376226197345645,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047418126760238416,
      "loss": 5.9355,
      "mean_token_accuracy": 0.17737175077199935,
      "num_tokens": 35969871.0,
      "step": 14190
    },
    {
      "entropy": 6.2128662109375,
      "epoch": 1.6381996537795729,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047416207156890887,
      "loss": 6.1297,
      "mean_token_accuracy": 0.16441965103149414,
      "num_tokens": 35983341.0,
      "step": 14195
    },
    {
      "entropy": 6.218688583374023,
      "epoch": 1.6387766878245817,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004741428688367164,
      "loss": 5.9772,
      "mean_token_accuracy": 0.17187047749757767,
      "num_tokens": 35995306.0,
      "step": 14200
    },
    {
      "entropy": 6.1825262069702145,
      "epoch": 1.6393537218695902,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047412365940645225,
      "loss": 5.9889,
      "mean_token_accuracy": 0.17393578141927718,
      "num_tokens": 36007424.0,
      "step": 14205
    },
    {
      "entropy": 6.192206621170044,
      "epoch": 1.639930755914599,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047410444327876286,
      "loss": 6.0419,
      "mean_token_accuracy": 0.17737023979425431,
      "num_tokens": 36020752.0,
      "step": 14210
    },
    {
      "entropy": 6.26286940574646,
      "epoch": 1.6405077899596077,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047408522045429426,
      "loss": 5.9983,
      "mean_token_accuracy": 0.17167010009288788,
      "num_tokens": 36031802.0,
      "step": 14215
    },
    {
      "entropy": 6.210866975784302,
      "epoch": 1.6410848240046163,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004740659909336933,
      "loss": 6.0716,
      "mean_token_accuracy": 0.16674784570932388,
      "num_tokens": 36044784.0,
      "step": 14220
    },
    {
      "entropy": 6.199074697494507,
      "epoch": 1.641661858049625,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047404675471760655,
      "loss": 5.963,
      "mean_token_accuracy": 0.17144973427057267,
      "num_tokens": 36057301.0,
      "step": 14225
    },
    {
      "entropy": 6.251472282409668,
      "epoch": 1.6422388920946336,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004740275118066811,
      "loss": 6.0088,
      "mean_token_accuracy": 0.16767774373292924,
      "num_tokens": 36070301.0,
      "step": 14230
    },
    {
      "entropy": 6.180870008468628,
      "epoch": 1.6428159261396422,
      "grad_norm": 1.125,
      "learning_rate": 0.00047400826220156416,
      "loss": 5.9847,
      "mean_token_accuracy": 0.1756862685084343,
      "num_tokens": 36083122.0,
      "step": 14235
    },
    {
      "entropy": 6.184627819061279,
      "epoch": 1.6433929601846509,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047398900590290313,
      "loss": 6.0728,
      "mean_token_accuracy": 0.16237133741378784,
      "num_tokens": 36095888.0,
      "step": 14240
    },
    {
      "entropy": 6.296611881256103,
      "epoch": 1.6439699942296595,
      "grad_norm": 1.375,
      "learning_rate": 0.00047396974291134575,
      "loss": 6.0508,
      "mean_token_accuracy": 0.1665154293179512,
      "num_tokens": 36108865.0,
      "step": 14245
    },
    {
      "entropy": 6.273109245300293,
      "epoch": 1.6445470282746681,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047395047322754,
      "loss": 6.0135,
      "mean_token_accuracy": 0.16950811445713043,
      "num_tokens": 36121318.0,
      "step": 14250
    },
    {
      "entropy": 6.260707759857178,
      "epoch": 1.645124062319677,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047393119685213374,
      "loss": 6.0843,
      "mean_token_accuracy": 0.16530290395021438,
      "num_tokens": 36134097.0,
      "step": 14255
    },
    {
      "entropy": 6.126479148864746,
      "epoch": 1.6457010963646854,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004739119137857756,
      "loss": 5.966,
      "mean_token_accuracy": 0.1744974821805954,
      "num_tokens": 36146382.0,
      "step": 14260
    },
    {
      "entropy": 6.288453054428101,
      "epoch": 1.6462781304096943,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047389262402911404,
      "loss": 6.0372,
      "mean_token_accuracy": 0.16469367891550063,
      "num_tokens": 36159323.0,
      "step": 14265
    },
    {
      "entropy": 6.246522855758667,
      "epoch": 1.6468551644547027,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047387332758279784,
      "loss": 5.9836,
      "mean_token_accuracy": 0.17046671211719513,
      "num_tokens": 36171545.0,
      "step": 14270
    },
    {
      "entropy": 6.169871234893799,
      "epoch": 1.6474321984997116,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047385402444747606,
      "loss": 5.9681,
      "mean_token_accuracy": 0.1721335768699646,
      "num_tokens": 36184175.0,
      "step": 14275
    },
    {
      "entropy": 6.172840404510498,
      "epoch": 1.64800923254472,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047383471462379803,
      "loss": 5.9272,
      "mean_token_accuracy": 0.1770294949412346,
      "num_tokens": 36196444.0,
      "step": 14280
    },
    {
      "entropy": 6.168339014053345,
      "epoch": 1.6485862665897288,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000473815398112413,
      "loss": 5.9565,
      "mean_token_accuracy": 0.17079036831855773,
      "num_tokens": 36208899.0,
      "step": 14285
    },
    {
      "entropy": 6.194360208511353,
      "epoch": 1.6491633006347375,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004737960749139708,
      "loss": 6.0127,
      "mean_token_accuracy": 0.17395621836185454,
      "num_tokens": 36223039.0,
      "step": 14290
    },
    {
      "entropy": 6.253053760528564,
      "epoch": 1.6497403346797461,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004737767450291215,
      "loss": 6.0861,
      "mean_token_accuracy": 0.1641833260655403,
      "num_tokens": 36234917.0,
      "step": 14295
    },
    {
      "entropy": 6.087620258331299,
      "epoch": 1.6503173687247548,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047375740845851506,
      "loss": 5.932,
      "mean_token_accuracy": 0.1802587553858757,
      "num_tokens": 36246791.0,
      "step": 14300
    },
    {
      "entropy": 6.263202333450318,
      "epoch": 1.6508944027697634,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004737380652028019,
      "loss": 6.064,
      "mean_token_accuracy": 0.1652047798037529,
      "num_tokens": 36259666.0,
      "step": 14305
    },
    {
      "entropy": 6.163006353378296,
      "epoch": 1.651471436814772,
      "grad_norm": 1.578125,
      "learning_rate": 0.00047371871526263263,
      "loss": 5.9392,
      "mean_token_accuracy": 0.1741989493370056,
      "num_tokens": 36272208.0,
      "step": 14310
    },
    {
      "entropy": 6.196961736679077,
      "epoch": 1.6520484708597807,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004736993586386581,
      "loss": 6.001,
      "mean_token_accuracy": 0.17262399047613144,
      "num_tokens": 36285108.0,
      "step": 14315
    },
    {
      "entropy": 6.171053171157837,
      "epoch": 1.6526255049047895,
      "grad_norm": 1.125,
      "learning_rate": 0.00047367999533152934,
      "loss": 6.0504,
      "mean_token_accuracy": 0.17122725397348404,
      "num_tokens": 36297228.0,
      "step": 14320
    },
    {
      "entropy": 6.215073585510254,
      "epoch": 1.653202538949798,
      "grad_norm": 1.5,
      "learning_rate": 0.00047366062534189756,
      "loss": 6.0138,
      "mean_token_accuracy": 0.16877247989177704,
      "num_tokens": 36310291.0,
      "step": 14325
    },
    {
      "entropy": 6.182399272918701,
      "epoch": 1.6537795729948068,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047364124867041446,
      "loss": 5.9492,
      "mean_token_accuracy": 0.17518317103385925,
      "num_tokens": 36322932.0,
      "step": 14330
    },
    {
      "entropy": 6.172776556015014,
      "epoch": 1.6543566070398152,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047362186531773156,
      "loss": 6.0128,
      "mean_token_accuracy": 0.16843672841787338,
      "num_tokens": 36334896.0,
      "step": 14335
    },
    {
      "entropy": 6.232485723495484,
      "epoch": 1.654933641084824,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004736024752845008,
      "loss": 6.0213,
      "mean_token_accuracy": 0.17199326455593109,
      "num_tokens": 36347866.0,
      "step": 14340
    },
    {
      "entropy": 6.139515066146851,
      "epoch": 1.6555106751298325,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004735830785713746,
      "loss": 5.995,
      "mean_token_accuracy": 0.17099547684192656,
      "num_tokens": 36360930.0,
      "step": 14345
    },
    {
      "entropy": 6.205392551422119,
      "epoch": 1.6560877091748414,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004735636751790051,
      "loss": 5.9755,
      "mean_token_accuracy": 0.17642167806625367,
      "num_tokens": 36373798.0,
      "step": 14350
    },
    {
      "entropy": 6.117266082763672,
      "epoch": 1.65666474321985,
      "grad_norm": 1.15625,
      "learning_rate": 0.000473544265108045,
      "loss": 5.9795,
      "mean_token_accuracy": 0.17760858237743377,
      "num_tokens": 36387654.0,
      "step": 14355
    },
    {
      "entropy": 6.146162366867065,
      "epoch": 1.6572417772648587,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047352484835914716,
      "loss": 6.0151,
      "mean_token_accuracy": 0.1729048788547516,
      "num_tokens": 36399963.0,
      "step": 14360
    },
    {
      "entropy": 6.284958457946777,
      "epoch": 1.6578188113098673,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004735054249329647,
      "loss": 6.097,
      "mean_token_accuracy": 0.16359793841838838,
      "num_tokens": 36412397.0,
      "step": 14365
    },
    {
      "entropy": 6.2170695781707765,
      "epoch": 1.658395845354876,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047348599483015087,
      "loss": 5.9616,
      "mean_token_accuracy": 0.17074348777532578,
      "num_tokens": 36425180.0,
      "step": 14370
    },
    {
      "entropy": 6.156053113937378,
      "epoch": 1.6589728793998846,
      "grad_norm": 1.125,
      "learning_rate": 0.00047346655805135916,
      "loss": 5.9944,
      "mean_token_accuracy": 0.1827524021267891,
      "num_tokens": 36437447.0,
      "step": 14375
    },
    {
      "entropy": 6.207530832290649,
      "epoch": 1.6595499134448932,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004734471145972434,
      "loss": 5.9638,
      "mean_token_accuracy": 0.17235314399003981,
      "num_tokens": 36449778.0,
      "step": 14380
    },
    {
      "entropy": 6.072767066955566,
      "epoch": 1.6601269474899019,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047342766446845753,
      "loss": 5.8999,
      "mean_token_accuracy": 0.18535937517881393,
      "num_tokens": 36461759.0,
      "step": 14385
    },
    {
      "entropy": 6.237001991271972,
      "epoch": 1.6607039815349105,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004734082076656557,
      "loss": 6.0597,
      "mean_token_accuracy": 0.16153217405080794,
      "num_tokens": 36474613.0,
      "step": 14390
    },
    {
      "entropy": 6.307389640808106,
      "epoch": 1.6612810155799194,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047338874418949235,
      "loss": 6.0366,
      "mean_token_accuracy": 0.16603167057037355,
      "num_tokens": 36486569.0,
      "step": 14395
    },
    {
      "entropy": 6.231861972808838,
      "epoch": 1.6618580496249278,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047336927404062213,
      "loss": 6.0664,
      "mean_token_accuracy": 0.16800082325935364,
      "num_tokens": 36500144.0,
      "step": 14400
    },
    {
      "entropy": 6.178709411621094,
      "epoch": 1.6624350836699366,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047334979721969995,
      "loss": 6.0193,
      "mean_token_accuracy": 0.1640546903014183,
      "num_tokens": 36512665.0,
      "step": 14405
    },
    {
      "entropy": 6.169596242904663,
      "epoch": 1.663012117714945,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004733303137273808,
      "loss": 5.9651,
      "mean_token_accuracy": 0.17677154392004013,
      "num_tokens": 36525876.0,
      "step": 14410
    },
    {
      "entropy": 6.179593181610107,
      "epoch": 1.663589151759954,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047331082356432015,
      "loss": 5.9214,
      "mean_token_accuracy": 0.17804815024137496,
      "num_tokens": 36538253.0,
      "step": 14415
    },
    {
      "entropy": 6.135897254943847,
      "epoch": 1.6641661858049623,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004732913267311734,
      "loss": 6.0013,
      "mean_token_accuracy": 0.1763201966881752,
      "num_tokens": 36550303.0,
      "step": 14420
    },
    {
      "entropy": 6.142184925079346,
      "epoch": 1.6647432198499712,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004732718232285964,
      "loss": 5.9174,
      "mean_token_accuracy": 0.17775216698646545,
      "num_tokens": 36564835.0,
      "step": 14425
    },
    {
      "entropy": 6.293658065795898,
      "epoch": 1.6653202538949798,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047325231305724507,
      "loss": 6.0542,
      "mean_token_accuracy": 0.16572537571191787,
      "num_tokens": 36577648.0,
      "step": 14430
    },
    {
      "entropy": 6.143080520629883,
      "epoch": 1.6658972879399885,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004732327962177757,
      "loss": 6.0065,
      "mean_token_accuracy": 0.1656201735138893,
      "num_tokens": 36590451.0,
      "step": 14435
    },
    {
      "entropy": 6.249016666412354,
      "epoch": 1.6664743219849971,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004732132727108447,
      "loss": 6.0086,
      "mean_token_accuracy": 0.1643291711807251,
      "num_tokens": 36602767.0,
      "step": 14440
    },
    {
      "entropy": 6.200948715209961,
      "epoch": 1.6670513560300058,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00047319374253710874,
      "loss": 5.9674,
      "mean_token_accuracy": 0.17258659601211548,
      "num_tokens": 36614938.0,
      "step": 14445
    },
    {
      "entropy": 6.135825490951538,
      "epoch": 1.6676283900750144,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004731742056972247,
      "loss": 5.9129,
      "mean_token_accuracy": 0.18108502328395842,
      "num_tokens": 36626880.0,
      "step": 14450
    },
    {
      "entropy": 6.0941671371459964,
      "epoch": 1.668205424120023,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004731546621918497,
      "loss": 6.0048,
      "mean_token_accuracy": 0.1749402955174446,
      "num_tokens": 36640729.0,
      "step": 14455
    },
    {
      "entropy": 6.1896399974823,
      "epoch": 1.668782458165032,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000473135112021641,
      "loss": 5.9795,
      "mean_token_accuracy": 0.18192052841186523,
      "num_tokens": 36656359.0,
      "step": 14460
    },
    {
      "entropy": 6.253931474685669,
      "epoch": 1.6693594922100403,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047311555518725617,
      "loss": 6.0093,
      "mean_token_accuracy": 0.17626387625932693,
      "num_tokens": 36667985.0,
      "step": 14465
    },
    {
      "entropy": 6.254989910125732,
      "epoch": 1.6699365262550492,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047309599168935323,
      "loss": 6.0841,
      "mean_token_accuracy": 0.16237382888793944,
      "num_tokens": 36681477.0,
      "step": 14470
    },
    {
      "entropy": 6.217514991760254,
      "epoch": 1.6705135603000576,
      "grad_norm": 1.125,
      "learning_rate": 0.00047307642152858993,
      "loss": 5.9697,
      "mean_token_accuracy": 0.17538907527923583,
      "num_tokens": 36695532.0,
      "step": 14475
    },
    {
      "entropy": 6.237057113647461,
      "epoch": 1.6710905943450665,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047305684470562453,
      "loss": 5.9919,
      "mean_token_accuracy": 0.17043597251176834,
      "num_tokens": 36708914.0,
      "step": 14480
    },
    {
      "entropy": 6.220160579681396,
      "epoch": 1.6716676283900749,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004730372612211156,
      "loss": 6.0678,
      "mean_token_accuracy": 0.16463319212198257,
      "num_tokens": 36720849.0,
      "step": 14485
    },
    {
      "entropy": 6.137443113327026,
      "epoch": 1.6722446624350837,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047301767107572174,
      "loss": 5.963,
      "mean_token_accuracy": 0.17474720180034636,
      "num_tokens": 36733786.0,
      "step": 14490
    },
    {
      "entropy": 6.197306680679321,
      "epoch": 1.6728216964800924,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004729980742701018,
      "loss": 6.0055,
      "mean_token_accuracy": 0.1708031564950943,
      "num_tokens": 36746101.0,
      "step": 14495
    },
    {
      "entropy": 6.181519412994385,
      "epoch": 1.673398730525101,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004729784708049151,
      "loss": 5.9216,
      "mean_token_accuracy": 0.16888677328824997,
      "num_tokens": 36758644.0,
      "step": 14500
    },
    {
      "entropy": 6.249948453903198,
      "epoch": 1.6739757645701097,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004729588606808209,
      "loss": 6.087,
      "mean_token_accuracy": 0.1653010293841362,
      "num_tokens": 36770301.0,
      "step": 14505
    },
    {
      "entropy": 6.198924446105957,
      "epoch": 1.6745527986151183,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047293924389847864,
      "loss": 5.9519,
      "mean_token_accuracy": 0.17446827590465547,
      "num_tokens": 36782337.0,
      "step": 14510
    },
    {
      "entropy": 6.207420587539673,
      "epoch": 1.675129832660127,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004729196204585483,
      "loss": 5.9959,
      "mean_token_accuracy": 0.17194482684135437,
      "num_tokens": 36793817.0,
      "step": 14515
    },
    {
      "entropy": 6.177169704437256,
      "epoch": 1.6757068667051356,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047289999036168983,
      "loss": 6.039,
      "mean_token_accuracy": 0.16699618101119995,
      "num_tokens": 36806577.0,
      "step": 14520
    },
    {
      "entropy": 6.210075759887696,
      "epoch": 1.6762839007501442,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004728803536085634,
      "loss": 6.0028,
      "mean_token_accuracy": 0.17215853780508042,
      "num_tokens": 36819896.0,
      "step": 14525
    },
    {
      "entropy": 6.1746727466583256,
      "epoch": 1.6768609347951529,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047286071019982974,
      "loss": 5.9904,
      "mean_token_accuracy": 0.17391820698976518,
      "num_tokens": 36831995.0,
      "step": 14530
    },
    {
      "entropy": 6.2465407371521,
      "epoch": 1.6774379688401617,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047284106013614926,
      "loss": 5.9573,
      "mean_token_accuracy": 0.16565890461206437,
      "num_tokens": 36844452.0,
      "step": 14535
    },
    {
      "entropy": 6.209101057052612,
      "epoch": 1.6780150028851701,
      "grad_norm": 1.6953125,
      "learning_rate": 0.000472821403418183,
      "loss": 5.9756,
      "mean_token_accuracy": 0.17331359833478927,
      "num_tokens": 36857578.0,
      "step": 14540
    },
    {
      "entropy": 6.175076818466186,
      "epoch": 1.678592036930179,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004728017400465921,
      "loss": 6.0304,
      "mean_token_accuracy": 0.16885432004928588,
      "num_tokens": 36870406.0,
      "step": 14545
    },
    {
      "entropy": 6.130092000961303,
      "epoch": 1.6791690709751874,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047278207002203796,
      "loss": 5.9245,
      "mean_token_accuracy": 0.17331396490335466,
      "num_tokens": 36881149.0,
      "step": 14550
    },
    {
      "entropy": 6.1751199722290036,
      "epoch": 1.6797461050201963,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047276239334518216,
      "loss": 5.9379,
      "mean_token_accuracy": 0.17374935895204544,
      "num_tokens": 36893175.0,
      "step": 14555
    },
    {
      "entropy": 6.142501306533814,
      "epoch": 1.6803231390652047,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047274271001668646,
      "loss": 5.9181,
      "mean_token_accuracy": 0.17434564381837844,
      "num_tokens": 36904993.0,
      "step": 14560
    },
    {
      "entropy": 6.180646562576294,
      "epoch": 1.6809001731102136,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047272302003721286,
      "loss": 6.0333,
      "mean_token_accuracy": 0.1675330862402916,
      "num_tokens": 36918597.0,
      "step": 14565
    },
    {
      "entropy": 6.212793445587158,
      "epoch": 1.6814772071552222,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047270332340742377,
      "loss": 6.0112,
      "mean_token_accuracy": 0.16955852657556533,
      "num_tokens": 36931404.0,
      "step": 14570
    },
    {
      "entropy": 6.227953624725342,
      "epoch": 1.6820542412002308,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047268362012798157,
      "loss": 6.0562,
      "mean_token_accuracy": 0.16889184713363647,
      "num_tokens": 36942657.0,
      "step": 14575
    },
    {
      "entropy": 6.193990039825439,
      "epoch": 1.6826312752452395,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004726639101995491,
      "loss": 6.0049,
      "mean_token_accuracy": 0.17326837331056594,
      "num_tokens": 36956130.0,
      "step": 14580
    },
    {
      "entropy": 6.190242528915405,
      "epoch": 1.6832083092902481,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00047264419362278906,
      "loss": 6.0364,
      "mean_token_accuracy": 0.1697862759232521,
      "num_tokens": 36968865.0,
      "step": 14585
    },
    {
      "entropy": 6.251032209396362,
      "epoch": 1.6837853433352568,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047262447039836484,
      "loss": 6.0575,
      "mean_token_accuracy": 0.17273726612329482,
      "num_tokens": 36980466.0,
      "step": 14590
    },
    {
      "entropy": 6.191921329498291,
      "epoch": 1.6843623773802654,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047260474052693963,
      "loss": 5.9968,
      "mean_token_accuracy": 0.16677181795239449,
      "num_tokens": 36992263.0,
      "step": 14595
    },
    {
      "entropy": 6.230498313903809,
      "epoch": 1.684939411425274,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047258500400917724,
      "loss": 6.0293,
      "mean_token_accuracy": 0.1694782108068466,
      "num_tokens": 37003422.0,
      "step": 14600
    },
    {
      "entropy": 6.182830190658569,
      "epoch": 1.6855164454702827,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047256526084574137,
      "loss": 5.9325,
      "mean_token_accuracy": 0.17168499678373336,
      "num_tokens": 37015064.0,
      "step": 14605
    },
    {
      "entropy": 6.1292472839355465,
      "epoch": 1.6860934795152915,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047254551103729597,
      "loss": 5.8957,
      "mean_token_accuracy": 0.17820012718439102,
      "num_tokens": 37027750.0,
      "step": 14610
    },
    {
      "entropy": 6.176606321334839,
      "epoch": 1.6866705135603,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004725257545845055,
      "loss": 5.9895,
      "mean_token_accuracy": 0.16939300894737244,
      "num_tokens": 37039501.0,
      "step": 14615
    },
    {
      "entropy": 6.18608136177063,
      "epoch": 1.6872475476053088,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047250599148803443,
      "loss": 6.0366,
      "mean_token_accuracy": 0.16540022939443588,
      "num_tokens": 37051179.0,
      "step": 14620
    },
    {
      "entropy": 6.157530164718628,
      "epoch": 1.6878245816503172,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047248622174854746,
      "loss": 5.9878,
      "mean_token_accuracy": 0.17260215878486634,
      "num_tokens": 37064098.0,
      "step": 14625
    },
    {
      "entropy": 6.220685195922852,
      "epoch": 1.688401615695326,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004724664453667094,
      "loss": 6.0498,
      "mean_token_accuracy": 0.1647424504160881,
      "num_tokens": 37076315.0,
      "step": 14630
    },
    {
      "entropy": 6.247399950027466,
      "epoch": 1.6889786497403345,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004724466623431857,
      "loss": 6.0876,
      "mean_token_accuracy": 0.17160662710666658,
      "num_tokens": 37088354.0,
      "step": 14635
    },
    {
      "entropy": 6.210821914672851,
      "epoch": 1.6895556837853434,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004724268726786415,
      "loss": 6.0055,
      "mean_token_accuracy": 0.17390506565570832,
      "num_tokens": 37101296.0,
      "step": 14640
    },
    {
      "entropy": 6.227218341827393,
      "epoch": 1.690132717830352,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004724070763737424,
      "loss": 6.0537,
      "mean_token_accuracy": 0.1699638247489929,
      "num_tokens": 37114095.0,
      "step": 14645
    },
    {
      "entropy": 6.1917479038238525,
      "epoch": 1.6907097518753607,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004723872734291545,
      "loss": 5.9346,
      "mean_token_accuracy": 0.18094567209482193,
      "num_tokens": 37125963.0,
      "step": 14650
    },
    {
      "entropy": 6.196824836730957,
      "epoch": 1.6912867859203693,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047236746384554364,
      "loss": 6.0509,
      "mean_token_accuracy": 0.1704024627804756,
      "num_tokens": 37139210.0,
      "step": 14655
    },
    {
      "entropy": 6.091272068023682,
      "epoch": 1.691863819965378,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004723476476235762,
      "loss": 5.8794,
      "mean_token_accuracy": 0.18381695002317427,
      "num_tokens": 37151592.0,
      "step": 14660
    },
    {
      "entropy": 6.172664785385132,
      "epoch": 1.6924408540103866,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004723278247639186,
      "loss": 5.9941,
      "mean_token_accuracy": 0.17310368716716767,
      "num_tokens": 37164722.0,
      "step": 14665
    },
    {
      "entropy": 6.125967741012573,
      "epoch": 1.6930178880553952,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004723079952672377,
      "loss": 5.9133,
      "mean_token_accuracy": 0.18194041550159454,
      "num_tokens": 37177188.0,
      "step": 14670
    },
    {
      "entropy": 6.17428822517395,
      "epoch": 1.693594922100404,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047228815913420035,
      "loss": 5.9824,
      "mean_token_accuracy": 0.18060059249401092,
      "num_tokens": 37189702.0,
      "step": 14675
    },
    {
      "entropy": 6.192619276046753,
      "epoch": 1.6941719561454125,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004722683163654738,
      "loss": 5.9703,
      "mean_token_accuracy": 0.17456228137016297,
      "num_tokens": 37202327.0,
      "step": 14680
    },
    {
      "entropy": 6.133909845352173,
      "epoch": 1.6947489901904214,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004722484669617254,
      "loss": 5.9691,
      "mean_token_accuracy": 0.1698708176612854,
      "num_tokens": 37215557.0,
      "step": 14685
    },
    {
      "entropy": 6.163161373138427,
      "epoch": 1.6953260242354298,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047222861092362277,
      "loss": 5.9868,
      "mean_token_accuracy": 0.17944956123828887,
      "num_tokens": 37228863.0,
      "step": 14690
    },
    {
      "entropy": 6.152451753616333,
      "epoch": 1.6959030582804386,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047220874825183387,
      "loss": 6.053,
      "mean_token_accuracy": 0.1653454542160034,
      "num_tokens": 37241591.0,
      "step": 14695
    },
    {
      "entropy": 6.241807079315185,
      "epoch": 1.696480092325447,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047218887894702656,
      "loss": 6.0978,
      "mean_token_accuracy": 0.1691400408744812,
      "num_tokens": 37254440.0,
      "step": 14700
    },
    {
      "entropy": 6.200218772888183,
      "epoch": 1.697057126370456,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004721690030098693,
      "loss": 5.9561,
      "mean_token_accuracy": 0.1696375474333763,
      "num_tokens": 37266803.0,
      "step": 14705
    },
    {
      "entropy": 6.223460960388183,
      "epoch": 1.6976341604154646,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004721491204410305,
      "loss": 6.0859,
      "mean_token_accuracy": 0.16403590589761735,
      "num_tokens": 37280339.0,
      "step": 14710
    },
    {
      "entropy": 6.173096132278443,
      "epoch": 1.6982111944604732,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047212923124117915,
      "loss": 5.9152,
      "mean_token_accuracy": 0.17313134521245957,
      "num_tokens": 37293816.0,
      "step": 14715
    },
    {
      "entropy": 6.225034618377686,
      "epoch": 1.6987882285054818,
      "grad_norm": 1.125,
      "learning_rate": 0.0004721093354109839,
      "loss": 6.0009,
      "mean_token_accuracy": 0.16971071511507035,
      "num_tokens": 37306642.0,
      "step": 14720
    },
    {
      "entropy": 6.239164209365844,
      "epoch": 1.6993652625504905,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047208943295111406,
      "loss": 6.0137,
      "mean_token_accuracy": 0.17579566240310668,
      "num_tokens": 37320153.0,
      "step": 14725
    },
    {
      "entropy": 6.223319721221924,
      "epoch": 1.6999422965954991,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047206952386223905,
      "loss": 6.0968,
      "mean_token_accuracy": 0.1658716931939125,
      "num_tokens": 37333660.0,
      "step": 14730
    },
    {
      "entropy": 6.166567659378051,
      "epoch": 1.7005193306405078,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004720496081450285,
      "loss": 5.9713,
      "mean_token_accuracy": 0.17064529210329055,
      "num_tokens": 37346589.0,
      "step": 14735
    },
    {
      "entropy": 6.170449876785279,
      "epoch": 1.7010963646855164,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047202968580015224,
      "loss": 5.9415,
      "mean_token_accuracy": 0.1764495849609375,
      "num_tokens": 37359309.0,
      "step": 14740
    },
    {
      "entropy": 6.197453308105469,
      "epoch": 1.701673398730525,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047200975682828045,
      "loss": 6.0406,
      "mean_token_accuracy": 0.1724436789751053,
      "num_tokens": 37372547.0,
      "step": 14745
    },
    {
      "entropy": 6.125006818771363,
      "epoch": 1.702250432775534,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004719898212300832,
      "loss": 5.9859,
      "mean_token_accuracy": 0.17096855342388154,
      "num_tokens": 37384987.0,
      "step": 14750
    },
    {
      "entropy": 6.245584535598755,
      "epoch": 1.7028274668205423,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047196987900623134,
      "loss": 6.0115,
      "mean_token_accuracy": 0.16955127120018004,
      "num_tokens": 37397065.0,
      "step": 14755
    },
    {
      "entropy": 6.186125230789185,
      "epoch": 1.7034045008655512,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047194993015739527,
      "loss": 5.9603,
      "mean_token_accuracy": 0.172216759622097,
      "num_tokens": 37409467.0,
      "step": 14760
    },
    {
      "entropy": 6.185346984863282,
      "epoch": 1.7039815349105596,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047192997468424624,
      "loss": 5.9449,
      "mean_token_accuracy": 0.1823701336979866,
      "num_tokens": 37422268.0,
      "step": 14765
    },
    {
      "entropy": 6.149506998062134,
      "epoch": 1.7045585689555685,
      "grad_norm": 1.0,
      "learning_rate": 0.0004719100125874553,
      "loss": 5.9557,
      "mean_token_accuracy": 0.1771388977766037,
      "num_tokens": 37434637.0,
      "step": 14770
    },
    {
      "entropy": 6.119137573242187,
      "epoch": 1.7051356030005769,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004718900438676939,
      "loss": 5.9581,
      "mean_token_accuracy": 0.174500934779644,
      "num_tokens": 37447683.0,
      "step": 14775
    },
    {
      "entropy": 6.236277151107788,
      "epoch": 1.7057126370455857,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004718700685256337,
      "loss": 6.0487,
      "mean_token_accuracy": 0.16718055456876754,
      "num_tokens": 37459960.0,
      "step": 14780
    },
    {
      "entropy": 6.221787023544311,
      "epoch": 1.7062896710905944,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004718500865619465,
      "loss": 6.093,
      "mean_token_accuracy": 0.16780304312705993,
      "num_tokens": 37471262.0,
      "step": 14785
    },
    {
      "entropy": 6.110309410095215,
      "epoch": 1.706866705135603,
      "grad_norm": 1.125,
      "learning_rate": 0.0004718300979773044,
      "loss": 5.9203,
      "mean_token_accuracy": 0.17355912029743195,
      "num_tokens": 37484843.0,
      "step": 14790
    },
    {
      "entropy": 6.285755443572998,
      "epoch": 1.7074437391806117,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047181010277237977,
      "loss": 6.0215,
      "mean_token_accuracy": 0.16503700464963914,
      "num_tokens": 37497197.0,
      "step": 14795
    },
    {
      "entropy": 6.174851512908935,
      "epoch": 1.7080207732256203,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004717901009478451,
      "loss": 5.9313,
      "mean_token_accuracy": 0.1688641771674156,
      "num_tokens": 37508812.0,
      "step": 14800
    },
    {
      "entropy": 6.1470293521881105,
      "epoch": 1.708597807270629,
      "grad_norm": 1.125,
      "learning_rate": 0.00047177009250437313,
      "loss": 5.9586,
      "mean_token_accuracy": 0.17310403436422347,
      "num_tokens": 37520861.0,
      "step": 14805
    },
    {
      "entropy": 6.251069355010986,
      "epoch": 1.7091748413156376,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047175007744263683,
      "loss": 5.9998,
      "mean_token_accuracy": 0.1649973139166832,
      "num_tokens": 37533011.0,
      "step": 14810
    },
    {
      "entropy": 6.119192981719971,
      "epoch": 1.7097518753606464,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047173005576330935,
      "loss": 5.9441,
      "mean_token_accuracy": 0.17625575065612792,
      "num_tokens": 37546613.0,
      "step": 14815
    },
    {
      "entropy": 6.16118278503418,
      "epoch": 1.7103289094056549,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047171002746706424,
      "loss": 5.9002,
      "mean_token_accuracy": 0.17466118931770325,
      "num_tokens": 37559429.0,
      "step": 14820
    },
    {
      "entropy": 6.221500015258789,
      "epoch": 1.7109059434506637,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047168999255457506,
      "loss": 6.017,
      "mean_token_accuracy": 0.173841255903244,
      "num_tokens": 37571530.0,
      "step": 14825
    },
    {
      "entropy": 6.189529895782471,
      "epoch": 1.7114829774956721,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047166995102651565,
      "loss": 6.0432,
      "mean_token_accuracy": 0.16787166744470597,
      "num_tokens": 37584653.0,
      "step": 14830
    },
    {
      "entropy": 6.195649671554565,
      "epoch": 1.712060011540681,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004716499028835601,
      "loss": 6.0326,
      "mean_token_accuracy": 0.1693824365735054,
      "num_tokens": 37597361.0,
      "step": 14835
    },
    {
      "entropy": 6.213965988159179,
      "epoch": 1.7126370455856894,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004716298481263828,
      "loss": 6.0356,
      "mean_token_accuracy": 0.1728340968489647,
      "num_tokens": 37611525.0,
      "step": 14840
    },
    {
      "entropy": 6.2376471042633055,
      "epoch": 1.7132140796306983,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004716097867556583,
      "loss": 5.9796,
      "mean_token_accuracy": 0.17407371550798417,
      "num_tokens": 37624012.0,
      "step": 14845
    },
    {
      "entropy": 6.172725534439087,
      "epoch": 1.713791113675707,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047158971877206114,
      "loss": 6.0562,
      "mean_token_accuracy": 0.1694328248500824,
      "num_tokens": 37636877.0,
      "step": 14850
    },
    {
      "entropy": 6.2098681926727295,
      "epoch": 1.7143681477207156,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004715696441762665,
      "loss": 5.9833,
      "mean_token_accuracy": 0.17341266125440596,
      "num_tokens": 37649791.0,
      "step": 14855
    },
    {
      "entropy": 6.206290245056152,
      "epoch": 1.7149451817657242,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00047154956296894954,
      "loss": 5.9851,
      "mean_token_accuracy": 0.1722115233540535,
      "num_tokens": 37661699.0,
      "step": 14860
    },
    {
      "entropy": 6.157696485519409,
      "epoch": 1.7155222158107328,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004715294751507856,
      "loss": 6.0041,
      "mean_token_accuracy": 0.1655634269118309,
      "num_tokens": 37674498.0,
      "step": 14865
    },
    {
      "entropy": 6.284539747238159,
      "epoch": 1.7160992498557415,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004715093807224505,
      "loss": 5.9837,
      "mean_token_accuracy": 0.17345443814992906,
      "num_tokens": 37687196.0,
      "step": 14870
    },
    {
      "entropy": 6.139006853103638,
      "epoch": 1.71667628390075,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047148927968462,
      "loss": 5.8751,
      "mean_token_accuracy": 0.17929705679416658,
      "num_tokens": 37698888.0,
      "step": 14875
    },
    {
      "entropy": 6.109095811843872,
      "epoch": 1.7172533179457588,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047146917203797,
      "loss": 6.0011,
      "mean_token_accuracy": 0.17196657955646516,
      "num_tokens": 37711560.0,
      "step": 14880
    },
    {
      "entropy": 6.135826158523559,
      "epoch": 1.7178303519907674,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004714490577831771,
      "loss": 5.9135,
      "mean_token_accuracy": 0.1796274498105049,
      "num_tokens": 37723846.0,
      "step": 14885
    },
    {
      "entropy": 6.148435306549072,
      "epoch": 1.7184073860357763,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004714289369209177,
      "loss": 5.8807,
      "mean_token_accuracy": 0.17652539312839508,
      "num_tokens": 37735535.0,
      "step": 14890
    },
    {
      "entropy": 6.17125096321106,
      "epoch": 1.7189844200807847,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047140880945186863,
      "loss": 6.0086,
      "mean_token_accuracy": 0.1709775820374489,
      "num_tokens": 37748015.0,
      "step": 14895
    },
    {
      "entropy": 6.104178237915039,
      "epoch": 1.7195614541257935,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047138867537670676,
      "loss": 5.9444,
      "mean_token_accuracy": 0.17848032861948013,
      "num_tokens": 37760278.0,
      "step": 14900
    },
    {
      "entropy": 6.270360851287842,
      "epoch": 1.720138488170802,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047136853469610933,
      "loss": 6.0015,
      "mean_token_accuracy": 0.17952914386987687,
      "num_tokens": 37772274.0,
      "step": 14905
    },
    {
      "entropy": 6.214459609985352,
      "epoch": 1.7207155222158108,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047134838741075383,
      "loss": 5.9732,
      "mean_token_accuracy": 0.1751771464943886,
      "num_tokens": 37785919.0,
      "step": 14910
    },
    {
      "entropy": 6.1037694931030275,
      "epoch": 1.7212925562608192,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047132823352131787,
      "loss": 5.9339,
      "mean_token_accuracy": 0.17146482765674592,
      "num_tokens": 37797770.0,
      "step": 14915
    },
    {
      "entropy": 6.1130475997924805,
      "epoch": 1.721869590305828,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004713080730284793,
      "loss": 5.9481,
      "mean_token_accuracy": 0.17310337573289872,
      "num_tokens": 37809962.0,
      "step": 14920
    },
    {
      "entropy": 6.2526124000549315,
      "epoch": 1.7224466243508367,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047128790593291617,
      "loss": 5.9994,
      "mean_token_accuracy": 0.1671294689178467,
      "num_tokens": 37822345.0,
      "step": 14925
    },
    {
      "entropy": 6.138121795654297,
      "epoch": 1.7230236583958454,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047126773223530677,
      "loss": 5.9431,
      "mean_token_accuracy": 0.17680928111076355,
      "num_tokens": 37835568.0,
      "step": 14930
    },
    {
      "entropy": 6.201503133773803,
      "epoch": 1.723600692440854,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047124755193632975,
      "loss": 6.0517,
      "mean_token_accuracy": 0.17126816511154175,
      "num_tokens": 37849356.0,
      "step": 14935
    },
    {
      "entropy": 6.169628143310547,
      "epoch": 1.7241777264858626,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047122736503666377,
      "loss": 5.921,
      "mean_token_accuracy": 0.1799556404352188,
      "num_tokens": 37861970.0,
      "step": 14940
    },
    {
      "entropy": 6.2190672874450685,
      "epoch": 1.7247547605308713,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004712071715369878,
      "loss": 6.0388,
      "mean_token_accuracy": 0.16779190748929979,
      "num_tokens": 37874066.0,
      "step": 14945
    },
    {
      "entropy": 6.174004983901978,
      "epoch": 1.72533179457588,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004711869714379812,
      "loss": 5.9734,
      "mean_token_accuracy": 0.16788453310728074,
      "num_tokens": 37885529.0,
      "step": 14950
    },
    {
      "entropy": 6.262821245193481,
      "epoch": 1.7259088286208888,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004711667647403232,
      "loss": 6.0308,
      "mean_token_accuracy": 0.16671091616153716,
      "num_tokens": 37899517.0,
      "step": 14955
    },
    {
      "entropy": 6.144222211837769,
      "epoch": 1.7264858626658972,
      "grad_norm": 1.125,
      "learning_rate": 0.00047114655144469354,
      "loss": 5.9628,
      "mean_token_accuracy": 0.17398159503936766,
      "num_tokens": 37912586.0,
      "step": 14960
    },
    {
      "entropy": 6.210615587234497,
      "epoch": 1.727062896710906,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047112633155177203,
      "loss": 6.0085,
      "mean_token_accuracy": 0.17672025710344313,
      "num_tokens": 37924559.0,
      "step": 14965
    },
    {
      "entropy": 6.063754081726074,
      "epoch": 1.7276399307559145,
      "grad_norm": 1.5,
      "learning_rate": 0.0004711061050622388,
      "loss": 5.8824,
      "mean_token_accuracy": 0.18309446573257446,
      "num_tokens": 37937470.0,
      "step": 14970
    },
    {
      "entropy": 6.220537996292114,
      "epoch": 1.7282169648009233,
      "grad_norm": 1.53125,
      "learning_rate": 0.00047108587197677404,
      "loss": 5.9697,
      "mean_token_accuracy": 0.17395588904619216,
      "num_tokens": 37949502.0,
      "step": 14975
    },
    {
      "entropy": 6.201508903503418,
      "epoch": 1.7287939988459318,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047106563229605845,
      "loss": 6.0049,
      "mean_token_accuracy": 0.17538221329450607,
      "num_tokens": 37962298.0,
      "step": 14980
    },
    {
      "entropy": 6.133077192306518,
      "epoch": 1.7293710328909406,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047104538602077276,
      "loss": 5.9029,
      "mean_token_accuracy": 0.18140460401773453,
      "num_tokens": 37975614.0,
      "step": 14985
    },
    {
      "entropy": 6.117460155487061,
      "epoch": 1.7299480669359493,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004710251331515978,
      "loss": 5.9297,
      "mean_token_accuracy": 0.18087606728076935,
      "num_tokens": 37990117.0,
      "step": 14990
    },
    {
      "entropy": 6.173279285430908,
      "epoch": 1.730525100980958,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047100487368921485,
      "loss": 5.9803,
      "mean_token_accuracy": 0.1675584450364113,
      "num_tokens": 38003654.0,
      "step": 14995
    },
    {
      "entropy": 6.2653642177581785,
      "epoch": 1.7311021350259665,
      "grad_norm": 1.125,
      "learning_rate": 0.0004709846076343055,
      "loss": 6.0323,
      "mean_token_accuracy": 0.16838781535625458,
      "num_tokens": 38016299.0,
      "step": 15000
    },
    {
      "epoch": 1.7311021350259665,
      "eval_entropy": 6.015740028592311,
      "eval_loss": 6.092541694641113,
      "eval_mean_token_accuracy": 0.17235964559546105,
      "eval_num_tokens": 38016299.0,
      "eval_runtime": 22.9119,
      "eval_samples_per_second": 1228.006,
      "eval_steps_per_second": 153.501,
      "step": 15000
    },
    {
      "entropy": 6.1085117816925045,
      "epoch": 1.7316791690709752,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047096433498755103,
      "loss": 6.0068,
      "mean_token_accuracy": 0.1716531977057457,
      "num_tokens": 38027793.0,
      "step": 15005
    },
    {
      "entropy": 6.200501108169556,
      "epoch": 1.7322562031159838,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00047094405574963364,
      "loss": 5.9653,
      "mean_token_accuracy": 0.17174111008644105,
      "num_tokens": 38040167.0,
      "step": 15010
    },
    {
      "entropy": 6.204786014556885,
      "epoch": 1.7328332371609925,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047092376992123516,
      "loss": 5.9464,
      "mean_token_accuracy": 0.17328376471996307,
      "num_tokens": 38053334.0,
      "step": 15015
    },
    {
      "entropy": 6.1448546886444095,
      "epoch": 1.733410271206001,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047090347750303803,
      "loss": 5.9411,
      "mean_token_accuracy": 0.1776982307434082,
      "num_tokens": 38065831.0,
      "step": 15020
    },
    {
      "entropy": 6.134150981903076,
      "epoch": 1.7339873052510097,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004708831784957247,
      "loss": 6.0045,
      "mean_token_accuracy": 0.1737317144870758,
      "num_tokens": 38079298.0,
      "step": 15025
    },
    {
      "entropy": 6.268193197250366,
      "epoch": 1.7345643392960186,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004708628728999781,
      "loss": 6.0826,
      "mean_token_accuracy": 0.16167911887168884,
      "num_tokens": 38092473.0,
      "step": 15030
    },
    {
      "entropy": 6.199768304824829,
      "epoch": 1.735141373341027,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004708425607164809,
      "loss": 5.9486,
      "mean_token_accuracy": 0.17793972045183182,
      "num_tokens": 38106148.0,
      "step": 15035
    },
    {
      "entropy": 6.170068645477295,
      "epoch": 1.7357184073860359,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004708222419459165,
      "loss": 6.014,
      "mean_token_accuracy": 0.169931860268116,
      "num_tokens": 38119061.0,
      "step": 15040
    },
    {
      "entropy": 6.144618701934815,
      "epoch": 1.7362954414310443,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004708019165889683,
      "loss": 5.9982,
      "mean_token_accuracy": 0.17071737796068193,
      "num_tokens": 38131390.0,
      "step": 15045
    },
    {
      "entropy": 6.163508749008178,
      "epoch": 1.7368724754760532,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004707815846463199,
      "loss": 5.9492,
      "mean_token_accuracy": 0.16905829608440398,
      "num_tokens": 38145393.0,
      "step": 15050
    },
    {
      "entropy": 6.286931085586548,
      "epoch": 1.7374495095210616,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004707612461186552,
      "loss": 6.0697,
      "mean_token_accuracy": 0.16377016454935073,
      "num_tokens": 38158341.0,
      "step": 15055
    },
    {
      "entropy": 6.291764163970948,
      "epoch": 1.7380265435660704,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047074090100665805,
      "loss": 6.0316,
      "mean_token_accuracy": 0.16984689831733704,
      "num_tokens": 38170796.0,
      "step": 15060
    },
    {
      "entropy": 6.197325992584228,
      "epoch": 1.738603577611079,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047072054931101306,
      "loss": 5.9847,
      "mean_token_accuracy": 0.16967191696166992,
      "num_tokens": 38184280.0,
      "step": 15065
    },
    {
      "entropy": 6.204004144668579,
      "epoch": 1.7391806116560877,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004707001910324046,
      "loss": 6.0022,
      "mean_token_accuracy": 0.16744782626628876,
      "num_tokens": 38195709.0,
      "step": 15070
    },
    {
      "entropy": 6.228013944625855,
      "epoch": 1.7397576457010964,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047067982617151737,
      "loss": 6.0434,
      "mean_token_accuracy": 0.1737017586827278,
      "num_tokens": 38208904.0,
      "step": 15075
    },
    {
      "entropy": 6.160681819915771,
      "epoch": 1.740334679746105,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004706594547290365,
      "loss": 6.0243,
      "mean_token_accuracy": 0.16645268648862838,
      "num_tokens": 38222827.0,
      "step": 15080
    },
    {
      "entropy": 6.190474128723144,
      "epoch": 1.7409117137911136,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047063907670564695,
      "loss": 6.0204,
      "mean_token_accuracy": 0.17178211212158204,
      "num_tokens": 38235972.0,
      "step": 15085
    },
    {
      "entropy": 6.1540998935699465,
      "epoch": 1.7414887478361223,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004706186921020342,
      "loss": 5.866,
      "mean_token_accuracy": 0.1827749416232109,
      "num_tokens": 38247998.0,
      "step": 15090
    },
    {
      "entropy": 6.19387412071228,
      "epoch": 1.7420657818811311,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047059830091888407,
      "loss": 6.0708,
      "mean_token_accuracy": 0.16434583216905593,
      "num_tokens": 38260460.0,
      "step": 15095
    },
    {
      "entropy": 6.1568567752838135,
      "epoch": 1.7426428159261396,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004705779031568821,
      "loss": 5.9044,
      "mean_token_accuracy": 0.1804940477013588,
      "num_tokens": 38271829.0,
      "step": 15100
    },
    {
      "entropy": 6.143571090698242,
      "epoch": 1.7432198499711484,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047055749881671463,
      "loss": 5.9359,
      "mean_token_accuracy": 0.17703107744455338,
      "num_tokens": 38283602.0,
      "step": 15105
    },
    {
      "entropy": 6.206677961349487,
      "epoch": 1.7437968840161568,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004705370878990677,
      "loss": 6.0189,
      "mean_token_accuracy": 0.1702786475419998,
      "num_tokens": 38296112.0,
      "step": 15110
    },
    {
      "entropy": 6.2077109813690186,
      "epoch": 1.7443739180611657,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047051667040462806,
      "loss": 5.9929,
      "mean_token_accuracy": 0.17309020906686784,
      "num_tokens": 38309159.0,
      "step": 15115
    },
    {
      "entropy": 6.175277519226074,
      "epoch": 1.7449509521061741,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047049624633408224,
      "loss": 5.9504,
      "mean_token_accuracy": 0.175517301261425,
      "num_tokens": 38321468.0,
      "step": 15120
    },
    {
      "entropy": 6.14253249168396,
      "epoch": 1.745527986151183,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047047581568811724,
      "loss": 5.9485,
      "mean_token_accuracy": 0.16967506855726242,
      "num_tokens": 38333866.0,
      "step": 15125
    },
    {
      "entropy": 6.1963437557220455,
      "epoch": 1.7461050201961916,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047045537846742043,
      "loss": 6.0153,
      "mean_token_accuracy": 0.1668930858373642,
      "num_tokens": 38345965.0,
      "step": 15130
    },
    {
      "entropy": 6.144978857040405,
      "epoch": 1.7466820542412003,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000470434934672679,
      "loss": 5.9415,
      "mean_token_accuracy": 0.17128112018108368,
      "num_tokens": 38358536.0,
      "step": 15135
    },
    {
      "entropy": 6.198805427551269,
      "epoch": 1.747259088286209,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047041448430458054,
      "loss": 5.9195,
      "mean_token_accuracy": 0.18110038191080094,
      "num_tokens": 38371209.0,
      "step": 15140
    },
    {
      "entropy": 6.191978836059571,
      "epoch": 1.7478361223312175,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047039402736381305,
      "loss": 5.9956,
      "mean_token_accuracy": 0.17341268211603164,
      "num_tokens": 38384270.0,
      "step": 15145
    },
    {
      "entropy": 6.147642660140991,
      "epoch": 1.7484131563762262,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004703735638510645,
      "loss": 5.8653,
      "mean_token_accuracy": 0.18669939041137695,
      "num_tokens": 38396328.0,
      "step": 15150
    },
    {
      "entropy": 6.206048631668091,
      "epoch": 1.7489901904212348,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004703530937670232,
      "loss": 6.0139,
      "mean_token_accuracy": 0.17168012112379075,
      "num_tokens": 38408460.0,
      "step": 15155
    },
    {
      "entropy": 6.130110359191894,
      "epoch": 1.7495672244662435,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004703326171123776,
      "loss": 5.9651,
      "mean_token_accuracy": 0.17124782651662826,
      "num_tokens": 38421211.0,
      "step": 15160
    },
    {
      "entropy": 6.12564492225647,
      "epoch": 1.750144258511252,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004703121338878164,
      "loss": 6.0243,
      "mean_token_accuracy": 0.1676825299859047,
      "num_tokens": 38433720.0,
      "step": 15165
    },
    {
      "entropy": 6.250373315811157,
      "epoch": 1.750721292556261,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004702916440940286,
      "loss": 5.9478,
      "mean_token_accuracy": 0.17314032316207886,
      "num_tokens": 38445776.0,
      "step": 15170
    },
    {
      "entropy": 6.181895303726196,
      "epoch": 1.7512983266012694,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004702711477317034,
      "loss": 5.913,
      "mean_token_accuracy": 0.1707316517829895,
      "num_tokens": 38457218.0,
      "step": 15175
    },
    {
      "entropy": 6.1652435779571535,
      "epoch": 1.7518753606462782,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004702506448015301,
      "loss": 6.0179,
      "mean_token_accuracy": 0.1684328079223633,
      "num_tokens": 38469944.0,
      "step": 15180
    },
    {
      "entropy": 6.242323160171509,
      "epoch": 1.7524523946912867,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047023013530419843,
      "loss": 5.9937,
      "mean_token_accuracy": 0.17218188494443892,
      "num_tokens": 38482683.0,
      "step": 15185
    },
    {
      "entropy": 6.154385089874268,
      "epoch": 1.7530294287362955,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004702096192403981,
      "loss": 5.9115,
      "mean_token_accuracy": 0.17809037119150162,
      "num_tokens": 38494652.0,
      "step": 15190
    },
    {
      "entropy": 6.202943277359009,
      "epoch": 1.753606462781304,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004701890966108192,
      "loss": 5.9842,
      "mean_token_accuracy": 0.17837140411138536,
      "num_tokens": 38506567.0,
      "step": 15195
    },
    {
      "entropy": 6.211185359954834,
      "epoch": 1.7541834968263128,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000470168567416152,
      "loss": 5.9751,
      "mean_token_accuracy": 0.17064291387796401,
      "num_tokens": 38519399.0,
      "step": 15200
    },
    {
      "entropy": 6.1117274284362795,
      "epoch": 1.7547605308713214,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004701480316570869,
      "loss": 5.9036,
      "mean_token_accuracy": 0.17843794226646423,
      "num_tokens": 38533100.0,
      "step": 15205
    },
    {
      "entropy": 6.187893581390381,
      "epoch": 1.75533756491633,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004701274893343147,
      "loss": 5.988,
      "mean_token_accuracy": 0.1717816784977913,
      "num_tokens": 38545412.0,
      "step": 15210
    },
    {
      "entropy": 6.175931596755982,
      "epoch": 1.7559145989613387,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047010694044852643,
      "loss": 5.9527,
      "mean_token_accuracy": 0.1764634683728218,
      "num_tokens": 38557428.0,
      "step": 15215
    },
    {
      "entropy": 6.153143310546875,
      "epoch": 1.7564916330063474,
      "grad_norm": 1.140625,
      "learning_rate": 0.000470086385000413,
      "loss": 6.0424,
      "mean_token_accuracy": 0.1752571254968643,
      "num_tokens": 38569656.0,
      "step": 15220
    },
    {
      "entropy": 6.204163408279419,
      "epoch": 1.757068667051356,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004700658229906661,
      "loss": 6.0245,
      "mean_token_accuracy": 0.16561446189880372,
      "num_tokens": 38582280.0,
      "step": 15225
    },
    {
      "entropy": 6.144233369827271,
      "epoch": 1.7576457010963646,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047004525441997694,
      "loss": 5.9333,
      "mean_token_accuracy": 0.17786281704902648,
      "num_tokens": 38594578.0,
      "step": 15230
    },
    {
      "entropy": 6.135379791259766,
      "epoch": 1.7582227351413735,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004700246792890376,
      "loss": 5.9719,
      "mean_token_accuracy": 0.1712510481476784,
      "num_tokens": 38606803.0,
      "step": 15235
    },
    {
      "entropy": 6.212940263748169,
      "epoch": 1.758799769186382,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004700040975985401,
      "loss": 5.9359,
      "mean_token_accuracy": 0.17288441061973572,
      "num_tokens": 38618851.0,
      "step": 15240
    },
    {
      "entropy": 6.14959135055542,
      "epoch": 1.7593768032313908,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046998350934917654,
      "loss": 5.9038,
      "mean_token_accuracy": 0.1738117203116417,
      "num_tokens": 38630505.0,
      "step": 15245
    },
    {
      "entropy": 6.134607458114624,
      "epoch": 1.7599538372763992,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046996291454163956,
      "loss": 5.9548,
      "mean_token_accuracy": 0.17708190381526948,
      "num_tokens": 38642601.0,
      "step": 15250
    },
    {
      "entropy": 6.1670698642730715,
      "epoch": 1.760530871321408,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004699423131766218,
      "loss": 5.957,
      "mean_token_accuracy": 0.1744084030389786,
      "num_tokens": 38655713.0,
      "step": 15255
    },
    {
      "entropy": 6.291418743133545,
      "epoch": 1.7611079053664165,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004699217052548161,
      "loss": 6.0439,
      "mean_token_accuracy": 0.1678642749786377,
      "num_tokens": 38667909.0,
      "step": 15260
    },
    {
      "entropy": 6.227230596542358,
      "epoch": 1.7616849394114253,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046990109077691577,
      "loss": 6.0525,
      "mean_token_accuracy": 0.16827605068683624,
      "num_tokens": 38681729.0,
      "step": 15265
    },
    {
      "entropy": 6.264706802368164,
      "epoch": 1.762261973456434,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046988046974361406,
      "loss": 6.0429,
      "mean_token_accuracy": 0.16819683015346526,
      "num_tokens": 38694957.0,
      "step": 15270
    },
    {
      "entropy": 6.217157077789307,
      "epoch": 1.7628390075014426,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004698598421556045,
      "loss": 6.0052,
      "mean_token_accuracy": 0.17774756401777267,
      "num_tokens": 38707914.0,
      "step": 15275
    },
    {
      "entropy": 6.1865074157714846,
      "epoch": 1.7634160415464513,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004698392080135809,
      "loss": 5.9422,
      "mean_token_accuracy": 0.1725253701210022,
      "num_tokens": 38719011.0,
      "step": 15280
    },
    {
      "entropy": 6.149824905395508,
      "epoch": 1.76399307559146,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004698185673182374,
      "loss": 5.9464,
      "mean_token_accuracy": 0.17740254253149032,
      "num_tokens": 38731473.0,
      "step": 15285
    },
    {
      "entropy": 6.132081985473633,
      "epoch": 1.7645701096364685,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046979792007026817,
      "loss": 6.0465,
      "mean_token_accuracy": 0.1713622272014618,
      "num_tokens": 38745061.0,
      "step": 15290
    },
    {
      "entropy": 6.269248390197754,
      "epoch": 1.7651471436814772,
      "grad_norm": 1.0,
      "learning_rate": 0.0004697772662703676,
      "loss": 5.9737,
      "mean_token_accuracy": 0.17214820832014083,
      "num_tokens": 38757980.0,
      "step": 15295
    },
    {
      "entropy": 6.163847303390503,
      "epoch": 1.7657241777264858,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046975660591923047,
      "loss": 6.0862,
      "mean_token_accuracy": 0.1635837137699127,
      "num_tokens": 38771918.0,
      "step": 15300
    },
    {
      "entropy": 6.155009746551514,
      "epoch": 1.7663012117714945,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004697359390175516,
      "loss": 5.9462,
      "mean_token_accuracy": 0.17715550512075423,
      "num_tokens": 38783702.0,
      "step": 15305
    },
    {
      "entropy": 6.124956846237183,
      "epoch": 1.7668782458165033,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046971526556602625,
      "loss": 5.9259,
      "mean_token_accuracy": 0.1846996769309044,
      "num_tokens": 38797574.0,
      "step": 15310
    },
    {
      "entropy": 6.176227331161499,
      "epoch": 1.7674552798615117,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004696945855653495,
      "loss": 5.8942,
      "mean_token_accuracy": 0.17883059829473497,
      "num_tokens": 38810302.0,
      "step": 15315
    },
    {
      "entropy": 6.057977819442749,
      "epoch": 1.7680323139065206,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004696738990162172,
      "loss": 5.8604,
      "mean_token_accuracy": 0.17707136422395706,
      "num_tokens": 38823968.0,
      "step": 15320
    },
    {
      "entropy": 6.17450442314148,
      "epoch": 1.768609347951529,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000469653205919325,
      "loss": 5.9655,
      "mean_token_accuracy": 0.1705733969807625,
      "num_tokens": 38837641.0,
      "step": 15325
    },
    {
      "entropy": 6.156165504455567,
      "epoch": 1.7691863819965379,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046963250627536884,
      "loss": 5.9271,
      "mean_token_accuracy": 0.17344410419464112,
      "num_tokens": 38849556.0,
      "step": 15330
    },
    {
      "entropy": 6.155330181121826,
      "epoch": 1.7697634160415463,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004696118000850451,
      "loss": 5.9648,
      "mean_token_accuracy": 0.1715877339243889,
      "num_tokens": 38863276.0,
      "step": 15335
    },
    {
      "entropy": 6.142491006851197,
      "epoch": 1.7703404500865552,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046959108734905003,
      "loss": 5.993,
      "mean_token_accuracy": 0.17025951147079468,
      "num_tokens": 38876230.0,
      "step": 15340
    },
    {
      "entropy": 6.242098093032837,
      "epoch": 1.7709174841315638,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046957036806808045,
      "loss": 6.0773,
      "mean_token_accuracy": 0.16869438737630843,
      "num_tokens": 38891269.0,
      "step": 15345
    },
    {
      "entropy": 6.189268636703491,
      "epoch": 1.7714945181765724,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004695496422428332,
      "loss": 6.0132,
      "mean_token_accuracy": 0.16962060034275056,
      "num_tokens": 38904081.0,
      "step": 15350
    },
    {
      "entropy": 6.161999893188477,
      "epoch": 1.772071552221581,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004695289098740054,
      "loss": 5.9393,
      "mean_token_accuracy": 0.17272746115922927,
      "num_tokens": 38916422.0,
      "step": 15355
    },
    {
      "entropy": 6.297621154785157,
      "epoch": 1.7726485862665897,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004695081709622943,
      "loss": 6.0301,
      "mean_token_accuracy": 0.167911534011364,
      "num_tokens": 38928554.0,
      "step": 15360
    },
    {
      "entropy": 6.165102243423462,
      "epoch": 1.7732256203115984,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046948742550839744,
      "loss": 5.9369,
      "mean_token_accuracy": 0.17219328731298447,
      "num_tokens": 38942252.0,
      "step": 15365
    },
    {
      "entropy": 6.2235502243042,
      "epoch": 1.773802654356607,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004694666735130127,
      "loss": 6.027,
      "mean_token_accuracy": 0.16650664806365967,
      "num_tokens": 38955604.0,
      "step": 15370
    },
    {
      "entropy": 6.221412277221679,
      "epoch": 1.7743796884016156,
      "grad_norm": 1.0625,
      "learning_rate": 0.000469445914976838,
      "loss": 5.9556,
      "mean_token_accuracy": 0.1788329854607582,
      "num_tokens": 38968439.0,
      "step": 15375
    },
    {
      "entropy": 6.1664549827575685,
      "epoch": 1.7749567224466243,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004694251499005715,
      "loss": 5.991,
      "mean_token_accuracy": 0.17186417579650878,
      "num_tokens": 38980251.0,
      "step": 15380
    },
    {
      "entropy": 6.158185148239136,
      "epoch": 1.7755337564916331,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004694043782849116,
      "loss": 5.9851,
      "mean_token_accuracy": 0.17879897505044937,
      "num_tokens": 38992242.0,
      "step": 15385
    },
    {
      "entropy": 6.133122539520263,
      "epoch": 1.7761107905366416,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046938360013055715,
      "loss": 5.9235,
      "mean_token_accuracy": 0.16716435104608535,
      "num_tokens": 39004372.0,
      "step": 15390
    },
    {
      "entropy": 6.1651139736175535,
      "epoch": 1.7766878245816504,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046936281543820673,
      "loss": 5.9582,
      "mean_token_accuracy": 0.17229029834270476,
      "num_tokens": 39015928.0,
      "step": 15395
    },
    {
      "entropy": 6.143525266647339,
      "epoch": 1.7772648586266588,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046934202420855967,
      "loss": 5.9444,
      "mean_token_accuracy": 0.17930613160133363,
      "num_tokens": 39028948.0,
      "step": 15400
    },
    {
      "entropy": 6.121102571487427,
      "epoch": 1.7778418926716677,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046932122644231507,
      "loss": 5.9866,
      "mean_token_accuracy": 0.16983349472284318,
      "num_tokens": 39042486.0,
      "step": 15405
    },
    {
      "entropy": 6.198353385925293,
      "epoch": 1.7784189267166761,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046930042214017256,
      "loss": 6.0505,
      "mean_token_accuracy": 0.17363871037960052,
      "num_tokens": 39054660.0,
      "step": 15410
    },
    {
      "entropy": 6.216564559936524,
      "epoch": 1.778995960761685,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004692796113028319,
      "loss": 6.0322,
      "mean_token_accuracy": 0.16882157772779466,
      "num_tokens": 39067272.0,
      "step": 15415
    },
    {
      "entropy": 6.256233835220337,
      "epoch": 1.7795729948066936,
      "grad_norm": 1.09375,
      "learning_rate": 0.000469258793930993,
      "loss": 6.0006,
      "mean_token_accuracy": 0.16928353309631347,
      "num_tokens": 39080146.0,
      "step": 15420
    },
    {
      "entropy": 6.122313690185547,
      "epoch": 1.7801500288517023,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046923797002535605,
      "loss": 5.8629,
      "mean_token_accuracy": 0.18094716668128968,
      "num_tokens": 39093093.0,
      "step": 15425
    },
    {
      "entropy": 6.2192699909210205,
      "epoch": 1.780727062896711,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004692171395866214,
      "loss": 5.9706,
      "mean_token_accuracy": 0.17284361124038697,
      "num_tokens": 39105239.0,
      "step": 15430
    },
    {
      "entropy": 6.184670639038086,
      "epoch": 1.7813040969417195,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046919630261548986,
      "loss": 6.0024,
      "mean_token_accuracy": 0.170201113820076,
      "num_tokens": 39118265.0,
      "step": 15435
    },
    {
      "entropy": 6.164061641693115,
      "epoch": 1.7818811309867282,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046917545911266207,
      "loss": 6.0576,
      "mean_token_accuracy": 0.15933084338903428,
      "num_tokens": 39131999.0,
      "step": 15440
    },
    {
      "entropy": 6.2334857940673825,
      "epoch": 1.7824581650317368,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046915460907883923,
      "loss": 6.006,
      "mean_token_accuracy": 0.16682424694299697,
      "num_tokens": 39145529.0,
      "step": 15445
    },
    {
      "entropy": 6.085377550125122,
      "epoch": 1.7830351990767457,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046913375251472246,
      "loss": 5.7955,
      "mean_token_accuracy": 0.17601656764745713,
      "num_tokens": 39157736.0,
      "step": 15450
    },
    {
      "entropy": 6.199916887283325,
      "epoch": 1.783612233121754,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046911288942101345,
      "loss": 6.0223,
      "mean_token_accuracy": 0.16780002415180206,
      "num_tokens": 39169126.0,
      "step": 15455
    },
    {
      "entropy": 6.279119920730591,
      "epoch": 1.784189267166763,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004690920197984138,
      "loss": 6.0603,
      "mean_token_accuracy": 0.1718040958046913,
      "num_tokens": 39183190.0,
      "step": 15460
    },
    {
      "entropy": 6.169201850891113,
      "epoch": 1.7847663012117714,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004690711436476254,
      "loss": 5.8998,
      "mean_token_accuracy": 0.18014869689941407,
      "num_tokens": 39194874.0,
      "step": 15465
    },
    {
      "entropy": 6.184126091003418,
      "epoch": 1.7853433352567802,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046905026096935056,
      "loss": 6.0038,
      "mean_token_accuracy": 0.17372843772172927,
      "num_tokens": 39208373.0,
      "step": 15470
    },
    {
      "entropy": 6.179333114624024,
      "epoch": 1.7859203693017887,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004690293717642916,
      "loss": 5.9343,
      "mean_token_accuracy": 0.1735247179865837,
      "num_tokens": 39219848.0,
      "step": 15475
    },
    {
      "entropy": 6.177401161193847,
      "epoch": 1.7864974033467975,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000469008476033151,
      "loss": 5.9503,
      "mean_token_accuracy": 0.17796252220869063,
      "num_tokens": 39232737.0,
      "step": 15480
    },
    {
      "entropy": 6.235263347625732,
      "epoch": 1.7870744373918062,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046898757377663176,
      "loss": 6.0718,
      "mean_token_accuracy": 0.1667163610458374,
      "num_tokens": 39245220.0,
      "step": 15485
    },
    {
      "entropy": 6.240390205383301,
      "epoch": 1.7876514714368148,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046896666499543683,
      "loss": 6.0229,
      "mean_token_accuracy": 0.16968497931957244,
      "num_tokens": 39257828.0,
      "step": 15490
    },
    {
      "entropy": 6.198735141754151,
      "epoch": 1.7882285054818234,
      "grad_norm": 1.0,
      "learning_rate": 0.00046894574969026946,
      "loss": 6.0147,
      "mean_token_accuracy": 0.16953890174627304,
      "num_tokens": 39270679.0,
      "step": 15495
    },
    {
      "entropy": 6.205756330490113,
      "epoch": 1.788805539526832,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046892482786183313,
      "loss": 5.9901,
      "mean_token_accuracy": 0.17093409299850465,
      "num_tokens": 39282634.0,
      "step": 15500
    },
    {
      "entropy": 6.1952447414398195,
      "epoch": 1.7893825735718407,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046890389951083155,
      "loss": 5.919,
      "mean_token_accuracy": 0.1857565686106682,
      "num_tokens": 39295383.0,
      "step": 15505
    },
    {
      "entropy": 6.221450328826904,
      "epoch": 1.7899596076168494,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046888296463796857,
      "loss": 5.98,
      "mean_token_accuracy": 0.16859333962202072,
      "num_tokens": 39308636.0,
      "step": 15510
    },
    {
      "entropy": 6.261836147308349,
      "epoch": 1.790536641661858,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004688620232439485,
      "loss": 6.0906,
      "mean_token_accuracy": 0.16861074566841125,
      "num_tokens": 39321330.0,
      "step": 15515
    },
    {
      "entropy": 6.249241495132447,
      "epoch": 1.7911136757068666,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046884107532947547,
      "loss": 6.0814,
      "mean_token_accuracy": 0.1648913025856018,
      "num_tokens": 39334920.0,
      "step": 15520
    },
    {
      "entropy": 6.190037250518799,
      "epoch": 1.7916907097518755,
      "grad_norm": 1.453125,
      "learning_rate": 0.00046882012089525415,
      "loss": 6.0063,
      "mean_token_accuracy": 0.1752471312880516,
      "num_tokens": 39347827.0,
      "step": 15525
    },
    {
      "entropy": 6.118485975265503,
      "epoch": 1.792267743796884,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004687991599419895,
      "loss": 5.931,
      "mean_token_accuracy": 0.17592341154813768,
      "num_tokens": 39359732.0,
      "step": 15530
    },
    {
      "entropy": 6.050671529769898,
      "epoch": 1.7928447778418928,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046877819247038624,
      "loss": 5.809,
      "mean_token_accuracy": 0.19305932074785231,
      "num_tokens": 39372056.0,
      "step": 15535
    },
    {
      "entropy": 6.158288478851318,
      "epoch": 1.7934218118869012,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004687572184811497,
      "loss": 5.9377,
      "mean_token_accuracy": 0.17697646468877792,
      "num_tokens": 39383302.0,
      "step": 15540
    },
    {
      "entropy": 6.123100423812867,
      "epoch": 1.79399884593191,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046873623797498545,
      "loss": 5.9548,
      "mean_token_accuracy": 0.17160976827144622,
      "num_tokens": 39395666.0,
      "step": 15545
    },
    {
      "entropy": 6.189558124542236,
      "epoch": 1.7945758799769185,
      "grad_norm": 1.046875,
      "learning_rate": 0.000468715250952599,
      "loss": 5.8837,
      "mean_token_accuracy": 0.17876604646444322,
      "num_tokens": 39407192.0,
      "step": 15550
    },
    {
      "entropy": 6.257903671264648,
      "epoch": 1.7951529140219273,
      "grad_norm": 1.4375,
      "learning_rate": 0.00046869425741469635,
      "loss": 6.0728,
      "mean_token_accuracy": 0.17084383666515351,
      "num_tokens": 39418331.0,
      "step": 15555
    },
    {
      "entropy": 6.2055830478668215,
      "epoch": 1.795729948066936,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004686732573619835,
      "loss": 5.988,
      "mean_token_accuracy": 0.17247890532016755,
      "num_tokens": 39431789.0,
      "step": 15560
    },
    {
      "entropy": 6.204622411727906,
      "epoch": 1.7963069821119446,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004686522507951669,
      "loss": 5.9718,
      "mean_token_accuracy": 0.17110926359891893,
      "num_tokens": 39444516.0,
      "step": 15565
    },
    {
      "entropy": 6.140855884552002,
      "epoch": 1.7968840161569533,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004686312377149531,
      "loss": 5.9363,
      "mean_token_accuracy": 0.1744287610054016,
      "num_tokens": 39456490.0,
      "step": 15570
    },
    {
      "entropy": 6.162423133850098,
      "epoch": 1.797461050201962,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046861021812204874,
      "loss": 5.9423,
      "mean_token_accuracy": 0.17548976093530655,
      "num_tokens": 39468232.0,
      "step": 15575
    },
    {
      "entropy": 6.207177591323853,
      "epoch": 1.7980380842469705,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046858919201716085,
      "loss": 5.9696,
      "mean_token_accuracy": 0.1753180965781212,
      "num_tokens": 39481037.0,
      "step": 15580
    },
    {
      "entropy": 6.023946523666382,
      "epoch": 1.7986151182919792,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004685681594009966,
      "loss": 5.8929,
      "mean_token_accuracy": 0.17770785838365555,
      "num_tokens": 39493872.0,
      "step": 15585
    },
    {
      "entropy": 6.174573087692261,
      "epoch": 1.799192152336988,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00046854712027426357,
      "loss": 5.9913,
      "mean_token_accuracy": 0.1661468744277954,
      "num_tokens": 39505997.0,
      "step": 15590
    },
    {
      "entropy": 6.296399641036987,
      "epoch": 1.7997691863819965,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046852607463766923,
      "loss": 6.0673,
      "mean_token_accuracy": 0.17015460729599,
      "num_tokens": 39519114.0,
      "step": 15595
    },
    {
      "entropy": 6.1773223876953125,
      "epoch": 1.8003462204270053,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004685050224919215,
      "loss": 5.9343,
      "mean_token_accuracy": 0.17734288573265075,
      "num_tokens": 39530629.0,
      "step": 15600
    },
    {
      "entropy": 6.025782537460327,
      "epoch": 1.8009232544720137,
      "grad_norm": 1.4375,
      "learning_rate": 0.00046848396383772844,
      "loss": 5.9404,
      "mean_token_accuracy": 0.17582221031188966,
      "num_tokens": 39543549.0,
      "step": 15605
    },
    {
      "entropy": 6.16446590423584,
      "epoch": 1.8015002885170226,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004684628986757984,
      "loss": 6.0031,
      "mean_token_accuracy": 0.16606678664684296,
      "num_tokens": 39555487.0,
      "step": 15610
    },
    {
      "entropy": 6.241364574432373,
      "epoch": 1.802077322562031,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004684418270068398,
      "loss": 5.9433,
      "mean_token_accuracy": 0.17523407340049743,
      "num_tokens": 39568594.0,
      "step": 15615
    },
    {
      "entropy": 6.1396221160888675,
      "epoch": 1.8026543566070399,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004684207488315615,
      "loss": 5.8951,
      "mean_token_accuracy": 0.17363064587116242,
      "num_tokens": 39582069.0,
      "step": 15620
    },
    {
      "entropy": 6.10598931312561,
      "epoch": 1.8032313906520485,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004683996641506724,
      "loss": 5.9651,
      "mean_token_accuracy": 0.16897824108600618,
      "num_tokens": 39595316.0,
      "step": 15625
    },
    {
      "entropy": 6.183296871185303,
      "epoch": 1.8038084246970572,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00046837857296488163,
      "loss": 5.8655,
      "mean_token_accuracy": 0.1799294888973236,
      "num_tokens": 39609363.0,
      "step": 15630
    },
    {
      "entropy": 6.151688289642334,
      "epoch": 1.8043854587420658,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046835747527489857,
      "loss": 6.0143,
      "mean_token_accuracy": 0.17060075402259828,
      "num_tokens": 39621597.0,
      "step": 15635
    },
    {
      "entropy": 6.264867830276489,
      "epoch": 1.8049624927870744,
      "grad_norm": 1.171875,
      "learning_rate": 0.000468336371081433,
      "loss": 6.0372,
      "mean_token_accuracy": 0.16418321877717973,
      "num_tokens": 39633913.0,
      "step": 15640
    },
    {
      "entropy": 6.190672540664673,
      "epoch": 1.805539526832083,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046831526038519444,
      "loss": 5.9697,
      "mean_token_accuracy": 0.18033531457185745,
      "num_tokens": 39646616.0,
      "step": 15645
    },
    {
      "entropy": 6.185319280624389,
      "epoch": 1.8061165608770917,
      "grad_norm": 1.25,
      "learning_rate": 0.0004682941431868933,
      "loss": 5.9902,
      "mean_token_accuracy": 0.16639287024736404,
      "num_tokens": 39659636.0,
      "step": 15650
    },
    {
      "entropy": 6.214107084274292,
      "epoch": 1.8066935949221004,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046827301948723963,
      "loss": 5.933,
      "mean_token_accuracy": 0.1785548821091652,
      "num_tokens": 39672824.0,
      "step": 15655
    },
    {
      "entropy": 6.174220275878906,
      "epoch": 1.807270628967109,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046825188928694393,
      "loss": 5.959,
      "mean_token_accuracy": 0.174652099609375,
      "num_tokens": 39684988.0,
      "step": 15660
    },
    {
      "entropy": 6.207254886627197,
      "epoch": 1.8078476630121179,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004682307525867169,
      "loss": 5.9706,
      "mean_token_accuracy": 0.17355331927537918,
      "num_tokens": 39697313.0,
      "step": 15665
    },
    {
      "entropy": 6.142776536941528,
      "epoch": 1.8084246970571263,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004682096093872695,
      "loss": 5.956,
      "mean_token_accuracy": 0.1750632643699646,
      "num_tokens": 39710005.0,
      "step": 15670
    },
    {
      "entropy": 6.08503737449646,
      "epoch": 1.8090017311021351,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046818845968931284,
      "loss": 5.9029,
      "mean_token_accuracy": 0.18228058516979218,
      "num_tokens": 39724495.0,
      "step": 15675
    },
    {
      "entropy": 6.212007617950439,
      "epoch": 1.8095787651471436,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046816730349355843,
      "loss": 5.9818,
      "mean_token_accuracy": 0.17354291826486587,
      "num_tokens": 39736978.0,
      "step": 15680
    },
    {
      "entropy": 6.24097089767456,
      "epoch": 1.8101557991921524,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046814614080071756,
      "loss": 6.0214,
      "mean_token_accuracy": 0.1683731570839882,
      "num_tokens": 39748694.0,
      "step": 15685
    },
    {
      "entropy": 6.172418165206909,
      "epoch": 1.8107328332371608,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046812497161150224,
      "loss": 5.9623,
      "mean_token_accuracy": 0.17081085592508316,
      "num_tokens": 39760388.0,
      "step": 15690
    },
    {
      "entropy": 6.155089426040649,
      "epoch": 1.8113098672821697,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046810379592662445,
      "loss": 5.928,
      "mean_token_accuracy": 0.18098287135362626,
      "num_tokens": 39773424.0,
      "step": 15695
    },
    {
      "entropy": 6.211879873275757,
      "epoch": 1.8118869013271783,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046808261374679637,
      "loss": 5.933,
      "mean_token_accuracy": 0.17311359345912933,
      "num_tokens": 39786472.0,
      "step": 15700
    },
    {
      "entropy": 6.130142259597778,
      "epoch": 1.812463935372187,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004680614250727305,
      "loss": 5.9377,
      "mean_token_accuracy": 0.17745109647512436,
      "num_tokens": 39799968.0,
      "step": 15705
    },
    {
      "entropy": 6.183506298065185,
      "epoch": 1.8130409694171956,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004680402299051395,
      "loss": 5.9661,
      "mean_token_accuracy": 0.18038853406906127,
      "num_tokens": 39812332.0,
      "step": 15710
    },
    {
      "entropy": 6.239239120483399,
      "epoch": 1.8136180034622043,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004680190282447363,
      "loss": 6.0214,
      "mean_token_accuracy": 0.1716531366109848,
      "num_tokens": 39825545.0,
      "step": 15715
    },
    {
      "entropy": 6.2020550727844235,
      "epoch": 1.814195037507213,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004679978200922339,
      "loss": 6.1047,
      "mean_token_accuracy": 0.1656905472278595,
      "num_tokens": 39839691.0,
      "step": 15720
    },
    {
      "entropy": 6.119612550735473,
      "epoch": 1.8147720715522215,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004679766054483457,
      "loss": 5.8801,
      "mean_token_accuracy": 0.17325713187456132,
      "num_tokens": 39852225.0,
      "step": 15725
    },
    {
      "entropy": 6.127541589736938,
      "epoch": 1.8153491055972304,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004679553843137852,
      "loss": 5.8596,
      "mean_token_accuracy": 0.17421128004789352,
      "num_tokens": 39866451.0,
      "step": 15730
    },
    {
      "entropy": 6.135132932662964,
      "epoch": 1.8159261396422388,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046793415668926625,
      "loss": 5.9613,
      "mean_token_accuracy": 0.17361737489700318,
      "num_tokens": 39878564.0,
      "step": 15735
    },
    {
      "entropy": 6.16522970199585,
      "epoch": 1.8165031736872477,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004679129225755028,
      "loss": 5.9885,
      "mean_token_accuracy": 0.17147571891546248,
      "num_tokens": 39890588.0,
      "step": 15740
    },
    {
      "entropy": 6.190404224395752,
      "epoch": 1.817080207732256,
      "grad_norm": 1.03125,
      "learning_rate": 0.000467891681973209,
      "loss": 5.8903,
      "mean_token_accuracy": 0.18329087197780608,
      "num_tokens": 39901868.0,
      "step": 15745
    },
    {
      "entropy": 6.251967859268189,
      "epoch": 1.817657241777265,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046787043488309926,
      "loss": 5.9849,
      "mean_token_accuracy": 0.1769744262099266,
      "num_tokens": 39914475.0,
      "step": 15750
    },
    {
      "entropy": 6.047009134292603,
      "epoch": 1.8182342758222734,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004678491813058882,
      "loss": 5.748,
      "mean_token_accuracy": 0.1842229500412941,
      "num_tokens": 39927488.0,
      "step": 15755
    },
    {
      "entropy": 6.215115213394165,
      "epoch": 1.8188113098672822,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004678279212422908,
      "loss": 6.0181,
      "mean_token_accuracy": 0.16714513599872588,
      "num_tokens": 39940753.0,
      "step": 15760
    },
    {
      "entropy": 6.234380197525025,
      "epoch": 1.8193883439122909,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004678066546930221,
      "loss": 5.9672,
      "mean_token_accuracy": 0.17353275716304778,
      "num_tokens": 39954779.0,
      "step": 15765
    },
    {
      "entropy": 6.184096765518189,
      "epoch": 1.8199653779572995,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046778538165879725,
      "loss": 5.9409,
      "mean_token_accuracy": 0.17122038900852204,
      "num_tokens": 39966962.0,
      "step": 15770
    },
    {
      "entropy": 6.163376998901367,
      "epoch": 1.8205424120023082,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046776410214033185,
      "loss": 5.9491,
      "mean_token_accuracy": 0.17713500261306764,
      "num_tokens": 39978590.0,
      "step": 15775
    },
    {
      "entropy": 6.15793628692627,
      "epoch": 1.8211194460473168,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004677428161383417,
      "loss": 5.88,
      "mean_token_accuracy": 0.1817947342991829,
      "num_tokens": 39990929.0,
      "step": 15780
    },
    {
      "entropy": 6.175368118286133,
      "epoch": 1.8216964800923254,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004677215236535426,
      "loss": 6.0283,
      "mean_token_accuracy": 0.17091631442308425,
      "num_tokens": 40003301.0,
      "step": 15785
    },
    {
      "entropy": 6.222087144851685,
      "epoch": 1.822273514137334,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004677002246866508,
      "loss": 5.9704,
      "mean_token_accuracy": 0.1720083698630333,
      "num_tokens": 40016325.0,
      "step": 15790
    },
    {
      "entropy": 6.244450187683105,
      "epoch": 1.8228505481823427,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046767891923838264,
      "loss": 5.9356,
      "mean_token_accuracy": 0.17754150480031966,
      "num_tokens": 40030666.0,
      "step": 15795
    },
    {
      "entropy": 6.187222290039062,
      "epoch": 1.8234275822273514,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004676576073094548,
      "loss": 5.9125,
      "mean_token_accuracy": 0.17317680567502974,
      "num_tokens": 40041192.0,
      "step": 15800
    },
    {
      "entropy": 6.223333692550659,
      "epoch": 1.8240046162723602,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046763628890058396,
      "loss": 5.9296,
      "mean_token_accuracy": 0.17504908591508866,
      "num_tokens": 40053620.0,
      "step": 15805
    },
    {
      "entropy": 6.141034555435181,
      "epoch": 1.8245816503173686,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046761496401248734,
      "loss": 5.8998,
      "mean_token_accuracy": 0.1789110466837883,
      "num_tokens": 40066268.0,
      "step": 15810
    },
    {
      "entropy": 6.186302757263183,
      "epoch": 1.8251586843623775,
      "grad_norm": 1.046875,
      "learning_rate": 0.000467593632645882,
      "loss": 6.0241,
      "mean_token_accuracy": 0.16849912106990814,
      "num_tokens": 40079251.0,
      "step": 15815
    },
    {
      "entropy": 6.213567876815796,
      "epoch": 1.825735718407386,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004675722948014855,
      "loss": 6.0002,
      "mean_token_accuracy": 0.17895146310329438,
      "num_tokens": 40093778.0,
      "step": 15820
    },
    {
      "entropy": 6.081861591339111,
      "epoch": 1.8263127524523948,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046755095048001556,
      "loss": 5.8074,
      "mean_token_accuracy": 0.18788816779851913,
      "num_tokens": 40106604.0,
      "step": 15825
    },
    {
      "entropy": 6.135801076889038,
      "epoch": 1.8268897864974032,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004675295996821899,
      "loss": 5.9696,
      "mean_token_accuracy": 0.17356823831796647,
      "num_tokens": 40119841.0,
      "step": 15830
    },
    {
      "entropy": 6.1972580909729,
      "epoch": 1.827466820542412,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000467508242408727,
      "loss": 6.041,
      "mean_token_accuracy": 0.17398447543382645,
      "num_tokens": 40132790.0,
      "step": 15835
    },
    {
      "entropy": 6.215589427947998,
      "epoch": 1.8280438545874207,
      "grad_norm": 1.0,
      "learning_rate": 0.0004674868786603448,
      "loss": 5.9635,
      "mean_token_accuracy": 0.16788264364004135,
      "num_tokens": 40145175.0,
      "step": 15840
    },
    {
      "entropy": 6.226518249511718,
      "epoch": 1.8286208886324293,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004674655084377622,
      "loss": 5.9869,
      "mean_token_accuracy": 0.17462982833385468,
      "num_tokens": 40158311.0,
      "step": 15845
    },
    {
      "entropy": 6.224665594100952,
      "epoch": 1.829197922677438,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004674441317416978,
      "loss": 6.1106,
      "mean_token_accuracy": 0.16693505197763442,
      "num_tokens": 40170486.0,
      "step": 15850
    },
    {
      "entropy": 6.167939329147339,
      "epoch": 1.8297749567224466,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046742274857287057,
      "loss": 6.0662,
      "mean_token_accuracy": 0.16844267249107361,
      "num_tokens": 40184065.0,
      "step": 15855
    },
    {
      "entropy": 6.220897960662842,
      "epoch": 1.8303519907674553,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004674013589319998,
      "loss": 5.9538,
      "mean_token_accuracy": 0.1779506802558899,
      "num_tokens": 40196387.0,
      "step": 15860
    },
    {
      "entropy": 6.269541692733765,
      "epoch": 1.830929024812464,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004673799628198049,
      "loss": 6.1145,
      "mean_token_accuracy": 0.15836625099182128,
      "num_tokens": 40208983.0,
      "step": 15865
    },
    {
      "entropy": 6.27610673904419,
      "epoch": 1.8315060588574728,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046735856023700546,
      "loss": 6.0433,
      "mean_token_accuracy": 0.16987049877643584,
      "num_tokens": 40223094.0,
      "step": 15870
    },
    {
      "entropy": 6.187994766235351,
      "epoch": 1.8320830929024812,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004673371511843215,
      "loss": 5.914,
      "mean_token_accuracy": 0.17617291510105132,
      "num_tokens": 40235027.0,
      "step": 15875
    },
    {
      "entropy": 6.1627685546875,
      "epoch": 1.83266012694749,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004673157356624729,
      "loss": 5.9602,
      "mean_token_accuracy": 0.17981325685977936,
      "num_tokens": 40247070.0,
      "step": 15880
    },
    {
      "entropy": 6.085093355178833,
      "epoch": 1.8332371609924984,
      "grad_norm": 1.0,
      "learning_rate": 0.0004672943136721801,
      "loss": 5.9094,
      "mean_token_accuracy": 0.17942582666873932,
      "num_tokens": 40259691.0,
      "step": 15885
    },
    {
      "entropy": 6.181700277328491,
      "epoch": 1.8338141950375073,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004672728852141636,
      "loss": 5.9844,
      "mean_token_accuracy": 0.16414399445056915,
      "num_tokens": 40272068.0,
      "step": 15890
    },
    {
      "entropy": 6.173003387451172,
      "epoch": 1.8343912290825157,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046725145028914404,
      "loss": 5.9237,
      "mean_token_accuracy": 0.17495870888233184,
      "num_tokens": 40285224.0,
      "step": 15895
    },
    {
      "entropy": 6.254984664916992,
      "epoch": 1.8349682631275246,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004672300088978425,
      "loss": 6.0556,
      "mean_token_accuracy": 0.16879668682813645,
      "num_tokens": 40297280.0,
      "step": 15900
    },
    {
      "entropy": 6.179126167297364,
      "epoch": 1.8355452971725332,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004672085610409801,
      "loss": 5.9257,
      "mean_token_accuracy": 0.17829905301332474,
      "num_tokens": 40309097.0,
      "step": 15905
    },
    {
      "entropy": 6.140904378890991,
      "epoch": 1.8361223312175419,
      "grad_norm": 1.0,
      "learning_rate": 0.00046718710671927815,
      "loss": 5.9543,
      "mean_token_accuracy": 0.17014556378126144,
      "num_tokens": 40321665.0,
      "step": 15910
    },
    {
      "entropy": 6.16058201789856,
      "epoch": 1.8366993652625505,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00046716564593345843,
      "loss": 5.9626,
      "mean_token_accuracy": 0.1762630298733711,
      "num_tokens": 40333889.0,
      "step": 15915
    },
    {
      "entropy": 6.244926881790161,
      "epoch": 1.8372763993075591,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00046714417868424265,
      "loss": 5.9576,
      "mean_token_accuracy": 0.16962620615959167,
      "num_tokens": 40346448.0,
      "step": 15920
    },
    {
      "entropy": 6.189945459365845,
      "epoch": 1.8378534333525678,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046712270497235284,
      "loss": 5.8959,
      "mean_token_accuracy": 0.17132564932107924,
      "num_tokens": 40359033.0,
      "step": 15925
    },
    {
      "entropy": 6.126277685165405,
      "epoch": 1.8384304673975764,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004671012247985112,
      "loss": 5.9224,
      "mean_token_accuracy": 0.1740038052201271,
      "num_tokens": 40372598.0,
      "step": 15930
    },
    {
      "entropy": 6.175677299499512,
      "epoch": 1.839007501442585,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046707973816344044,
      "loss": 6.0011,
      "mean_token_accuracy": 0.1706273466348648,
      "num_tokens": 40385009.0,
      "step": 15935
    },
    {
      "entropy": 6.201859378814698,
      "epoch": 1.8395845354875937,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046705824506786304,
      "loss": 6.0223,
      "mean_token_accuracy": 0.17398494482040405,
      "num_tokens": 40396368.0,
      "step": 15940
    },
    {
      "entropy": 6.116965675354004,
      "epoch": 1.8401615695326026,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046703674551250193,
      "loss": 5.8762,
      "mean_token_accuracy": 0.17988387793302535,
      "num_tokens": 40408084.0,
      "step": 15945
    },
    {
      "entropy": 6.157942533493042,
      "epoch": 1.840738603577611,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004670152394980803,
      "loss": 5.9251,
      "mean_token_accuracy": 0.1743283450603485,
      "num_tokens": 40420290.0,
      "step": 15950
    },
    {
      "entropy": 6.171168231964112,
      "epoch": 1.8413156376226198,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004669937270253214,
      "loss": 5.9854,
      "mean_token_accuracy": 0.17471616715192795,
      "num_tokens": 40434561.0,
      "step": 15955
    },
    {
      "entropy": 6.124910688400268,
      "epoch": 1.8418926716676283,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000466972208094949,
      "loss": 5.8828,
      "mean_token_accuracy": 0.17960798293352126,
      "num_tokens": 40447065.0,
      "step": 15960
    },
    {
      "entropy": 6.188348245620728,
      "epoch": 1.8424697057126371,
      "grad_norm": 1.125,
      "learning_rate": 0.00046695068270768665,
      "loss": 5.9996,
      "mean_token_accuracy": 0.16989807039499283,
      "num_tokens": 40460441.0,
      "step": 15965
    },
    {
      "entropy": 6.174532318115235,
      "epoch": 1.8430467397576455,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046692915086425846,
      "loss": 5.9807,
      "mean_token_accuracy": 0.1772579774260521,
      "num_tokens": 40473381.0,
      "step": 15970
    },
    {
      "entropy": 6.231026363372803,
      "epoch": 1.8436237738026544,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046690761256538857,
      "loss": 6.0313,
      "mean_token_accuracy": 0.17102903723716736,
      "num_tokens": 40486346.0,
      "step": 15975
    },
    {
      "entropy": 6.098371887207032,
      "epoch": 1.844200807847663,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004668860678118015,
      "loss": 5.8672,
      "mean_token_accuracy": 0.17897507697343826,
      "num_tokens": 40497660.0,
      "step": 15980
    },
    {
      "entropy": 6.24845781326294,
      "epoch": 1.8447778418926717,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046686451660422185,
      "loss": 6.0393,
      "mean_token_accuracy": 0.1717912584543228,
      "num_tokens": 40510351.0,
      "step": 15985
    },
    {
      "entropy": 6.221736192703247,
      "epoch": 1.8453548759376803,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046684295894337455,
      "loss": 5.9571,
      "mean_token_accuracy": 0.1732628121972084,
      "num_tokens": 40524051.0,
      "step": 15990
    },
    {
      "entropy": 6.100642919540405,
      "epoch": 1.845931909982689,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004668213948299845,
      "loss": 5.9538,
      "mean_token_accuracy": 0.17815927267074586,
      "num_tokens": 40536502.0,
      "step": 15995
    },
    {
      "entropy": 6.198820400238037,
      "epoch": 1.8465089440276976,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004667998242647772,
      "loss": 5.9157,
      "mean_token_accuracy": 0.1729410320520401,
      "num_tokens": 40549383.0,
      "step": 16000
    },
    {
      "entropy": 6.182983350753784,
      "epoch": 1.8470859780727062,
      "grad_norm": 1.09375,
      "learning_rate": 0.000466778247248478,
      "loss": 5.9766,
      "mean_token_accuracy": 0.17112861424684525,
      "num_tokens": 40562125.0,
      "step": 16005
    },
    {
      "entropy": 6.132656097412109,
      "epoch": 1.847663012117715,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046675666378181275,
      "loss": 6.0265,
      "mean_token_accuracy": 0.1694214567542076,
      "num_tokens": 40576293.0,
      "step": 16010
    },
    {
      "entropy": 6.145465326309204,
      "epoch": 1.8482400461627235,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004667350738655074,
      "loss": 5.8876,
      "mean_token_accuracy": 0.1753913402557373,
      "num_tokens": 40589009.0,
      "step": 16015
    },
    {
      "entropy": 6.318290424346924,
      "epoch": 1.8488170802077324,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046671347750028806,
      "loss": 6.0442,
      "mean_token_accuracy": 0.16501498222351074,
      "num_tokens": 40601826.0,
      "step": 16020
    },
    {
      "entropy": 6.224329137802124,
      "epoch": 1.8493941142527408,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004666918746868811,
      "loss": 6.0172,
      "mean_token_accuracy": 0.17779606431722642,
      "num_tokens": 40613453.0,
      "step": 16025
    },
    {
      "entropy": 6.127078628540039,
      "epoch": 1.8499711482977497,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004666702654260133,
      "loss": 5.9601,
      "mean_token_accuracy": 0.16933163106441498,
      "num_tokens": 40626205.0,
      "step": 16030
    },
    {
      "entropy": 6.186867809295654,
      "epoch": 1.850548182342758,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046664864971841113,
      "loss": 5.8765,
      "mean_token_accuracy": 0.17959264069795608,
      "num_tokens": 40637853.0,
      "step": 16035
    },
    {
      "entropy": 6.1430762767791744,
      "epoch": 1.851125216387767,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046662702756480195,
      "loss": 5.9732,
      "mean_token_accuracy": 0.17338251173496247,
      "num_tokens": 40651059.0,
      "step": 16040
    },
    {
      "entropy": 6.171749830245972,
      "epoch": 1.8517022504327756,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046660539896591286,
      "loss": 6.0049,
      "mean_token_accuracy": 0.16729219853878022,
      "num_tokens": 40663434.0,
      "step": 16045
    },
    {
      "entropy": 6.188680171966553,
      "epoch": 1.8522792844777842,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004665837639224713,
      "loss": 5.9412,
      "mean_token_accuracy": 0.17630462646484374,
      "num_tokens": 40675448.0,
      "step": 16050
    },
    {
      "entropy": 6.2071984767913815,
      "epoch": 1.8528563185227929,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046656212243520505,
      "loss": 6.034,
      "mean_token_accuracy": 0.17146788984537126,
      "num_tokens": 40687203.0,
      "step": 16055
    },
    {
      "entropy": 6.226963758468628,
      "epoch": 1.8534333525678015,
      "grad_norm": 1.125,
      "learning_rate": 0.00046654047450484193,
      "loss": 5.9936,
      "mean_token_accuracy": 0.17073075026273726,
      "num_tokens": 40700057.0,
      "step": 16060
    },
    {
      "entropy": 6.183546304702759,
      "epoch": 1.8540103866128101,
      "grad_norm": 1.125,
      "learning_rate": 0.0004665188201321102,
      "loss": 6.039,
      "mean_token_accuracy": 0.16928800791502,
      "num_tokens": 40714252.0,
      "step": 16065
    },
    {
      "entropy": 6.1210003852844235,
      "epoch": 1.8545874206578188,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046649715931773795,
      "loss": 5.8985,
      "mean_token_accuracy": 0.18124406933784484,
      "num_tokens": 40727666.0,
      "step": 16070
    },
    {
      "entropy": 6.182778406143188,
      "epoch": 1.8551644547028274,
      "grad_norm": 1.15625,
      "learning_rate": 0.000466475492062454,
      "loss": 5.9239,
      "mean_token_accuracy": 0.18020354807376862,
      "num_tokens": 40740340.0,
      "step": 16075
    },
    {
      "entropy": 6.172518062591553,
      "epoch": 1.855741488747836,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046645381836698684,
      "loss": 5.8956,
      "mean_token_accuracy": 0.18397375345230102,
      "num_tokens": 40752275.0,
      "step": 16080
    },
    {
      "entropy": 6.164517211914062,
      "epoch": 1.856318522792845,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004664321382320657,
      "loss": 5.9992,
      "mean_token_accuracy": 0.1715526357293129,
      "num_tokens": 40765163.0,
      "step": 16085
    },
    {
      "entropy": 6.233036708831787,
      "epoch": 1.8568955568378533,
      "grad_norm": 1.578125,
      "learning_rate": 0.00046641045165841965,
      "loss": 5.9803,
      "mean_token_accuracy": 0.16842788308858872,
      "num_tokens": 40778306.0,
      "step": 16090
    },
    {
      "entropy": 6.210261678695678,
      "epoch": 1.8574725908828622,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046638875864677814,
      "loss": 5.9735,
      "mean_token_accuracy": 0.16739570796489717,
      "num_tokens": 40791409.0,
      "step": 16095
    },
    {
      "entropy": 6.180561685562134,
      "epoch": 1.8580496249278706,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004663670591978708,
      "loss": 5.9969,
      "mean_token_accuracy": 0.17395004779100418,
      "num_tokens": 40803899.0,
      "step": 16100
    },
    {
      "entropy": 6.112685298919677,
      "epoch": 1.8586266589728795,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004663453533124275,
      "loss": 5.9441,
      "mean_token_accuracy": 0.18069510757923127,
      "num_tokens": 40815796.0,
      "step": 16105
    },
    {
      "entropy": 6.170335197448731,
      "epoch": 1.859203693017888,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004663236409911783,
      "loss": 5.9028,
      "mean_token_accuracy": 0.18205611854791642,
      "num_tokens": 40829275.0,
      "step": 16110
    },
    {
      "entropy": 6.15189380645752,
      "epoch": 1.8597807270628968,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046630192223485345,
      "loss": 5.9457,
      "mean_token_accuracy": 0.17662346512079238,
      "num_tokens": 40841582.0,
      "step": 16115
    },
    {
      "entropy": 6.176106643676758,
      "epoch": 1.8603577611079054,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046628019704418345,
      "loss": 5.9109,
      "mean_token_accuracy": 0.17581360340118407,
      "num_tokens": 40853416.0,
      "step": 16120
    },
    {
      "entropy": 6.027369403839112,
      "epoch": 1.860934795152914,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000466258465419899,
      "loss": 5.7598,
      "mean_token_accuracy": 0.19171839207410812,
      "num_tokens": 40865728.0,
      "step": 16125
    },
    {
      "entropy": 6.148682498931885,
      "epoch": 1.8615118291979227,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004662367273627312,
      "loss": 5.9338,
      "mean_token_accuracy": 0.17275214791297913,
      "num_tokens": 40878222.0,
      "step": 16130
    },
    {
      "entropy": 6.234012603759766,
      "epoch": 1.8620888632429313,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000466214982873411,
      "loss": 6.0253,
      "mean_token_accuracy": 0.17280251532793045,
      "num_tokens": 40890238.0,
      "step": 16135
    },
    {
      "entropy": 6.209517860412598,
      "epoch": 1.86266589728794,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046619323195266975,
      "loss": 6.0477,
      "mean_token_accuracy": 0.17480884939432145,
      "num_tokens": 40903258.0,
      "step": 16140
    },
    {
      "entropy": 6.2364787578582765,
      "epoch": 1.8632429313329486,
      "grad_norm": 1.125,
      "learning_rate": 0.0004661714746012392,
      "loss": 5.9369,
      "mean_token_accuracy": 0.16862312704324722,
      "num_tokens": 40916620.0,
      "step": 16145
    },
    {
      "entropy": 6.113634252548218,
      "epoch": 1.8638199653779572,
      "grad_norm": 1.359375,
      "learning_rate": 0.000466149710819851,
      "loss": 5.9635,
      "mean_token_accuracy": 0.1765235558152199,
      "num_tokens": 40928608.0,
      "step": 16150
    },
    {
      "entropy": 6.15068359375,
      "epoch": 1.8643969994229659,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004661279406092373,
      "loss": 5.9255,
      "mean_token_accuracy": 0.1762104719877243,
      "num_tokens": 40940543.0,
      "step": 16155
    },
    {
      "entropy": 6.191907548904419,
      "epoch": 1.8649740334679747,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046610616397013027,
      "loss": 5.9913,
      "mean_token_accuracy": 0.1703713282942772,
      "num_tokens": 40953774.0,
      "step": 16160
    },
    {
      "entropy": 6.234385967254639,
      "epoch": 1.8655510675129832,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004660843809032623,
      "loss": 5.9527,
      "mean_token_accuracy": 0.17811035066843034,
      "num_tokens": 40965673.0,
      "step": 16165
    },
    {
      "entropy": 6.077642965316772,
      "epoch": 1.866128101557992,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004660625914093661,
      "loss": 5.881,
      "mean_token_accuracy": 0.17766595631837845,
      "num_tokens": 40979301.0,
      "step": 16170
    },
    {
      "entropy": 6.22216911315918,
      "epoch": 1.8667051356030004,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004660407954891745,
      "loss": 5.9462,
      "mean_token_accuracy": 0.1772797077894211,
      "num_tokens": 40991475.0,
      "step": 16175
    },
    {
      "entropy": 6.168555974960327,
      "epoch": 1.8672821696480093,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004660189931434207,
      "loss": 5.9799,
      "mean_token_accuracy": 0.1704336002469063,
      "num_tokens": 41005592.0,
      "step": 16180
    },
    {
      "entropy": 6.19687786102295,
      "epoch": 1.8678592036930177,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004659971843728379,
      "loss": 5.9664,
      "mean_token_accuracy": 0.16889609545469284,
      "num_tokens": 41017812.0,
      "step": 16185
    },
    {
      "entropy": 6.209896087646484,
      "epoch": 1.8684362377380266,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004659753691781597,
      "loss": 5.9521,
      "mean_token_accuracy": 0.18006757497787476,
      "num_tokens": 41029560.0,
      "step": 16190
    },
    {
      "entropy": 6.177838373184204,
      "epoch": 1.8690132717830352,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004659535475601198,
      "loss": 5.9454,
      "mean_token_accuracy": 0.17087184488773347,
      "num_tokens": 41043219.0,
      "step": 16195
    },
    {
      "entropy": 6.140656757354736,
      "epoch": 1.8695903058280439,
      "grad_norm": 1.484375,
      "learning_rate": 0.00046593171951945226,
      "loss": 5.8641,
      "mean_token_accuracy": 0.17858447432518004,
      "num_tokens": 41056314.0,
      "step": 16200
    },
    {
      "entropy": 6.19299349784851,
      "epoch": 1.8701673398730525,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046590988505689114,
      "loss": 5.9273,
      "mean_token_accuracy": 0.17550511360168458,
      "num_tokens": 41068760.0,
      "step": 16205
    },
    {
      "entropy": 6.167492961883545,
      "epoch": 1.8707443739180611,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046588804417317084,
      "loss": 5.9695,
      "mean_token_accuracy": 0.1799878180027008,
      "num_tokens": 41080867.0,
      "step": 16210
    },
    {
      "entropy": 6.0867242336273195,
      "epoch": 1.8713214079630698,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046586619686902597,
      "loss": 5.913,
      "mean_token_accuracy": 0.18001627773046494,
      "num_tokens": 41093573.0,
      "step": 16215
    },
    {
      "entropy": 6.13196496963501,
      "epoch": 1.8718984420080784,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046584434314519144,
      "loss": 5.8422,
      "mean_token_accuracy": 0.1831797555088997,
      "num_tokens": 41106381.0,
      "step": 16220
    },
    {
      "entropy": 6.2189610481262205,
      "epoch": 1.8724754760530873,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004658224830024022,
      "loss": 5.9183,
      "mean_token_accuracy": 0.18200977444648742,
      "num_tokens": 41119791.0,
      "step": 16225
    },
    {
      "entropy": 6.113285255432129,
      "epoch": 1.8730525100980957,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004658006164413935,
      "loss": 5.9609,
      "mean_token_accuracy": 0.16842600852251052,
      "num_tokens": 41132934.0,
      "step": 16230
    },
    {
      "entropy": 6.019165849685669,
      "epoch": 1.8736295441431046,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004657787434629009,
      "loss": 5.8091,
      "mean_token_accuracy": 0.18185718804597856,
      "num_tokens": 41144751.0,
      "step": 16235
    },
    {
      "entropy": 6.140448999404907,
      "epoch": 1.874206578188113,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00046575686406765993,
      "loss": 5.9227,
      "mean_token_accuracy": 0.1719075158238411,
      "num_tokens": 41157563.0,
      "step": 16240
    },
    {
      "entropy": 6.152461767196655,
      "epoch": 1.8747836122331218,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046573497825640664,
      "loss": 5.8967,
      "mean_token_accuracy": 0.18032123595476152,
      "num_tokens": 41170130.0,
      "step": 16245
    },
    {
      "entropy": 6.168968582153321,
      "epoch": 1.8753606462781303,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004657130860298771,
      "loss": 5.8969,
      "mean_token_accuracy": 0.1778804361820221,
      "num_tokens": 41182091.0,
      "step": 16250
    },
    {
      "entropy": 6.133495235443116,
      "epoch": 1.8759376803231391,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004656911873888077,
      "loss": 5.9168,
      "mean_token_accuracy": 0.17606219202280043,
      "num_tokens": 41194429.0,
      "step": 16255
    },
    {
      "entropy": 6.2173878192901615,
      "epoch": 1.8765147143681478,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004656692823339349,
      "loss": 5.9854,
      "mean_token_accuracy": 0.16420651823282242,
      "num_tokens": 41207732.0,
      "step": 16260
    },
    {
      "entropy": 6.2502669334411625,
      "epoch": 1.8770917484131564,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004656473708659955,
      "loss": 6.0303,
      "mean_token_accuracy": 0.1654380440711975,
      "num_tokens": 41220973.0,
      "step": 16265
    },
    {
      "entropy": 6.1933471202850345,
      "epoch": 1.877668782458165,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046562545298572646,
      "loss": 5.9925,
      "mean_token_accuracy": 0.17013400942087173,
      "num_tokens": 41232590.0,
      "step": 16270
    },
    {
      "entropy": 6.12314920425415,
      "epoch": 1.8782458165031737,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000465603528693865,
      "loss": 5.8899,
      "mean_token_accuracy": 0.1786983624100685,
      "num_tokens": 41244806.0,
      "step": 16275
    },
    {
      "entropy": 6.236997890472412,
      "epoch": 1.8788228505481823,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046558159799114853,
      "loss": 6.0272,
      "mean_token_accuracy": 0.16638400703668593,
      "num_tokens": 41257818.0,
      "step": 16280
    },
    {
      "entropy": 6.0872478008270265,
      "epoch": 1.879399884593191,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004655596608783147,
      "loss": 5.8242,
      "mean_token_accuracy": 0.1838735967874527,
      "num_tokens": 41270396.0,
      "step": 16285
    },
    {
      "entropy": 6.1313518524169925,
      "epoch": 1.8799769186381996,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046553771735610136,
      "loss": 5.9607,
      "mean_token_accuracy": 0.17496560364961625,
      "num_tokens": 41282647.0,
      "step": 16290
    },
    {
      "entropy": 6.18224835395813,
      "epoch": 1.8805539526832082,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004655157674252465,
      "loss": 5.9172,
      "mean_token_accuracy": 0.18396375477313995,
      "num_tokens": 41294995.0,
      "step": 16295
    },
    {
      "entropy": 6.158409690856933,
      "epoch": 1.881130986728217,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046549381108648843,
      "loss": 5.9932,
      "mean_token_accuracy": 0.17139537930488585,
      "num_tokens": 41307420.0,
      "step": 16300
    },
    {
      "entropy": 6.162729358673095,
      "epoch": 1.8817080207732255,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004654718483405656,
      "loss": 5.9355,
      "mean_token_accuracy": 0.17805388569831848,
      "num_tokens": 41320145.0,
      "step": 16305
    },
    {
      "entropy": 6.2361412525177,
      "epoch": 1.8822850548182344,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046544987918821685,
      "loss": 5.9926,
      "mean_token_accuracy": 0.17488617151975633,
      "num_tokens": 41331740.0,
      "step": 16310
    },
    {
      "entropy": 6.168468570709228,
      "epoch": 1.8828620888632428,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000465427903630181,
      "loss": 5.9755,
      "mean_token_accuracy": 0.16915089190006255,
      "num_tokens": 41344120.0,
      "step": 16315
    },
    {
      "entropy": 6.1352907657623295,
      "epoch": 1.8834391229082517,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004654059216671972,
      "loss": 5.8874,
      "mean_token_accuracy": 0.17766397893428804,
      "num_tokens": 41357272.0,
      "step": 16320
    },
    {
      "entropy": 6.189902353286743,
      "epoch": 1.88401615695326,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004653839333000048,
      "loss": 6.0291,
      "mean_token_accuracy": 0.16860764026641845,
      "num_tokens": 41370557.0,
      "step": 16325
    },
    {
      "entropy": 6.182813930511474,
      "epoch": 1.884593190998269,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046536193852934334,
      "loss": 5.9071,
      "mean_token_accuracy": 0.1765513151884079,
      "num_tokens": 41381678.0,
      "step": 16330
    },
    {
      "entropy": 6.20900149345398,
      "epoch": 1.8851702250432776,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046533993735595273,
      "loss": 5.9589,
      "mean_token_accuracy": 0.16868072897195815,
      "num_tokens": 41395025.0,
      "step": 16335
    },
    {
      "entropy": 6.253957605361938,
      "epoch": 1.8857472590882862,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046531792978057277,
      "loss": 5.9907,
      "mean_token_accuracy": 0.1724797859787941,
      "num_tokens": 41407159.0,
      "step": 16340
    },
    {
      "entropy": 6.121765041351319,
      "epoch": 1.8863242931332949,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004652959158039438,
      "loss": 5.8978,
      "mean_token_accuracy": 0.1762823984026909,
      "num_tokens": 41420120.0,
      "step": 16345
    },
    {
      "entropy": 6.150393295288086,
      "epoch": 1.8869013271783035,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004652738954268062,
      "loss": 5.9013,
      "mean_token_accuracy": 0.17812399864196776,
      "num_tokens": 41433637.0,
      "step": 16350
    },
    {
      "entropy": 6.205445194244385,
      "epoch": 1.8874783612233121,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046525186864990073,
      "loss": 6.0426,
      "mean_token_accuracy": 0.17360662370920182,
      "num_tokens": 41447547.0,
      "step": 16355
    },
    {
      "entropy": 6.127155447006226,
      "epoch": 1.8880553952683208,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046522983547396806,
      "loss": 5.8896,
      "mean_token_accuracy": 0.17929900735616683,
      "num_tokens": 41460906.0,
      "step": 16360
    },
    {
      "entropy": 6.155298233032227,
      "epoch": 1.8886324293133296,
      "grad_norm": 1.375,
      "learning_rate": 0.0004652077958997494,
      "loss": 5.941,
      "mean_token_accuracy": 0.1667128935456276,
      "num_tokens": 41473928.0,
      "step": 16365
    },
    {
      "entropy": 6.162806749343872,
      "epoch": 1.889209463358338,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046518574992798604,
      "loss": 6.0096,
      "mean_token_accuracy": 0.17328576743602753,
      "num_tokens": 41487201.0,
      "step": 16370
    },
    {
      "entropy": 6.163497114181519,
      "epoch": 1.889786497403347,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046516369755941945,
      "loss": 5.8721,
      "mean_token_accuracy": 0.17560300529003142,
      "num_tokens": 41498697.0,
      "step": 16375
    },
    {
      "entropy": 6.264917039871216,
      "epoch": 1.8903635314483553,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004651416387947914,
      "loss": 5.9829,
      "mean_token_accuracy": 0.16863669604063034,
      "num_tokens": 41510533.0,
      "step": 16380
    },
    {
      "entropy": 6.157265329360962,
      "epoch": 1.8909405654933642,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004651195736348437,
      "loss": 5.9176,
      "mean_token_accuracy": 0.17344749420881272,
      "num_tokens": 41522753.0,
      "step": 16385
    },
    {
      "entropy": 6.146238327026367,
      "epoch": 1.8915175995383726,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004650975020803186,
      "loss": 5.9228,
      "mean_token_accuracy": 0.17551787197589874,
      "num_tokens": 41534700.0,
      "step": 16390
    },
    {
      "entropy": 6.221690368652344,
      "epoch": 1.8920946335833815,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004650754241319584,
      "loss": 5.9359,
      "mean_token_accuracy": 0.17332299798727036,
      "num_tokens": 41548126.0,
      "step": 16395
    },
    {
      "entropy": 6.158237266540527,
      "epoch": 1.8926716676283901,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046505333979050573,
      "loss": 5.9252,
      "mean_token_accuracy": 0.1761897936463356,
      "num_tokens": 41560423.0,
      "step": 16400
    },
    {
      "entropy": 6.146600866317749,
      "epoch": 1.8932487016733988,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004650312490567035,
      "loss": 6.0169,
      "mean_token_accuracy": 0.16882299184799193,
      "num_tokens": 41573714.0,
      "step": 16405
    },
    {
      "entropy": 6.250226545333862,
      "epoch": 1.8938257357184074,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004650091519312945,
      "loss": 5.985,
      "mean_token_accuracy": 0.1710459277033806,
      "num_tokens": 41586785.0,
      "step": 16410
    },
    {
      "entropy": 6.212391519546509,
      "epoch": 1.894402769763416,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046498704841502203,
      "loss": 5.9826,
      "mean_token_accuracy": 0.17754384130239487,
      "num_tokens": 41599001.0,
      "step": 16415
    },
    {
      "entropy": 6.129633092880249,
      "epoch": 1.8949798038084247,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004649649385086296,
      "loss": 5.9911,
      "mean_token_accuracy": 0.1726169615983963,
      "num_tokens": 41611876.0,
      "step": 16420
    },
    {
      "entropy": 6.136781024932861,
      "epoch": 1.8955568378534333,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004649428222128608,
      "loss": 5.9232,
      "mean_token_accuracy": 0.17759547829627992,
      "num_tokens": 41624074.0,
      "step": 16425
    },
    {
      "entropy": 6.216894340515137,
      "epoch": 1.896133871898442,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046492069952845953,
      "loss": 6.0637,
      "mean_token_accuracy": 0.16873060911893845,
      "num_tokens": 41638063.0,
      "step": 16430
    },
    {
      "entropy": 6.204340267181396,
      "epoch": 1.8967109059434506,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046489857045617,
      "loss": 5.9292,
      "mean_token_accuracy": 0.18534722775220872,
      "num_tokens": 41651123.0,
      "step": 16435
    },
    {
      "entropy": 6.053315353393555,
      "epoch": 1.8972879399884595,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046487643499673625,
      "loss": 5.8301,
      "mean_token_accuracy": 0.18508904576301574,
      "num_tokens": 41664025.0,
      "step": 16440
    },
    {
      "entropy": 6.114342880249024,
      "epoch": 1.8978649740334679,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004648542931509029,
      "loss": 5.87,
      "mean_token_accuracy": 0.1821804463863373,
      "num_tokens": 41675740.0,
      "step": 16445
    },
    {
      "entropy": 6.208328580856323,
      "epoch": 1.8984420080784767,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004648321449194148,
      "loss": 5.9579,
      "mean_token_accuracy": 0.17101528793573378,
      "num_tokens": 41688103.0,
      "step": 16450
    },
    {
      "entropy": 6.221075868606567,
      "epoch": 1.8990190421234852,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046480999030301673,
      "loss": 5.9473,
      "mean_token_accuracy": 0.1706663742661476,
      "num_tokens": 41701569.0,
      "step": 16455
    },
    {
      "entropy": 6.136055183410645,
      "epoch": 1.899596076168494,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046478782930245395,
      "loss": 5.8933,
      "mean_token_accuracy": 0.17741233110427856,
      "num_tokens": 41713373.0,
      "step": 16460
    },
    {
      "entropy": 6.180948972702026,
      "epoch": 1.9001731102135024,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046476566191847176,
      "loss": 6.0279,
      "mean_token_accuracy": 0.16507464200258254,
      "num_tokens": 41725550.0,
      "step": 16465
    },
    {
      "entropy": 6.21638593673706,
      "epoch": 1.9007501442585113,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004647434881518159,
      "loss": 5.9342,
      "mean_token_accuracy": 0.18668182343244552,
      "num_tokens": 41739064.0,
      "step": 16470
    },
    {
      "entropy": 6.164031457901001,
      "epoch": 1.90132717830352,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046472130800323194,
      "loss": 6.0003,
      "mean_token_accuracy": 0.18097071051597596,
      "num_tokens": 41752614.0,
      "step": 16475
    },
    {
      "entropy": 6.176392889022827,
      "epoch": 1.9019042123485286,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004646991214734661,
      "loss": 6.0015,
      "mean_token_accuracy": 0.17321633100509642,
      "num_tokens": 41766458.0,
      "step": 16480
    },
    {
      "entropy": 6.07353196144104,
      "epoch": 1.9024812463935372,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004646769285632645,
      "loss": 5.8824,
      "mean_token_accuracy": 0.18171819746494294,
      "num_tokens": 41778742.0,
      "step": 16485
    },
    {
      "entropy": 6.243658351898193,
      "epoch": 1.9030582804385459,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004646547292733737,
      "loss": 6.0335,
      "mean_token_accuracy": 0.16751776933670043,
      "num_tokens": 41793498.0,
      "step": 16490
    },
    {
      "entropy": 6.248891162872314,
      "epoch": 1.9036353144835545,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004646325236045402,
      "loss": 5.9928,
      "mean_token_accuracy": 0.17542592585086822,
      "num_tokens": 41807042.0,
      "step": 16495
    },
    {
      "entropy": 6.229089212417603,
      "epoch": 1.9042123485285631,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000464610311557511,
      "loss": 5.9644,
      "mean_token_accuracy": 0.17464442998170854,
      "num_tokens": 41818197.0,
      "step": 16500
    },
    {
      "entropy": 6.204925489425659,
      "epoch": 1.904789382573572,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004645880931330331,
      "loss": 5.9889,
      "mean_token_accuracy": 0.17287486046552658,
      "num_tokens": 41830183.0,
      "step": 16505
    },
    {
      "entropy": 6.2538042068481445,
      "epoch": 1.9053664166185804,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004645658683318539,
      "loss": 6.0072,
      "mean_token_accuracy": 0.17453489005565642,
      "num_tokens": 41843626.0,
      "step": 16510
    },
    {
      "entropy": 6.245009708404541,
      "epoch": 1.9059434506635893,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004645436371547209,
      "loss": 6.0322,
      "mean_token_accuracy": 0.1722807765007019,
      "num_tokens": 41855047.0,
      "step": 16515
    },
    {
      "entropy": 6.197148561477661,
      "epoch": 1.9065204847085977,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046452139960238186,
      "loss": 6.0196,
      "mean_token_accuracy": 0.17005757242441177,
      "num_tokens": 41868480.0,
      "step": 16520
    },
    {
      "entropy": 6.2311110496521,
      "epoch": 1.9070975187536066,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046449915567558467,
      "loss": 5.9877,
      "mean_token_accuracy": 0.16795886904001237,
      "num_tokens": 41882267.0,
      "step": 16525
    },
    {
      "entropy": 6.211942434310913,
      "epoch": 1.907674552798615,
      "grad_norm": 1.0,
      "learning_rate": 0.0004644769053750775,
      "loss": 6.032,
      "mean_token_accuracy": 0.16821773946285248,
      "num_tokens": 41894270.0,
      "step": 16530
    },
    {
      "entropy": 6.100690603256226,
      "epoch": 1.9082515868436238,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004644546487016087,
      "loss": 5.9181,
      "mean_token_accuracy": 0.17819420993328094,
      "num_tokens": 41908283.0,
      "step": 16535
    },
    {
      "entropy": 6.283512449264526,
      "epoch": 1.9088286208886325,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046443238565592687,
      "loss": 6.0565,
      "mean_token_accuracy": 0.17055184096097947,
      "num_tokens": 41921411.0,
      "step": 16540
    },
    {
      "entropy": 6.154033470153808,
      "epoch": 1.9094056549336411,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046441011623878087,
      "loss": 5.8953,
      "mean_token_accuracy": 0.178169909119606,
      "num_tokens": 41933221.0,
      "step": 16545
    },
    {
      "entropy": 6.128041076660156,
      "epoch": 1.9099826889786498,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004643878404509197,
      "loss": 5.9575,
      "mean_token_accuracy": 0.17057888954877853,
      "num_tokens": 41946497.0,
      "step": 16550
    },
    {
      "entropy": 6.192527151107788,
      "epoch": 1.9105597230236584,
      "grad_norm": 1.0,
      "learning_rate": 0.00046436555829309264,
      "loss": 5.9762,
      "mean_token_accuracy": 0.16172671169042588,
      "num_tokens": 41959118.0,
      "step": 16555
    },
    {
      "entropy": 6.263062429428101,
      "epoch": 1.911136757068667,
      "grad_norm": 1.171875,
      "learning_rate": 0.000464343269766049,
      "loss": 5.9648,
      "mean_token_accuracy": 0.17740431874990464,
      "num_tokens": 41972185.0,
      "step": 16560
    },
    {
      "entropy": 6.181894779205322,
      "epoch": 1.9117137911136757,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046432097487053857,
      "loss": 5.9002,
      "mean_token_accuracy": 0.1754305511713028,
      "num_tokens": 41984841.0,
      "step": 16565
    },
    {
      "entropy": 6.087489128112793,
      "epoch": 1.9122908251586843,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046429867360731124,
      "loss": 5.9289,
      "mean_token_accuracy": 0.17117201238870622,
      "num_tokens": 41998462.0,
      "step": 16570
    },
    {
      "entropy": 6.1801276206970215,
      "epoch": 1.912867859203693,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046427636597711695,
      "loss": 5.947,
      "mean_token_accuracy": 0.1818233013153076,
      "num_tokens": 42011300.0,
      "step": 16575
    },
    {
      "entropy": 6.257336902618408,
      "epoch": 1.9134448932487018,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046425405198070624,
      "loss": 5.9422,
      "mean_token_accuracy": 0.1738502189517021,
      "num_tokens": 42022334.0,
      "step": 16580
    },
    {
      "entropy": 6.189742517471314,
      "epoch": 1.9140219272937102,
      "grad_norm": 1.46875,
      "learning_rate": 0.00046423173161882944,
      "loss": 5.9521,
      "mean_token_accuracy": 0.1748774304986,
      "num_tokens": 42035495.0,
      "step": 16585
    },
    {
      "entropy": 6.19982271194458,
      "epoch": 1.914598961338719,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046420940489223735,
      "loss": 5.9485,
      "mean_token_accuracy": 0.17269486635923387,
      "num_tokens": 42046836.0,
      "step": 16590
    },
    {
      "entropy": 6.128095483779907,
      "epoch": 1.9151759953837275,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004641870718016809,
      "loss": 5.8899,
      "mean_token_accuracy": 0.17374359518289567,
      "num_tokens": 42058828.0,
      "step": 16595
    },
    {
      "entropy": 6.15652322769165,
      "epoch": 1.9157530294287364,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004641647323479113,
      "loss": 5.9151,
      "mean_token_accuracy": 0.17926120460033418,
      "num_tokens": 42070549.0,
      "step": 16600
    },
    {
      "entropy": 6.154821062088013,
      "epoch": 1.9163300634737448,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004641423865316798,
      "loss": 5.9151,
      "mean_token_accuracy": 0.1757856637239456,
      "num_tokens": 42082328.0,
      "step": 16605
    },
    {
      "entropy": 6.130866384506225,
      "epoch": 1.9169070975187537,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004641200343537381,
      "loss": 5.8957,
      "mean_token_accuracy": 0.18188864290714263,
      "num_tokens": 42094765.0,
      "step": 16610
    },
    {
      "entropy": 6.201978778839111,
      "epoch": 1.9174841315637623,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000464097675814838,
      "loss": 5.9441,
      "mean_token_accuracy": 0.17383870780467986,
      "num_tokens": 42107418.0,
      "step": 16615
    },
    {
      "entropy": 6.201455163955688,
      "epoch": 1.918061165608771,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004640753109157316,
      "loss": 5.9718,
      "mean_token_accuracy": 0.17375909686088561,
      "num_tokens": 42120623.0,
      "step": 16620
    },
    {
      "entropy": 6.1200353622436525,
      "epoch": 1.9186381996537796,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046405293965717093,
      "loss": 5.8518,
      "mean_token_accuracy": 0.18187272250652314,
      "num_tokens": 42133168.0,
      "step": 16625
    },
    {
      "entropy": 6.079050827026367,
      "epoch": 1.9192152336987882,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004640305620399086,
      "loss": 5.8905,
      "mean_token_accuracy": 0.18242004811763762,
      "num_tokens": 42146512.0,
      "step": 16630
    },
    {
      "entropy": 6.140944862365723,
      "epoch": 1.9197922677437969,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004640081780646971,
      "loss": 5.9428,
      "mean_token_accuracy": 0.17827149033546447,
      "num_tokens": 42158815.0,
      "step": 16635
    },
    {
      "entropy": 6.2126874923706055,
      "epoch": 1.9203693017888055,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046398578773228954,
      "loss": 5.9451,
      "mean_token_accuracy": 0.17334950119256973,
      "num_tokens": 42170614.0,
      "step": 16640
    },
    {
      "entropy": 6.183828401565552,
      "epoch": 1.9209463358338144,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046396339104343886,
      "loss": 5.9364,
      "mean_token_accuracy": 0.18198630660772325,
      "num_tokens": 42183758.0,
      "step": 16645
    },
    {
      "entropy": 6.1541972160339355,
      "epoch": 1.9215233698788228,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004639409879988984,
      "loss": 5.958,
      "mean_token_accuracy": 0.17005266547203063,
      "num_tokens": 42196111.0,
      "step": 16650
    },
    {
      "entropy": 6.209176349639892,
      "epoch": 1.9221004039238316,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004639185785994216,
      "loss": 5.969,
      "mean_token_accuracy": 0.17703946232795714,
      "num_tokens": 42209245.0,
      "step": 16655
    },
    {
      "entropy": 6.243718910217285,
      "epoch": 1.92267743796884,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046389616284576226,
      "loss": 5.9744,
      "mean_token_accuracy": 0.17565379291772842,
      "num_tokens": 42222356.0,
      "step": 16660
    },
    {
      "entropy": 6.098089170455933,
      "epoch": 1.923254472013849,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046387374073867435,
      "loss": 5.8151,
      "mean_token_accuracy": 0.19478982388973237,
      "num_tokens": 42236469.0,
      "step": 16665
    },
    {
      "entropy": 6.232636117935181,
      "epoch": 1.9238315060588573,
      "grad_norm": 1.0,
      "learning_rate": 0.00046385131227891197,
      "loss": 5.9513,
      "mean_token_accuracy": 0.1737977683544159,
      "num_tokens": 42248794.0,
      "step": 16670
    },
    {
      "entropy": 6.090934419631958,
      "epoch": 1.9244085401038662,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004638288774672295,
      "loss": 5.8567,
      "mean_token_accuracy": 0.1897292897105217,
      "num_tokens": 42262099.0,
      "step": 16675
    },
    {
      "entropy": 6.045392465591431,
      "epoch": 1.9249855741488748,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004638064363043816,
      "loss": 5.8532,
      "mean_token_accuracy": 0.17716535031795502,
      "num_tokens": 42274319.0,
      "step": 16680
    },
    {
      "entropy": 6.104020452499389,
      "epoch": 1.9255626081938835,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046378398879112293,
      "loss": 5.8706,
      "mean_token_accuracy": 0.17311550080776214,
      "num_tokens": 42285029.0,
      "step": 16685
    },
    {
      "entropy": 6.15468544960022,
      "epoch": 1.9261396422388921,
      "grad_norm": 1.125,
      "learning_rate": 0.0004637615349282086,
      "loss": 5.9771,
      "mean_token_accuracy": 0.17958744913339614,
      "num_tokens": 42297909.0,
      "step": 16690
    },
    {
      "entropy": 6.196270847320557,
      "epoch": 1.9267166762839008,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004637390747163938,
      "loss": 5.9793,
      "mean_token_accuracy": 0.17086661159992217,
      "num_tokens": 42309918.0,
      "step": 16695
    },
    {
      "entropy": 6.194420623779297,
      "epoch": 1.9272937103289094,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004637166081564339,
      "loss": 5.9136,
      "mean_token_accuracy": 0.1764407902956009,
      "num_tokens": 42322112.0,
      "step": 16700
    },
    {
      "entropy": 6.077479267120362,
      "epoch": 1.927870744373918,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046369413524908473,
      "loss": 5.8291,
      "mean_token_accuracy": 0.183780674636364,
      "num_tokens": 42335074.0,
      "step": 16705
    },
    {
      "entropy": 6.106912660598755,
      "epoch": 1.9284477784189267,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046367165599510207,
      "loss": 5.9189,
      "mean_token_accuracy": 0.1804255098104477,
      "num_tokens": 42348760.0,
      "step": 16710
    },
    {
      "entropy": 6.198479795455933,
      "epoch": 1.9290248124639353,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004636491703952419,
      "loss": 5.9325,
      "mean_token_accuracy": 0.16738573014736174,
      "num_tokens": 42361745.0,
      "step": 16715
    },
    {
      "entropy": 6.190168809890747,
      "epoch": 1.9296018465089442,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046362667845026057,
      "loss": 5.9574,
      "mean_token_accuracy": 0.17301880568265915,
      "num_tokens": 42375267.0,
      "step": 16720
    },
    {
      "entropy": 6.1772236824035645,
      "epoch": 1.9301788805539526,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046360418016091467,
      "loss": 5.9347,
      "mean_token_accuracy": 0.1741398021578789,
      "num_tokens": 42387589.0,
      "step": 16725
    },
    {
      "entropy": 6.238862705230713,
      "epoch": 1.9307559145989615,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046358167552796085,
      "loss": 5.9669,
      "mean_token_accuracy": 0.17430325001478195,
      "num_tokens": 42400144.0,
      "step": 16730
    },
    {
      "entropy": 6.183186388015747,
      "epoch": 1.9313329486439699,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046355916455215597,
      "loss": 5.983,
      "mean_token_accuracy": 0.16701369285583495,
      "num_tokens": 42412529.0,
      "step": 16735
    },
    {
      "entropy": 6.18750491142273,
      "epoch": 1.9319099826889787,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004635366472342573,
      "loss": 5.9602,
      "mean_token_accuracy": 0.1751283138990402,
      "num_tokens": 42424155.0,
      "step": 16740
    },
    {
      "entropy": 6.1764912605285645,
      "epoch": 1.9324870167339872,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004635141235750222,
      "loss": 5.8834,
      "mean_token_accuracy": 0.17514799833297728,
      "num_tokens": 42436299.0,
      "step": 16745
    },
    {
      "entropy": 6.095181894302368,
      "epoch": 1.933064050778996,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046349159357520815,
      "loss": 5.9456,
      "mean_token_accuracy": 0.1849423885345459,
      "num_tokens": 42450059.0,
      "step": 16750
    },
    {
      "entropy": 6.182287216186523,
      "epoch": 1.9336410848240047,
      "grad_norm": 1.078125,
      "learning_rate": 0.000463469057235573,
      "loss": 5.9222,
      "mean_token_accuracy": 0.17577754557132722,
      "num_tokens": 42463403.0,
      "step": 16755
    },
    {
      "entropy": 6.233158302307129,
      "epoch": 1.9342181188690133,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046344651455687476,
      "loss": 5.9644,
      "mean_token_accuracy": 0.1747601881623268,
      "num_tokens": 42476115.0,
      "step": 16760
    },
    {
      "entropy": 6.182958126068115,
      "epoch": 1.934795152914022,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046342396553987155,
      "loss": 5.9878,
      "mean_token_accuracy": 0.1705252096056938,
      "num_tokens": 42487992.0,
      "step": 16765
    },
    {
      "entropy": 6.142816019058228,
      "epoch": 1.9353721869590306,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046340141018532186,
      "loss": 5.8597,
      "mean_token_accuracy": 0.17995950430631638,
      "num_tokens": 42499961.0,
      "step": 16770
    },
    {
      "entropy": 6.194499063491821,
      "epoch": 1.9359492210040392,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004633788484939843,
      "loss": 5.9805,
      "mean_token_accuracy": 0.17590505182743071,
      "num_tokens": 42511432.0,
      "step": 16775
    },
    {
      "entropy": 6.071878337860108,
      "epoch": 1.9365262550490479,
      "grad_norm": 1.0,
      "learning_rate": 0.00046335628046661776,
      "loss": 5.8718,
      "mean_token_accuracy": 0.17307664901018144,
      "num_tokens": 42523904.0,
      "step": 16780
    },
    {
      "entropy": 6.2160711765289305,
      "epoch": 1.9371032890940567,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046333370610398135,
      "loss": 5.9613,
      "mean_token_accuracy": 0.17100536972284316,
      "num_tokens": 42537625.0,
      "step": 16785
    },
    {
      "entropy": 6.137755012512207,
      "epoch": 1.9376803231390651,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004633111254068342,
      "loss": 5.9004,
      "mean_token_accuracy": 0.18004844188690186,
      "num_tokens": 42551374.0,
      "step": 16790
    },
    {
      "entropy": 6.028134107589722,
      "epoch": 1.938257357184074,
      "grad_norm": 1.0,
      "learning_rate": 0.0004632885383759359,
      "loss": 5.7924,
      "mean_token_accuracy": 0.20065075755119324,
      "num_tokens": 42564325.0,
      "step": 16795
    },
    {
      "entropy": 6.137521982192993,
      "epoch": 1.9388343912290824,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046326594501204624,
      "loss": 5.9114,
      "mean_token_accuracy": 0.17819872498512268,
      "num_tokens": 42576353.0,
      "step": 16800
    },
    {
      "entropy": 6.118079900741577,
      "epoch": 1.9394114252740913,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000463243345315925,
      "loss": 5.8025,
      "mean_token_accuracy": 0.18807695508003236,
      "num_tokens": 42589199.0,
      "step": 16805
    },
    {
      "entropy": 6.178468322753906,
      "epoch": 1.9399884593190997,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046322073928833224,
      "loss": 6.0088,
      "mean_token_accuracy": 0.17217219322919847,
      "num_tokens": 42602145.0,
      "step": 16810
    },
    {
      "entropy": 6.221983146667481,
      "epoch": 1.9405654933641086,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004631981269300285,
      "loss": 5.9454,
      "mean_token_accuracy": 0.178158763051033,
      "num_tokens": 42615234.0,
      "step": 16815
    },
    {
      "entropy": 6.160770893096924,
      "epoch": 1.9411425274091172,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004631755082417742,
      "loss": 5.9128,
      "mean_token_accuracy": 0.1780958205461502,
      "num_tokens": 42626515.0,
      "step": 16820
    },
    {
      "entropy": 6.174253511428833,
      "epoch": 1.9417195614541258,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004631528832243302,
      "loss": 5.9021,
      "mean_token_accuracy": 0.18645912706851958,
      "num_tokens": 42638217.0,
      "step": 16825
    },
    {
      "entropy": 6.094896125793457,
      "epoch": 1.9422965954991345,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046313025187845735,
      "loss": 5.8707,
      "mean_token_accuracy": 0.17841373831033708,
      "num_tokens": 42650070.0,
      "step": 16830
    },
    {
      "entropy": 6.145742702484131,
      "epoch": 1.942873629544143,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046310761420491705,
      "loss": 5.9319,
      "mean_token_accuracy": 0.17956826388835906,
      "num_tokens": 42662849.0,
      "step": 16835
    },
    {
      "entropy": 6.14275050163269,
      "epoch": 1.9434506635891518,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004630849702044705,
      "loss": 5.917,
      "mean_token_accuracy": 0.17746737450361252,
      "num_tokens": 42676001.0,
      "step": 16840
    },
    {
      "entropy": 6.224986457824707,
      "epoch": 1.9440276976341604,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046306231987787937,
      "loss": 5.9169,
      "mean_token_accuracy": 0.17978755086660386,
      "num_tokens": 42688115.0,
      "step": 16845
    },
    {
      "entropy": 6.068338918685913,
      "epoch": 1.944604731679169,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046303966322590556,
      "loss": 5.8731,
      "mean_token_accuracy": 0.18306791484355928,
      "num_tokens": 42700123.0,
      "step": 16850
    },
    {
      "entropy": 6.214413022994995,
      "epoch": 1.9451817657241777,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004630170002493111,
      "loss": 6.0276,
      "mean_token_accuracy": 0.17338667064905167,
      "num_tokens": 42713942.0,
      "step": 16855
    },
    {
      "entropy": 6.180695724487305,
      "epoch": 1.9457587997691865,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046299433094885826,
      "loss": 5.9724,
      "mean_token_accuracy": 0.17321696877479553,
      "num_tokens": 42726596.0,
      "step": 16860
    },
    {
      "entropy": 6.252784490585327,
      "epoch": 1.946335833814195,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046297165532530944,
      "loss": 6.0228,
      "mean_token_accuracy": 0.16888666450977324,
      "num_tokens": 42738169.0,
      "step": 16865
    },
    {
      "entropy": 6.325664424896241,
      "epoch": 1.9469128678592038,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004629489733794274,
      "loss": 5.9675,
      "mean_token_accuracy": 0.17065389901399614,
      "num_tokens": 42751262.0,
      "step": 16870
    },
    {
      "entropy": 6.149333047866821,
      "epoch": 1.9474899019042122,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000462926285111975,
      "loss": 5.8931,
      "mean_token_accuracy": 0.17905056923627855,
      "num_tokens": 42763366.0,
      "step": 16875
    },
    {
      "entropy": 6.099828767776489,
      "epoch": 1.948066935949221,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046290359052371535,
      "loss": 5.9032,
      "mean_token_accuracy": 0.17817543298006058,
      "num_tokens": 42775801.0,
      "step": 16880
    },
    {
      "entropy": 6.253103494644165,
      "epoch": 1.9486439699942295,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004628808896154117,
      "loss": 5.9849,
      "mean_token_accuracy": 0.17391834259033204,
      "num_tokens": 42788610.0,
      "step": 16885
    },
    {
      "entropy": 6.271477079391479,
      "epoch": 1.9492210040392384,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004628581823878277,
      "loss": 6.0385,
      "mean_token_accuracy": 0.17721263468265533,
      "num_tokens": 42801517.0,
      "step": 16890
    },
    {
      "entropy": 6.157790231704712,
      "epoch": 1.949798038084247,
      "grad_norm": 1.140625,
      "learning_rate": 0.000462835468841727,
      "loss": 5.8881,
      "mean_token_accuracy": 0.17232766896486282,
      "num_tokens": 42814007.0,
      "step": 16895
    },
    {
      "entropy": 6.1054267406463625,
      "epoch": 1.9503750721292556,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004628127489778737,
      "loss": 5.7962,
      "mean_token_accuracy": 0.17750702500343324,
      "num_tokens": 42826960.0,
      "step": 16900
    },
    {
      "entropy": 6.1394459247589115,
      "epoch": 1.9509521061742643,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004627900227970318,
      "loss": 5.893,
      "mean_token_accuracy": 0.1754269227385521,
      "num_tokens": 42838950.0,
      "step": 16905
    },
    {
      "entropy": 6.155321359634399,
      "epoch": 1.951529140219273,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046276729029996576,
      "loss": 5.9663,
      "mean_token_accuracy": 0.17171598076820374,
      "num_tokens": 42853470.0,
      "step": 16910
    },
    {
      "entropy": 6.183119106292724,
      "epoch": 1.9521061742642816,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046274455148744025,
      "loss": 5.8519,
      "mean_token_accuracy": 0.1780314564704895,
      "num_tokens": 42865946.0,
      "step": 16915
    },
    {
      "entropy": 6.166663265228271,
      "epoch": 1.9526832083092902,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046272180636022,
      "loss": 6.0068,
      "mean_token_accuracy": 0.16525886952877045,
      "num_tokens": 42878248.0,
      "step": 16920
    },
    {
      "entropy": 6.115409708023071,
      "epoch": 1.953260242354299,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046269905491907,
      "loss": 5.8592,
      "mean_token_accuracy": 0.17854195535182954,
      "num_tokens": 42890750.0,
      "step": 16925
    },
    {
      "entropy": 6.077497720718384,
      "epoch": 1.9538372763993075,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004626762971647555,
      "loss": 5.8631,
      "mean_token_accuracy": 0.17530524879693984,
      "num_tokens": 42904737.0,
      "step": 16930
    },
    {
      "entropy": 6.197599220275879,
      "epoch": 1.9544143104443163,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046265353309804205,
      "loss": 5.9883,
      "mean_token_accuracy": 0.17158817648887634,
      "num_tokens": 42917374.0,
      "step": 16935
    },
    {
      "entropy": 6.1594137191772464,
      "epoch": 1.9549913444893248,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004626307627196952,
      "loss": 5.884,
      "mean_token_accuracy": 0.1758200407028198,
      "num_tokens": 42930068.0,
      "step": 16940
    },
    {
      "entropy": 6.209073162078857,
      "epoch": 1.9555683785343336,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004626079860304808,
      "loss": 6.0059,
      "mean_token_accuracy": 0.17115724682807923,
      "num_tokens": 42944014.0,
      "step": 16945
    },
    {
      "entropy": 6.179897308349609,
      "epoch": 1.956145412579342,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046258520303116496,
      "loss": 5.9366,
      "mean_token_accuracy": 0.16816047877073287,
      "num_tokens": 42956765.0,
      "step": 16950
    },
    {
      "entropy": 6.159287643432617,
      "epoch": 1.956722446624351,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004625624137225141,
      "loss": 5.9038,
      "mean_token_accuracy": 0.18344654738903046,
      "num_tokens": 42970661.0,
      "step": 16955
    },
    {
      "entropy": 6.117857265472412,
      "epoch": 1.9572994806693593,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004625396181052945,
      "loss": 5.9016,
      "mean_token_accuracy": 0.1764778360724449,
      "num_tokens": 42983359.0,
      "step": 16960
    },
    {
      "entropy": 6.217076301574707,
      "epoch": 1.9578765147143682,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046251681618027316,
      "loss": 6.002,
      "mean_token_accuracy": 0.17189738303422927,
      "num_tokens": 42995650.0,
      "step": 16965
    },
    {
      "entropy": 6.194344615936279,
      "epoch": 1.9584535487593768,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004624940079482167,
      "loss": 5.9653,
      "mean_token_accuracy": 0.17196790277957916,
      "num_tokens": 43009161.0,
      "step": 16970
    },
    {
      "entropy": 6.225995349884033,
      "epoch": 1.9590305828043855,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046247119340989254,
      "loss": 5.9577,
      "mean_token_accuracy": 0.17845648527145386,
      "num_tokens": 43022314.0,
      "step": 16975
    },
    {
      "entropy": 6.17625937461853,
      "epoch": 1.959607616849394,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004624483725660678,
      "loss": 5.8891,
      "mean_token_accuracy": 0.1830867514014244,
      "num_tokens": 43034458.0,
      "step": 16980
    },
    {
      "entropy": 6.090566635131836,
      "epoch": 1.9601846508944027,
      "grad_norm": 1.125,
      "learning_rate": 0.0004624255454175102,
      "loss": 5.8864,
      "mean_token_accuracy": 0.176594015955925,
      "num_tokens": 43047341.0,
      "step": 16985
    },
    {
      "entropy": 6.159727478027344,
      "epoch": 1.9607616849394114,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004624027119649875,
      "loss": 5.926,
      "mean_token_accuracy": 0.1786920428276062,
      "num_tokens": 43059363.0,
      "step": 16990
    },
    {
      "entropy": 6.152777481079101,
      "epoch": 1.96133871898442,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046237987220926766,
      "loss": 5.9021,
      "mean_token_accuracy": 0.1785392701625824,
      "num_tokens": 43070574.0,
      "step": 16995
    },
    {
      "entropy": 6.134773635864258,
      "epoch": 1.9619157530294289,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046235702615111886,
      "loss": 5.9459,
      "mean_token_accuracy": 0.1816381186246872,
      "num_tokens": 43082582.0,
      "step": 17000
    },
    {
      "entropy": 6.191130113601685,
      "epoch": 1.9624927870744373,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004623341737913096,
      "loss": 5.9773,
      "mean_token_accuracy": 0.17308419197797775,
      "num_tokens": 43096808.0,
      "step": 17005
    },
    {
      "entropy": 6.204410171508789,
      "epoch": 1.9630698211194462,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004623113151306085,
      "loss": 5.9926,
      "mean_token_accuracy": 0.1701839715242386,
      "num_tokens": 43108977.0,
      "step": 17010
    },
    {
      "entropy": 6.207643985748291,
      "epoch": 1.9636468551644546,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046228845016978425,
      "loss": 6.0832,
      "mean_token_accuracy": 0.16595102995634078,
      "num_tokens": 43121135.0,
      "step": 17015
    },
    {
      "entropy": 6.153323936462402,
      "epoch": 1.9642238892094634,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004622655789096061,
      "loss": 5.8775,
      "mean_token_accuracy": 0.18600623160600663,
      "num_tokens": 43132955.0,
      "step": 17020
    },
    {
      "entropy": 6.223880863189697,
      "epoch": 1.9648009232544719,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046224270135084315,
      "loss": 5.9037,
      "mean_token_accuracy": 0.17566201090812683,
      "num_tokens": 43145029.0,
      "step": 17025
    },
    {
      "entropy": 6.170547294616699,
      "epoch": 1.9653779572994807,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046221981749426503,
      "loss": 5.9251,
      "mean_token_accuracy": 0.1788855567574501,
      "num_tokens": 43157594.0,
      "step": 17030
    },
    {
      "entropy": 6.1581573486328125,
      "epoch": 1.9659549913444894,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046219692734064124,
      "loss": 5.9563,
      "mean_token_accuracy": 0.1711184337735176,
      "num_tokens": 43169922.0,
      "step": 17035
    },
    {
      "entropy": 6.223145818710327,
      "epoch": 1.966532025389498,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004621740308907419,
      "loss": 5.8886,
      "mean_token_accuracy": 0.17752826809883118,
      "num_tokens": 43183151.0,
      "step": 17040
    },
    {
      "entropy": 6.157525491714478,
      "epoch": 1.9671090594345066,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004621511281453369,
      "loss": 5.8509,
      "mean_token_accuracy": 0.17739802300930024,
      "num_tokens": 43195542.0,
      "step": 17045
    },
    {
      "entropy": 6.144079113006592,
      "epoch": 1.9676860934795153,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004621282191051967,
      "loss": 5.9255,
      "mean_token_accuracy": 0.1794307768344879,
      "num_tokens": 43208560.0,
      "step": 17050
    },
    {
      "entropy": 6.046969079971314,
      "epoch": 1.968263127524524,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004621053037710918,
      "loss": 5.7251,
      "mean_token_accuracy": 0.18884184211492538,
      "num_tokens": 43220959.0,
      "step": 17055
    },
    {
      "entropy": 6.097615051269531,
      "epoch": 1.9688401615695326,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000462082382143793,
      "loss": 5.9516,
      "mean_token_accuracy": 0.1726236328482628,
      "num_tokens": 43231458.0,
      "step": 17060
    },
    {
      "entropy": 6.196081924438476,
      "epoch": 1.9694171956145412,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046205945422407113,
      "loss": 5.889,
      "mean_token_accuracy": 0.17824373096227647,
      "num_tokens": 43243712.0,
      "step": 17065
    },
    {
      "entropy": 6.178509616851807,
      "epoch": 1.9699942296595498,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00046203652001269754,
      "loss": 5.988,
      "mean_token_accuracy": 0.1731117382645607,
      "num_tokens": 43256931.0,
      "step": 17070
    },
    {
      "entropy": 6.19140887260437,
      "epoch": 1.9705712637045587,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046201357951044337,
      "loss": 5.944,
      "mean_token_accuracy": 0.17455937564373017,
      "num_tokens": 43270420.0,
      "step": 17075
    },
    {
      "entropy": 6.276460886001587,
      "epoch": 1.9711482977495671,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046199063271808047,
      "loss": 5.9625,
      "mean_token_accuracy": 0.1733388841152191,
      "num_tokens": 43283183.0,
      "step": 17080
    },
    {
      "entropy": 6.080477094650268,
      "epoch": 1.971725331794576,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004619676796363804,
      "loss": 5.7994,
      "mean_token_accuracy": 0.19623544067144394,
      "num_tokens": 43295608.0,
      "step": 17085
    },
    {
      "entropy": 6.103295993804932,
      "epoch": 1.9723023658395844,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046194472026611546,
      "loss": 5.9407,
      "mean_token_accuracy": 0.1715902030467987,
      "num_tokens": 43307189.0,
      "step": 17090
    },
    {
      "entropy": 6.142501640319824,
      "epoch": 1.9728793998845933,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004619217546080576,
      "loss": 5.9024,
      "mean_token_accuracy": 0.17907895445823668,
      "num_tokens": 43319729.0,
      "step": 17095
    },
    {
      "entropy": 6.082439613342285,
      "epoch": 1.9734564339296017,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004618987826629794,
      "loss": 5.8798,
      "mean_token_accuracy": 0.1844079613685608,
      "num_tokens": 43331481.0,
      "step": 17100
    },
    {
      "entropy": 6.09782304763794,
      "epoch": 1.9740334679746105,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046187580443165344,
      "loss": 5.9038,
      "mean_token_accuracy": 0.17753845006227492,
      "num_tokens": 43344815.0,
      "step": 17105
    },
    {
      "entropy": 6.274027395248413,
      "epoch": 1.9746105020196192,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004618528199148527,
      "loss": 6.0092,
      "mean_token_accuracy": 0.16852036714553834,
      "num_tokens": 43358825.0,
      "step": 17110
    },
    {
      "entropy": 6.183201646804809,
      "epoch": 1.9751875360646278,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046182982911335016,
      "loss": 5.9215,
      "mean_token_accuracy": 0.17296264320611954,
      "num_tokens": 43372536.0,
      "step": 17115
    },
    {
      "entropy": 6.19670991897583,
      "epoch": 1.9757645701096365,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046180683202791905,
      "loss": 5.977,
      "mean_token_accuracy": 0.16794750094413757,
      "num_tokens": 43384658.0,
      "step": 17120
    },
    {
      "entropy": 6.223017454147339,
      "epoch": 1.976341604154645,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000461783828659333,
      "loss": 5.9652,
      "mean_token_accuracy": 0.17475801110267639,
      "num_tokens": 43396532.0,
      "step": 17125
    },
    {
      "entropy": 6.251249027252197,
      "epoch": 1.9769186381996537,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046176081900836565,
      "loss": 6.0197,
      "mean_token_accuracy": 0.1667226955294609,
      "num_tokens": 43409811.0,
      "step": 17130
    },
    {
      "entropy": 6.257318830490112,
      "epoch": 1.9774956722446624,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046173780307579086,
      "loss": 5.9527,
      "mean_token_accuracy": 0.1660043865442276,
      "num_tokens": 43422835.0,
      "step": 17135
    },
    {
      "entropy": 6.176904964447021,
      "epoch": 1.9780727062896712,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004617147808623829,
      "loss": 5.9182,
      "mean_token_accuracy": 0.17998040169477464,
      "num_tokens": 43434418.0,
      "step": 17140
    },
    {
      "entropy": 6.181827783584595,
      "epoch": 1.9786497403346797,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046169175236891605,
      "loss": 5.913,
      "mean_token_accuracy": 0.17536397278308868,
      "num_tokens": 43447447.0,
      "step": 17145
    },
    {
      "entropy": 6.174407005310059,
      "epoch": 1.9792267743796885,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004616687175961648,
      "loss": 5.8986,
      "mean_token_accuracy": 0.17711858451366425,
      "num_tokens": 43461205.0,
      "step": 17150
    },
    {
      "entropy": 6.184445858001709,
      "epoch": 1.979803808424697,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004616456765449039,
      "loss": 5.9199,
      "mean_token_accuracy": 0.17315379083156585,
      "num_tokens": 43473838.0,
      "step": 17155
    },
    {
      "entropy": 6.137933015823364,
      "epoch": 1.9803808424697058,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046162262921590845,
      "loss": 5.9878,
      "mean_token_accuracy": 0.1697661504149437,
      "num_tokens": 43485661.0,
      "step": 17160
    },
    {
      "entropy": 6.210676288604736,
      "epoch": 1.9809578765147142,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004615995756099535,
      "loss": 5.9493,
      "mean_token_accuracy": 0.17100608646869658,
      "num_tokens": 43499219.0,
      "step": 17165
    },
    {
      "entropy": 6.203417730331421,
      "epoch": 1.981534910559723,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004615765157278146,
      "loss": 5.911,
      "mean_token_accuracy": 0.1754206046462059,
      "num_tokens": 43512588.0,
      "step": 17170
    },
    {
      "entropy": 6.211092948913574,
      "epoch": 1.9821119446047317,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046155344957026726,
      "loss": 5.9797,
      "mean_token_accuracy": 0.17614224404096604,
      "num_tokens": 43525419.0,
      "step": 17175
    },
    {
      "entropy": 6.1743371963500975,
      "epoch": 1.9826889786497404,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004615303771380873,
      "loss": 5.9234,
      "mean_token_accuracy": 0.1734701171517372,
      "num_tokens": 43540244.0,
      "step": 17180
    },
    {
      "entropy": 6.204787731170654,
      "epoch": 1.983266012694749,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046150729843205077,
      "loss": 5.9666,
      "mean_token_accuracy": 0.17416103333234786,
      "num_tokens": 43552108.0,
      "step": 17185
    },
    {
      "entropy": 6.212602233886718,
      "epoch": 1.9838430467397576,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046148421345293384,
      "loss": 5.9216,
      "mean_token_accuracy": 0.17334341555833815,
      "num_tokens": 43563146.0,
      "step": 17190
    },
    {
      "entropy": 6.111716985702515,
      "epoch": 1.9844200807847663,
      "grad_norm": 1.125,
      "learning_rate": 0.0004614611222015131,
      "loss": 5.9191,
      "mean_token_accuracy": 0.17702849358320236,
      "num_tokens": 43577188.0,
      "step": 17195
    },
    {
      "entropy": 6.144491195678711,
      "epoch": 1.984997114829775,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046143802467856507,
      "loss": 5.8723,
      "mean_token_accuracy": 0.18262009471654891,
      "num_tokens": 43590366.0,
      "step": 17200
    },
    {
      "entropy": 6.17120041847229,
      "epoch": 1.9855741488747836,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046141492088486673,
      "loss": 5.9594,
      "mean_token_accuracy": 0.1693733289837837,
      "num_tokens": 43603830.0,
      "step": 17205
    },
    {
      "entropy": 6.050378942489624,
      "epoch": 1.9861511829197922,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004613918108211951,
      "loss": 5.7823,
      "mean_token_accuracy": 0.18310095369815826,
      "num_tokens": 43617730.0,
      "step": 17210
    },
    {
      "entropy": 6.2308698177337645,
      "epoch": 1.986728216964801,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004613686944883275,
      "loss": 5.9689,
      "mean_token_accuracy": 0.1691381812095642,
      "num_tokens": 43630420.0,
      "step": 17215
    },
    {
      "entropy": 6.210497236251831,
      "epoch": 1.9873052510098095,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046134557188704146,
      "loss": 5.9738,
      "mean_token_accuracy": 0.17556350380182267,
      "num_tokens": 43643089.0,
      "step": 17220
    },
    {
      "entropy": 6.154446697235107,
      "epoch": 1.9878822850548183,
      "grad_norm": 1.609375,
      "learning_rate": 0.00046132244301811466,
      "loss": 5.9741,
      "mean_token_accuracy": 0.17790891826152802,
      "num_tokens": 43656803.0,
      "step": 17225
    },
    {
      "entropy": 6.1573426723480225,
      "epoch": 1.9884593190998268,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046129930788232497,
      "loss": 5.8574,
      "mean_token_accuracy": 0.1861798122525215,
      "num_tokens": 43669163.0,
      "step": 17230
    },
    {
      "entropy": 6.191298723220825,
      "epoch": 1.9890363531448356,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046127616648045073,
      "loss": 5.9648,
      "mean_token_accuracy": 0.1757996752858162,
      "num_tokens": 43680724.0,
      "step": 17235
    },
    {
      "entropy": 6.214610624313354,
      "epoch": 1.989613387189844,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046125301881327007,
      "loss": 6.0189,
      "mean_token_accuracy": 0.17464605271816253,
      "num_tokens": 43694823.0,
      "step": 17240
    },
    {
      "entropy": 6.20328917503357,
      "epoch": 1.990190421234853,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046122986488156167,
      "loss": 5.9637,
      "mean_token_accuracy": 0.17324412018060684,
      "num_tokens": 43707406.0,
      "step": 17245
    },
    {
      "entropy": 6.132642793655395,
      "epoch": 1.9907674552798615,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046120670468610426,
      "loss": 5.9278,
      "mean_token_accuracy": 0.1730557858943939,
      "num_tokens": 43719663.0,
      "step": 17250
    },
    {
      "entropy": 6.16251392364502,
      "epoch": 1.9913444893248702,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046118353822767683,
      "loss": 5.9001,
      "mean_token_accuracy": 0.1793742373585701,
      "num_tokens": 43732212.0,
      "step": 17255
    },
    {
      "entropy": 6.22993745803833,
      "epoch": 1.9919215233698788,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004611603655070586,
      "loss": 5.9599,
      "mean_token_accuracy": 0.17345501035451888,
      "num_tokens": 43743686.0,
      "step": 17260
    },
    {
      "entropy": 6.104994010925293,
      "epoch": 1.9924985574148875,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046113718652502896,
      "loss": 5.8469,
      "mean_token_accuracy": 0.18492385149002075,
      "num_tokens": 43756578.0,
      "step": 17265
    },
    {
      "entropy": 6.171674585342407,
      "epoch": 1.993075591459896,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004611140012823675,
      "loss": 5.9855,
      "mean_token_accuracy": 0.1713765248656273,
      "num_tokens": 43770054.0,
      "step": 17270
    },
    {
      "entropy": 6.196346759796143,
      "epoch": 1.9936526255049047,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046109080977985395,
      "loss": 5.8591,
      "mean_token_accuracy": 0.18892365843057632,
      "num_tokens": 43782669.0,
      "step": 17275
    },
    {
      "entropy": 6.102440118789673,
      "epoch": 1.9942296595499136,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046106761201826854,
      "loss": 5.8437,
      "mean_token_accuracy": 0.1779232069849968,
      "num_tokens": 43795330.0,
      "step": 17280
    },
    {
      "entropy": 6.150694417953491,
      "epoch": 1.994806693594922,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046104440799839145,
      "loss": 5.9148,
      "mean_token_accuracy": 0.18312762677669525,
      "num_tokens": 43808380.0,
      "step": 17285
    },
    {
      "entropy": 6.215299272537232,
      "epoch": 1.9953837276399309,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000461021197721003,
      "loss": 5.9485,
      "mean_token_accuracy": 0.17506540715694427,
      "num_tokens": 43821346.0,
      "step": 17290
    },
    {
      "entropy": 6.132453584671021,
      "epoch": 1.9959607616849393,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046099798118688407,
      "loss": 5.9036,
      "mean_token_accuracy": 0.1724586859345436,
      "num_tokens": 43832989.0,
      "step": 17295
    },
    {
      "entropy": 6.185548782348633,
      "epoch": 1.9965377957299482,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004609747583968154,
      "loss": 5.9519,
      "mean_token_accuracy": 0.18089664727449417,
      "num_tokens": 43846607.0,
      "step": 17300
    },
    {
      "entropy": 6.105967426300049,
      "epoch": 1.9971148297749566,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004609515293515781,
      "loss": 5.8484,
      "mean_token_accuracy": 0.18329283446073533,
      "num_tokens": 43860244.0,
      "step": 17305
    },
    {
      "entropy": 6.129933452606201,
      "epoch": 1.9976918638199654,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004609282940519535,
      "loss": 6.0058,
      "mean_token_accuracy": 0.17611169964075088,
      "num_tokens": 43874243.0,
      "step": 17310
    },
    {
      "entropy": 6.097413730621338,
      "epoch": 1.998268897864974,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004609050524987231,
      "loss": 5.8545,
      "mean_token_accuracy": 0.18297624588012695,
      "num_tokens": 43887241.0,
      "step": 17315
    },
    {
      "entropy": 6.189095592498779,
      "epoch": 1.9988459319099827,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004608818046926686,
      "loss": 5.9222,
      "mean_token_accuracy": 0.17636662870645523,
      "num_tokens": 43901006.0,
      "step": 17320
    },
    {
      "entropy": 6.178951740264893,
      "epoch": 1.9994229659549914,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004608585506345719,
      "loss": 5.9286,
      "mean_token_accuracy": 0.17217187881469725,
      "num_tokens": 43913167.0,
      "step": 17325
    },
    {
      "entropy": 6.19399209022522,
      "epoch": 2.0,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004608352903252152,
      "loss": 5.9629,
      "mean_token_accuracy": 0.1758102521300316,
      "num_tokens": 43926234.0,
      "step": 17330
    },
    {
      "entropy": 6.110502243041992,
      "epoch": 2.000577034045009,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046081202376538084,
      "loss": 5.7721,
      "mean_token_accuracy": 0.1888761505484581,
      "num_tokens": 43939047.0,
      "step": 17335
    },
    {
      "entropy": 6.070292139053345,
      "epoch": 2.0011540680900173,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004607887509558513,
      "loss": 5.7957,
      "mean_token_accuracy": 0.18005116134881974,
      "num_tokens": 43951649.0,
      "step": 17340
    },
    {
      "entropy": 6.158303880691529,
      "epoch": 2.001731102135026,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004607654718974094,
      "loss": 5.841,
      "mean_token_accuracy": 0.17674712091684341,
      "num_tokens": 43964140.0,
      "step": 17345
    },
    {
      "entropy": 6.15272364616394,
      "epoch": 2.0023081361800346,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004607421865908382,
      "loss": 5.8814,
      "mean_token_accuracy": 0.17471698969602584,
      "num_tokens": 43976189.0,
      "step": 17350
    },
    {
      "entropy": 6.27158031463623,
      "epoch": 2.0028851702250434,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004607188950369207,
      "loss": 5.9176,
      "mean_token_accuracy": 0.17445416748523712,
      "num_tokens": 43990920.0,
      "step": 17355
    },
    {
      "entropy": 6.096147918701172,
      "epoch": 2.003462204270052,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004606955972364405,
      "loss": 5.7605,
      "mean_token_accuracy": 0.18938583135604858,
      "num_tokens": 44003033.0,
      "step": 17360
    },
    {
      "entropy": 6.162179279327392,
      "epoch": 2.0040392383150607,
      "grad_norm": 1.25,
      "learning_rate": 0.0004606722931901812,
      "loss": 5.8741,
      "mean_token_accuracy": 0.17361028790473937,
      "num_tokens": 44014731.0,
      "step": 17365
    },
    {
      "entropy": 6.082789134979248,
      "epoch": 2.004616272360069,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004606489828989264,
      "loss": 5.7656,
      "mean_token_accuracy": 0.1834121584892273,
      "num_tokens": 44026076.0,
      "step": 17370
    },
    {
      "entropy": 6.183905601501465,
      "epoch": 2.005193306405078,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004606256663634604,
      "loss": 5.9072,
      "mean_token_accuracy": 0.17910184860229492,
      "num_tokens": 44038244.0,
      "step": 17375
    },
    {
      "entropy": 6.195360851287842,
      "epoch": 2.0057703404500864,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004606023435845672,
      "loss": 5.8568,
      "mean_token_accuracy": 0.17678214460611344,
      "num_tokens": 44051541.0,
      "step": 17380
    },
    {
      "entropy": 6.15230393409729,
      "epoch": 2.0063473744950953,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004605790145630314,
      "loss": 5.8045,
      "mean_token_accuracy": 0.18346837162971497,
      "num_tokens": 44064670.0,
      "step": 17385
    },
    {
      "entropy": 6.082640838623047,
      "epoch": 2.0069244085401037,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004605556792996377,
      "loss": 5.7076,
      "mean_token_accuracy": 0.1859280839562416,
      "num_tokens": 44077140.0,
      "step": 17390
    },
    {
      "entropy": 6.120267152786255,
      "epoch": 2.0075014425851125,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004605323377951709,
      "loss": 5.8799,
      "mean_token_accuracy": 0.19111145436763763,
      "num_tokens": 44091641.0,
      "step": 17395
    },
    {
      "entropy": 6.125492906570434,
      "epoch": 2.008078476630121,
      "grad_norm": 1.125,
      "learning_rate": 0.000460508990050416,
      "loss": 5.8501,
      "mean_token_accuracy": 0.18562736511230468,
      "num_tokens": 44104003.0,
      "step": 17400
    },
    {
      "entropy": 6.126090717315674,
      "epoch": 2.00865551067513,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004604856360661584,
      "loss": 5.898,
      "mean_token_accuracy": 0.17917969226837158,
      "num_tokens": 44116735.0,
      "step": 17405
    },
    {
      "entropy": 6.148071813583374,
      "epoch": 2.0092325447201387,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004604622758431835,
      "loss": 5.8441,
      "mean_token_accuracy": 0.18214959353208543,
      "num_tokens": 44129842.0,
      "step": 17410
    },
    {
      "entropy": 6.194177579879761,
      "epoch": 2.009809578765147,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046043890938227724,
      "loss": 5.8307,
      "mean_token_accuracy": 0.1797258213162422,
      "num_tokens": 44141800.0,
      "step": 17415
    },
    {
      "entropy": 6.105053043365478,
      "epoch": 2.010386612810156,
      "grad_norm": 1.25,
      "learning_rate": 0.00046041553668422526,
      "loss": 5.7951,
      "mean_token_accuracy": 0.1857082352042198,
      "num_tokens": 44155705.0,
      "step": 17420
    },
    {
      "entropy": 6.091993379592895,
      "epoch": 2.0109636468551644,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046039215774981376,
      "loss": 5.8056,
      "mean_token_accuracy": 0.18239107429981233,
      "num_tokens": 44167780.0,
      "step": 17425
    },
    {
      "entropy": 6.097388124465942,
      "epoch": 2.0115406809001732,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00046036877257982917,
      "loss": 5.8277,
      "mean_token_accuracy": 0.1817726120352745,
      "num_tokens": 44179990.0,
      "step": 17430
    },
    {
      "entropy": 6.168509197235108,
      "epoch": 2.0121177149451817,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000460345381175058,
      "loss": 5.8447,
      "mean_token_accuracy": 0.1802748665213585,
      "num_tokens": 44192608.0,
      "step": 17435
    },
    {
      "entropy": 6.116131973266602,
      "epoch": 2.0126947489901905,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000460321983536287,
      "loss": 5.8098,
      "mean_token_accuracy": 0.18671364188194275,
      "num_tokens": 44206017.0,
      "step": 17440
    },
    {
      "entropy": 6.085149049758911,
      "epoch": 2.013271783035199,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046029857966430315,
      "loss": 5.7902,
      "mean_token_accuracy": 0.17645594626665115,
      "num_tokens": 44219028.0,
      "step": 17445
    },
    {
      "entropy": 6.176063776016235,
      "epoch": 2.013848817080208,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046027516955989364,
      "loss": 5.8655,
      "mean_token_accuracy": 0.1822313278913498,
      "num_tokens": 44230902.0,
      "step": 17450
    },
    {
      "entropy": 6.163855504989624,
      "epoch": 2.014425851125216,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046025175322384577,
      "loss": 5.8452,
      "mean_token_accuracy": 0.17778589874505996,
      "num_tokens": 44243108.0,
      "step": 17455
    },
    {
      "entropy": 6.158125209808349,
      "epoch": 2.015002885170225,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046022833065694727,
      "loss": 5.9337,
      "mean_token_accuracy": 0.1698328897356987,
      "num_tokens": 44255442.0,
      "step": 17460
    },
    {
      "entropy": 6.195490169525146,
      "epoch": 2.0155799192152335,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046020490185998575,
      "loss": 5.8477,
      "mean_token_accuracy": 0.18404364138841628,
      "num_tokens": 44269180.0,
      "step": 17465
    },
    {
      "entropy": 6.206870460510254,
      "epoch": 2.0161569532602424,
      "grad_norm": 1.375,
      "learning_rate": 0.0004601814668337495,
      "loss": 5.9163,
      "mean_token_accuracy": 0.17094495743513108,
      "num_tokens": 44280905.0,
      "step": 17470
    },
    {
      "entropy": 6.248182249069214,
      "epoch": 2.016733987305251,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046015802557902654,
      "loss": 5.882,
      "mean_token_accuracy": 0.17512987554073334,
      "num_tokens": 44292672.0,
      "step": 17475
    },
    {
      "entropy": 6.081629610061645,
      "epoch": 2.0173110213502596,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046013457809660537,
      "loss": 5.715,
      "mean_token_accuracy": 0.19308462589979172,
      "num_tokens": 44305605.0,
      "step": 17480
    },
    {
      "entropy": 6.119570970535278,
      "epoch": 2.0178880553952685,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046011112438727454,
      "loss": 5.8761,
      "mean_token_accuracy": 0.17826148569583894,
      "num_tokens": 44318792.0,
      "step": 17485
    },
    {
      "entropy": 6.13537449836731,
      "epoch": 2.018465089440277,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004600876644518231,
      "loss": 5.8707,
      "mean_token_accuracy": 0.1799207717180252,
      "num_tokens": 44331878.0,
      "step": 17490
    },
    {
      "entropy": 6.228895378112793,
      "epoch": 2.0190421234852858,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046006419829104,
      "loss": 5.9785,
      "mean_token_accuracy": 0.17157808989286422,
      "num_tokens": 44345305.0,
      "step": 17495
    },
    {
      "entropy": 6.129715013504028,
      "epoch": 2.019619157530294,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004600407259057145,
      "loss": 5.7518,
      "mean_token_accuracy": 0.1825810268521309,
      "num_tokens": 44358512.0,
      "step": 17500
    },
    {
      "entropy": 6.124386501312256,
      "epoch": 2.020196191575303,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004600172472966361,
      "loss": 5.843,
      "mean_token_accuracy": 0.18949357271194459,
      "num_tokens": 44371139.0,
      "step": 17505
    },
    {
      "entropy": 6.155154466629028,
      "epoch": 2.0207732256203115,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045999376246459446,
      "loss": 5.8865,
      "mean_token_accuracy": 0.17255913764238356,
      "num_tokens": 44383984.0,
      "step": 17510
    },
    {
      "entropy": 6.150503158569336,
      "epoch": 2.0213502596653203,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004599702714103795,
      "loss": 5.8612,
      "mean_token_accuracy": 0.17950722724199294,
      "num_tokens": 44396481.0,
      "step": 17515
    },
    {
      "entropy": 6.150287532806397,
      "epoch": 2.0219272937103288,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004599467741347814,
      "loss": 5.8971,
      "mean_token_accuracy": 0.17643668949604036,
      "num_tokens": 44410843.0,
      "step": 17520
    },
    {
      "entropy": 6.198572492599487,
      "epoch": 2.0225043277553376,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004599232706385904,
      "loss": 5.8949,
      "mean_token_accuracy": 0.17484099119901658,
      "num_tokens": 44423797.0,
      "step": 17525
    },
    {
      "entropy": 6.114831209182739,
      "epoch": 2.023081361800346,
      "grad_norm": 1.265625,
      "learning_rate": 0.000459899760922597,
      "loss": 5.8632,
      "mean_token_accuracy": 0.1789395898580551,
      "num_tokens": 44436408.0,
      "step": 17530
    },
    {
      "entropy": 6.156760549545288,
      "epoch": 2.023658395845355,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004598762449875921,
      "loss": 5.7316,
      "mean_token_accuracy": 0.19193607270717622,
      "num_tokens": 44449145.0,
      "step": 17535
    },
    {
      "entropy": 6.159425687789917,
      "epoch": 2.0242354298903633,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004598527228343665,
      "loss": 5.8089,
      "mean_token_accuracy": 0.17869019508361816,
      "num_tokens": 44462945.0,
      "step": 17540
    },
    {
      "entropy": 6.0815025806427006,
      "epoch": 2.024812463935372,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004598291944637112,
      "loss": 5.8005,
      "mean_token_accuracy": 0.18434444665908814,
      "num_tokens": 44474478.0,
      "step": 17545
    },
    {
      "entropy": 6.020714664459229,
      "epoch": 2.025389497980381,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045980565987641797,
      "loss": 5.7064,
      "mean_token_accuracy": 0.19393896609544753,
      "num_tokens": 44488418.0,
      "step": 17550
    },
    {
      "entropy": 6.174957609176635,
      "epoch": 2.0259665320253895,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045978211907327804,
      "loss": 5.8581,
      "mean_token_accuracy": 0.17347276359796523,
      "num_tokens": 44501424.0,
      "step": 17555
    },
    {
      "entropy": 6.183877754211426,
      "epoch": 2.0265435660703983,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004597585720550834,
      "loss": 5.9212,
      "mean_token_accuracy": 0.1721937283873558,
      "num_tokens": 44513137.0,
      "step": 17560
    },
    {
      "entropy": 6.12046594619751,
      "epoch": 2.0271206001154067,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000459735018822626,
      "loss": 5.7734,
      "mean_token_accuracy": 0.185493803024292,
      "num_tokens": 44525195.0,
      "step": 17565
    },
    {
      "entropy": 6.1231849670410154,
      "epoch": 2.0276976341604156,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045971145937669794,
      "loss": 5.9356,
      "mean_token_accuracy": 0.17028947025537491,
      "num_tokens": 44537599.0,
      "step": 17570
    },
    {
      "entropy": 6.153174829483032,
      "epoch": 2.028274668205424,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004596878937180917,
      "loss": 5.8098,
      "mean_token_accuracy": 0.18010269552469255,
      "num_tokens": 44549384.0,
      "step": 17575
    },
    {
      "entropy": 6.119613361358643,
      "epoch": 2.028851702250433,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004596643218475999,
      "loss": 5.7462,
      "mean_token_accuracy": 0.18514081835746765,
      "num_tokens": 44561162.0,
      "step": 17580
    },
    {
      "entropy": 6.186729907989502,
      "epoch": 2.0294287362954413,
      "grad_norm": 0.96875,
      "learning_rate": 0.00045964074376601534,
      "loss": 5.8589,
      "mean_token_accuracy": 0.18353078067302703,
      "num_tokens": 44576047.0,
      "step": 17585
    },
    {
      "entropy": 6.0910969257354735,
      "epoch": 2.03000577034045,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045961715947413106,
      "loss": 5.8613,
      "mean_token_accuracy": 0.18219864070415498,
      "num_tokens": 44589506.0,
      "step": 17590
    },
    {
      "entropy": 6.074866437911988,
      "epoch": 2.0305828043854586,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004595935689727404,
      "loss": 5.7725,
      "mean_token_accuracy": 0.18342991024255753,
      "num_tokens": 44604483.0,
      "step": 17595
    },
    {
      "entropy": 6.20255675315857,
      "epoch": 2.0311598384304674,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004595699722626367,
      "loss": 5.8297,
      "mean_token_accuracy": 0.18816420435905457,
      "num_tokens": 44617696.0,
      "step": 17600
    },
    {
      "entropy": 6.025491380691529,
      "epoch": 2.031736872475476,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045954636934461367,
      "loss": 5.726,
      "mean_token_accuracy": 0.19219298660755157,
      "num_tokens": 44631482.0,
      "step": 17605
    },
    {
      "entropy": 6.09293384552002,
      "epoch": 2.0323139065204847,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004595227602194652,
      "loss": 5.7905,
      "mean_token_accuracy": 0.18800532668828965,
      "num_tokens": 44645544.0,
      "step": 17610
    },
    {
      "entropy": 6.135196304321289,
      "epoch": 2.0328909405654936,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004594991448879853,
      "loss": 5.8147,
      "mean_token_accuracy": 0.17980901151895523,
      "num_tokens": 44658869.0,
      "step": 17615
    },
    {
      "entropy": 6.19051718711853,
      "epoch": 2.033467974610502,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004594755233509683,
      "loss": 5.8961,
      "mean_token_accuracy": 0.17451203912496566,
      "num_tokens": 44670507.0,
      "step": 17620
    },
    {
      "entropy": 6.114893007278442,
      "epoch": 2.034045008655511,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045945189560920875,
      "loss": 5.797,
      "mean_token_accuracy": 0.18449529111385346,
      "num_tokens": 44684063.0,
      "step": 17625
    },
    {
      "entropy": 6.043971204757691,
      "epoch": 2.0346220427005193,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004594282616635013,
      "loss": 5.7139,
      "mean_token_accuracy": 0.20263297110795975,
      "num_tokens": 44698818.0,
      "step": 17630
    },
    {
      "entropy": 6.090926933288574,
      "epoch": 2.035199076745528,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004594046215146409,
      "loss": 5.8386,
      "mean_token_accuracy": 0.18151872605085373,
      "num_tokens": 44711401.0,
      "step": 17635
    },
    {
      "entropy": 6.154509210586548,
      "epoch": 2.0357761107905366,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045938097516342257,
      "loss": 5.8259,
      "mean_token_accuracy": 0.1761723890900612,
      "num_tokens": 44723141.0,
      "step": 17640
    },
    {
      "entropy": 6.243648052215576,
      "epoch": 2.0363531448355454,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00045935732261064184,
      "loss": 5.9187,
      "mean_token_accuracy": 0.1695309042930603,
      "num_tokens": 44736180.0,
      "step": 17645
    },
    {
      "entropy": 6.039850425720215,
      "epoch": 2.036930178880554,
      "grad_norm": 1.0,
      "learning_rate": 0.00045933366385709405,
      "loss": 5.8278,
      "mean_token_accuracy": 0.18564317524433135,
      "num_tokens": 44749453.0,
      "step": 17650
    },
    {
      "entropy": 6.130382680892945,
      "epoch": 2.0375072129255627,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004593099989035751,
      "loss": 5.8627,
      "mean_token_accuracy": 0.18117485493421553,
      "num_tokens": 44762609.0,
      "step": 17655
    },
    {
      "entropy": 6.134745645523071,
      "epoch": 2.038084246970571,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004592863277508809,
      "loss": 5.7454,
      "mean_token_accuracy": 0.18589025884866714,
      "num_tokens": 44776239.0,
      "step": 17660
    },
    {
      "entropy": 6.080679655075073,
      "epoch": 2.03866128101558,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004592626503998076,
      "loss": 5.7573,
      "mean_token_accuracy": 0.1886604368686676,
      "num_tokens": 44788979.0,
      "step": 17665
    },
    {
      "entropy": 6.180213356018067,
      "epoch": 2.0392383150605884,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004592389668511515,
      "loss": 5.9234,
      "mean_token_accuracy": 0.17439328879117966,
      "num_tokens": 44801715.0,
      "step": 17670
    },
    {
      "entropy": 6.099127578735351,
      "epoch": 2.0398153491055973,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004592152771057093,
      "loss": 5.8356,
      "mean_token_accuracy": 0.18264831155538558,
      "num_tokens": 44815004.0,
      "step": 17675
    },
    {
      "entropy": 6.233977270126343,
      "epoch": 2.0403923831506057,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045919158116427785,
      "loss": 5.9033,
      "mean_token_accuracy": 0.17557424455881118,
      "num_tokens": 44827933.0,
      "step": 17680
    },
    {
      "entropy": 6.150445508956909,
      "epoch": 2.0409694171956145,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045916787902765396,
      "loss": 5.8353,
      "mean_token_accuracy": 0.1891056105494499,
      "num_tokens": 44840967.0,
      "step": 17685
    },
    {
      "entropy": 6.123977994918823,
      "epoch": 2.0415464512406234,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004591441706966349,
      "loss": 5.8714,
      "mean_token_accuracy": 0.17447023689746857,
      "num_tokens": 44855456.0,
      "step": 17690
    },
    {
      "entropy": 6.060744714736939,
      "epoch": 2.042123485285632,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004591204561720183,
      "loss": 5.7303,
      "mean_token_accuracy": 0.1825903132557869,
      "num_tokens": 44867607.0,
      "step": 17695
    },
    {
      "entropy": 6.08973126411438,
      "epoch": 2.0427005193306407,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004590967354546015,
      "loss": 5.8125,
      "mean_token_accuracy": 0.1795862376689911,
      "num_tokens": 44880505.0,
      "step": 17700
    },
    {
      "entropy": 6.203137969970703,
      "epoch": 2.043277553375649,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004590730085451824,
      "loss": 5.8752,
      "mean_token_accuracy": 0.17790340930223464,
      "num_tokens": 44894007.0,
      "step": 17705
    },
    {
      "entropy": 6.177742671966553,
      "epoch": 2.043854587420658,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045904927544455914,
      "loss": 5.874,
      "mean_token_accuracy": 0.17606507241725922,
      "num_tokens": 44907710.0,
      "step": 17710
    },
    {
      "entropy": 6.229929685592651,
      "epoch": 2.0444316214656664,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045902553615353005,
      "loss": 5.9058,
      "mean_token_accuracy": 0.1770712301135063,
      "num_tokens": 44921485.0,
      "step": 17715
    },
    {
      "entropy": 6.187620973587036,
      "epoch": 2.0450086555106752,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004590017906728933,
      "loss": 5.8958,
      "mean_token_accuracy": 0.18173779547214508,
      "num_tokens": 44934266.0,
      "step": 17720
    },
    {
      "entropy": 6.165438842773438,
      "epoch": 2.0455856895556837,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045897803900344774,
      "loss": 5.9277,
      "mean_token_accuracy": 0.16903379410505295,
      "num_tokens": 44946828.0,
      "step": 17725
    },
    {
      "entropy": 6.140221977233887,
      "epoch": 2.0461627236006925,
      "grad_norm": 1.375,
      "learning_rate": 0.0004589542811459923,
      "loss": 5.78,
      "mean_token_accuracy": 0.19217150062322616,
      "num_tokens": 44959381.0,
      "step": 17730
    },
    {
      "entropy": 6.086016225814819,
      "epoch": 2.046739757645701,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004589305171013259,
      "loss": 5.8231,
      "mean_token_accuracy": 0.18420177549123765,
      "num_tokens": 44973232.0,
      "step": 17735
    },
    {
      "entropy": 6.205133867263794,
      "epoch": 2.04731679169071,
      "grad_norm": 1.109375,
      "learning_rate": 0.000458906746870248,
      "loss": 5.8982,
      "mean_token_accuracy": 0.17185035198926926,
      "num_tokens": 44987007.0,
      "step": 17740
    },
    {
      "entropy": 6.075138139724731,
      "epoch": 2.047893825735718,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000458882970453558,
      "loss": 5.7455,
      "mean_token_accuracy": 0.1856932371854782,
      "num_tokens": 44999358.0,
      "step": 17745
    },
    {
      "entropy": 6.179540967941284,
      "epoch": 2.048470859780727,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004588591878520556,
      "loss": 5.8964,
      "mean_token_accuracy": 0.17415293008089067,
      "num_tokens": 45011737.0,
      "step": 17750
    },
    {
      "entropy": 6.154490518569946,
      "epoch": 2.049047893825736,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004588353990665407,
      "loss": 5.8713,
      "mean_token_accuracy": 0.18093432933092118,
      "num_tokens": 45024305.0,
      "step": 17755
    },
    {
      "entropy": 6.127319622039795,
      "epoch": 2.0496249278707444,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004588116040978136,
      "loss": 5.853,
      "mean_token_accuracy": 0.1747841089963913,
      "num_tokens": 45036208.0,
      "step": 17760
    },
    {
      "entropy": 6.112208271026612,
      "epoch": 2.050201961915753,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00045878780294667437,
      "loss": 5.7983,
      "mean_token_accuracy": 0.18367448598146438,
      "num_tokens": 45048106.0,
      "step": 17765
    },
    {
      "entropy": 6.141728353500366,
      "epoch": 2.0507789959607616,
      "grad_norm": 1.125,
      "learning_rate": 0.0004587639956139237,
      "loss": 5.8392,
      "mean_token_accuracy": 0.1764760509133339,
      "num_tokens": 45061938.0,
      "step": 17770
    },
    {
      "entropy": 6.178233098983765,
      "epoch": 2.0513560300057705,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004587401821003624,
      "loss": 5.9278,
      "mean_token_accuracy": 0.17385248839855194,
      "num_tokens": 45074534.0,
      "step": 17775
    },
    {
      "entropy": 6.14547176361084,
      "epoch": 2.051933064050779,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045871636240679133,
      "loss": 5.8893,
      "mean_token_accuracy": 0.17295192331075668,
      "num_tokens": 45086332.0,
      "step": 17780
    },
    {
      "entropy": 6.125074863433838,
      "epoch": 2.0525100980957878,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004586925365340117,
      "loss": 5.7951,
      "mean_token_accuracy": 0.18127622306346894,
      "num_tokens": 45098559.0,
      "step": 17785
    },
    {
      "entropy": 6.225871276855469,
      "epoch": 2.053087132140796,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004586687044828247,
      "loss": 5.8005,
      "mean_token_accuracy": 0.17934494316577912,
      "num_tokens": 45109742.0,
      "step": 17790
    },
    {
      "entropy": 6.0726524829864506,
      "epoch": 2.053664166185805,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004586448662540322,
      "loss": 5.8707,
      "mean_token_accuracy": 0.18291911482810974,
      "num_tokens": 45122438.0,
      "step": 17795
    },
    {
      "entropy": 6.1158092498779295,
      "epoch": 2.0542412002308135,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004586210218484358,
      "loss": 5.8316,
      "mean_token_accuracy": 0.1866321235895157,
      "num_tokens": 45136499.0,
      "step": 17800
    },
    {
      "entropy": 6.2247051239013675,
      "epoch": 2.0548182342758223,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045859717126683745,
      "loss": 5.8862,
      "mean_token_accuracy": 0.1752875715494156,
      "num_tokens": 45149313.0,
      "step": 17805
    },
    {
      "entropy": 6.095439720153808,
      "epoch": 2.0553952683208307,
      "grad_norm": 1.125,
      "learning_rate": 0.00045857331451003953,
      "loss": 5.8921,
      "mean_token_accuracy": 0.17588415890932083,
      "num_tokens": 45162258.0,
      "step": 17810
    },
    {
      "entropy": 6.119178581237793,
      "epoch": 2.0559723023658396,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045854945157884435,
      "loss": 5.8008,
      "mean_token_accuracy": 0.1820492997765541,
      "num_tokens": 45174868.0,
      "step": 17815
    },
    {
      "entropy": 6.185946893692017,
      "epoch": 2.056549336410848,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045852558247405455,
      "loss": 5.8766,
      "mean_token_accuracy": 0.17692538499832153,
      "num_tokens": 45185743.0,
      "step": 17820
    },
    {
      "entropy": 6.100543451309204,
      "epoch": 2.057126370455857,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045850170719647287,
      "loss": 5.7672,
      "mean_token_accuracy": 0.18553495556116104,
      "num_tokens": 45197950.0,
      "step": 17825
    },
    {
      "entropy": 6.0452721118927,
      "epoch": 2.0577034045008658,
      "grad_norm": 0.921875,
      "learning_rate": 0.00045847782574690254,
      "loss": 5.8056,
      "mean_token_accuracy": 0.18583724349737168,
      "num_tokens": 45210721.0,
      "step": 17830
    },
    {
      "entropy": 6.20824089050293,
      "epoch": 2.058280438545874,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045845393812614664,
      "loss": 5.8953,
      "mean_token_accuracy": 0.17906317710876465,
      "num_tokens": 45222896.0,
      "step": 17835
    },
    {
      "entropy": 6.056743192672729,
      "epoch": 2.058857472590883,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004584300443350086,
      "loss": 5.747,
      "mean_token_accuracy": 0.1850567489862442,
      "num_tokens": 45235787.0,
      "step": 17840
    },
    {
      "entropy": 6.053712987899781,
      "epoch": 2.0594345066358914,
      "grad_norm": 1.125,
      "learning_rate": 0.0004584061443742922,
      "loss": 5.7648,
      "mean_token_accuracy": 0.17741813659667968,
      "num_tokens": 45247675.0,
      "step": 17845
    },
    {
      "entropy": 6.156357526779175,
      "epoch": 2.0600115406809003,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045838223824480124,
      "loss": 5.8087,
      "mean_token_accuracy": 0.18205617219209672,
      "num_tokens": 45259786.0,
      "step": 17850
    },
    {
      "entropy": 6.1536884784698485,
      "epoch": 2.0605885747259087,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004583583259473397,
      "loss": 5.8618,
      "mean_token_accuracy": 0.18085115402936935,
      "num_tokens": 45273069.0,
      "step": 17855
    },
    {
      "entropy": 6.230299615859986,
      "epoch": 2.0611656087709176,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045833440748271203,
      "loss": 5.9296,
      "mean_token_accuracy": 0.18109008073806762,
      "num_tokens": 45285593.0,
      "step": 17860
    },
    {
      "entropy": 6.062781238555909,
      "epoch": 2.061742642815926,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045831048285172266,
      "loss": 5.7973,
      "mean_token_accuracy": 0.18186074793338775,
      "num_tokens": 45297938.0,
      "step": 17865
    },
    {
      "entropy": 6.117888736724853,
      "epoch": 2.062319676860935,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004582865520551762,
      "loss": 5.7422,
      "mean_token_accuracy": 0.18247939199209212,
      "num_tokens": 45311149.0,
      "step": 17870
    },
    {
      "entropy": 6.222822904586792,
      "epoch": 2.0628967109059433,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045826261509387755,
      "loss": 5.8788,
      "mean_token_accuracy": 0.1810583084821701,
      "num_tokens": 45324169.0,
      "step": 17875
    },
    {
      "entropy": 6.068373394012451,
      "epoch": 2.063473744950952,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045823867196863197,
      "loss": 5.8637,
      "mean_token_accuracy": 0.18275449872016908,
      "num_tokens": 45336724.0,
      "step": 17880
    },
    {
      "entropy": 6.125694417953492,
      "epoch": 2.0640507789959606,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004582147226802446,
      "loss": 5.797,
      "mean_token_accuracy": 0.18555672615766525,
      "num_tokens": 45349045.0,
      "step": 17885
    },
    {
      "entropy": 6.142637109756469,
      "epoch": 2.0646278130409694,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004581907672295211,
      "loss": 5.8401,
      "mean_token_accuracy": 0.18246954083442687,
      "num_tokens": 45360732.0,
      "step": 17890
    },
    {
      "entropy": 6.120809698104859,
      "epoch": 2.065204847085978,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045816680561726716,
      "loss": 5.8774,
      "mean_token_accuracy": 0.17281407117843628,
      "num_tokens": 45372538.0,
      "step": 17895
    },
    {
      "entropy": 6.143454504013062,
      "epoch": 2.0657818811309867,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004581428378442886,
      "loss": 5.8035,
      "mean_token_accuracy": 0.18414751142263414,
      "num_tokens": 45385240.0,
      "step": 17900
    },
    {
      "entropy": 6.121305084228515,
      "epoch": 2.0663589151759956,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045811886391139173,
      "loss": 5.8529,
      "mean_token_accuracy": 0.17965630143880845,
      "num_tokens": 45398863.0,
      "step": 17905
    },
    {
      "entropy": 6.131429195404053,
      "epoch": 2.066935949221004,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00045809488381938284,
      "loss": 5.8953,
      "mean_token_accuracy": 0.1728986233472824,
      "num_tokens": 45412870.0,
      "step": 17910
    },
    {
      "entropy": 6.128060531616211,
      "epoch": 2.067512983266013,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004580708975690684,
      "loss": 5.7775,
      "mean_token_accuracy": 0.18321677148342133,
      "num_tokens": 45424314.0,
      "step": 17915
    },
    {
      "entropy": 6.142408180236816,
      "epoch": 2.0680900173110213,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004580469051612554,
      "loss": 5.8329,
      "mean_token_accuracy": 0.18117624372243882,
      "num_tokens": 45437403.0,
      "step": 17920
    },
    {
      "entropy": 6.066494941711426,
      "epoch": 2.06866705135603,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045802290659675057,
      "loss": 5.8245,
      "mean_token_accuracy": 0.17801202535629274,
      "num_tokens": 45451239.0,
      "step": 17925
    },
    {
      "entropy": 6.210570287704468,
      "epoch": 2.0692440854010385,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045799890187636123,
      "loss": 5.8963,
      "mean_token_accuracy": 0.18084469437599182,
      "num_tokens": 45463157.0,
      "step": 17930
    },
    {
      "entropy": 6.11203784942627,
      "epoch": 2.0698211194460474,
      "grad_norm": 1.0,
      "learning_rate": 0.00045797489100089464,
      "loss": 5.821,
      "mean_token_accuracy": 0.18428647816181182,
      "num_tokens": 45475774.0,
      "step": 17935
    },
    {
      "entropy": 6.171018981933594,
      "epoch": 2.070398153491056,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004579508739711585,
      "loss": 5.9445,
      "mean_token_accuracy": 0.17268856465816498,
      "num_tokens": 45489292.0,
      "step": 17940
    },
    {
      "entropy": 6.215978527069092,
      "epoch": 2.0709751875360647,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045792685078796075,
      "loss": 5.879,
      "mean_token_accuracy": 0.17229412049055098,
      "num_tokens": 45501021.0,
      "step": 17945
    },
    {
      "entropy": 6.126899862289429,
      "epoch": 2.071552221581073,
      "grad_norm": 1.109375,
      "learning_rate": 0.000457902821452109,
      "loss": 5.8413,
      "mean_token_accuracy": 0.18157557994127274,
      "num_tokens": 45513369.0,
      "step": 17950
    },
    {
      "entropy": 6.097498798370362,
      "epoch": 2.072129255626082,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045787878596441193,
      "loss": 5.7638,
      "mean_token_accuracy": 0.1842612385749817,
      "num_tokens": 45525077.0,
      "step": 17955
    },
    {
      "entropy": 6.25384373664856,
      "epoch": 2.0727062896710904,
      "grad_norm": 1.0,
      "learning_rate": 0.0004578547443256776,
      "loss": 5.9371,
      "mean_token_accuracy": 0.16918957233428955,
      "num_tokens": 45538008.0,
      "step": 17960
    },
    {
      "entropy": 6.188442325592041,
      "epoch": 2.0732833237160992,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004578306965367148,
      "loss": 5.8137,
      "mean_token_accuracy": 0.17548404932022094,
      "num_tokens": 45550797.0,
      "step": 17965
    },
    {
      "entropy": 6.166191720962525,
      "epoch": 2.073860357761108,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00045780664259833235,
      "loss": 5.8254,
      "mean_token_accuracy": 0.17992091029882432,
      "num_tokens": 45564146.0,
      "step": 17970
    },
    {
      "entropy": 6.014360427856445,
      "epoch": 2.0744373918061165,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045778258251133924,
      "loss": 5.79,
      "mean_token_accuracy": 0.1914459377527237,
      "num_tokens": 45577246.0,
      "step": 17975
    },
    {
      "entropy": 6.115902042388916,
      "epoch": 2.0750144258511254,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045775851627654474,
      "loss": 5.8788,
      "mean_token_accuracy": 0.17276673167943954,
      "num_tokens": 45589042.0,
      "step": 17980
    },
    {
      "entropy": 6.1912237167358395,
      "epoch": 2.075591459896134,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004577344438947584,
      "loss": 5.8303,
      "mean_token_accuracy": 0.18754025548696518,
      "num_tokens": 45602472.0,
      "step": 17985
    },
    {
      "entropy": 6.226692819595337,
      "epoch": 2.0761684939411427,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045771036536678984,
      "loss": 5.907,
      "mean_token_accuracy": 0.16899723708629608,
      "num_tokens": 45614208.0,
      "step": 17990
    },
    {
      "entropy": 6.052938318252563,
      "epoch": 2.076745527986151,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045768628069344885,
      "loss": 5.7712,
      "mean_token_accuracy": 0.19054771065711976,
      "num_tokens": 45626396.0,
      "step": 17995
    },
    {
      "entropy": 6.127606773376465,
      "epoch": 2.07732256203116,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004576621898755455,
      "loss": 5.826,
      "mean_token_accuracy": 0.17500776201486587,
      "num_tokens": 45640482.0,
      "step": 18000
    },
    {
      "epoch": 2.07732256203116,
      "eval_entropy": 6.02167503915797,
      "eval_loss": 6.019467830657959,
      "eval_mean_token_accuracy": 0.1770217191234108,
      "eval_num_tokens": 45640482.0,
      "eval_runtime": 23.1881,
      "eval_samples_per_second": 1213.382,
      "eval_steps_per_second": 151.673,
      "step": 18000
    },
    {
      "entropy": 6.1832404136657715,
      "epoch": 2.0778995960761684,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045763809291389024,
      "loss": 5.9315,
      "mean_token_accuracy": 0.17637717425823213,
      "num_tokens": 45654414.0,
      "step": 18005
    },
    {
      "entropy": 6.029995059967041,
      "epoch": 2.0784766301211772,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004576139898092933,
      "loss": 5.7502,
      "mean_token_accuracy": 0.18946496695280074,
      "num_tokens": 45667429.0,
      "step": 18010
    },
    {
      "entropy": 5.998154067993164,
      "epoch": 2.0790536641661856,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004575898805625657,
      "loss": 5.7939,
      "mean_token_accuracy": 0.1818694919347763,
      "num_tokens": 45680269.0,
      "step": 18015
    },
    {
      "entropy": 6.230090618133545,
      "epoch": 2.0796306982111945,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045756576517451804,
      "loss": 5.9355,
      "mean_token_accuracy": 0.1757431924343109,
      "num_tokens": 45693950.0,
      "step": 18020
    },
    {
      "entropy": 6.189419364929199,
      "epoch": 2.080207732256203,
      "grad_norm": 1.0,
      "learning_rate": 0.00045754164364596156,
      "loss": 5.828,
      "mean_token_accuracy": 0.18028396964073182,
      "num_tokens": 45706656.0,
      "step": 18025
    },
    {
      "entropy": 6.09426703453064,
      "epoch": 2.080784766301212,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004575175159777076,
      "loss": 5.7658,
      "mean_token_accuracy": 0.18802410811185838,
      "num_tokens": 45718894.0,
      "step": 18030
    },
    {
      "entropy": 6.172612285614013,
      "epoch": 2.0813618003462206,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004574933821705676,
      "loss": 5.8457,
      "mean_token_accuracy": 0.18275435119867325,
      "num_tokens": 45731556.0,
      "step": 18035
    },
    {
      "entropy": 6.110608768463135,
      "epoch": 2.081938834391229,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004574692422253534,
      "loss": 5.8104,
      "mean_token_accuracy": 0.18253417015075685,
      "num_tokens": 45743618.0,
      "step": 18040
    },
    {
      "entropy": 6.127291965484619,
      "epoch": 2.082515868436238,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045744509614287687,
      "loss": 5.8399,
      "mean_token_accuracy": 0.18637202233076094,
      "num_tokens": 45757086.0,
      "step": 18045
    },
    {
      "entropy": 6.112274837493897,
      "epoch": 2.0830929024812463,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004574209439239501,
      "loss": 5.7975,
      "mean_token_accuracy": 0.17976162284612657,
      "num_tokens": 45770153.0,
      "step": 18050
    },
    {
      "entropy": 6.1801050186157225,
      "epoch": 2.083669936526255,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045739678556938563,
      "loss": 5.8091,
      "mean_token_accuracy": 0.17971853613853456,
      "num_tokens": 45783356.0,
      "step": 18055
    },
    {
      "entropy": 6.129978275299072,
      "epoch": 2.0842469705712636,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045737262107999575,
      "loss": 5.7957,
      "mean_token_accuracy": 0.18867485523223876,
      "num_tokens": 45795557.0,
      "step": 18060
    },
    {
      "entropy": 6.17446346282959,
      "epoch": 2.0848240046162725,
      "grad_norm": 1.125,
      "learning_rate": 0.0004573484504565934,
      "loss": 5.8808,
      "mean_token_accuracy": 0.18199313730001448,
      "num_tokens": 45808011.0,
      "step": 18065
    },
    {
      "entropy": 6.076695537567138,
      "epoch": 2.085401038661281,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004573242736999915,
      "loss": 5.8039,
      "mean_token_accuracy": 0.18437327444553375,
      "num_tokens": 45820514.0,
      "step": 18070
    },
    {
      "entropy": 6.132738733291626,
      "epoch": 2.0859780727062898,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045730009081100314,
      "loss": 5.8554,
      "mean_token_accuracy": 0.17328645884990693,
      "num_tokens": 45833062.0,
      "step": 18075
    },
    {
      "entropy": 6.142876052856446,
      "epoch": 2.086555106751298,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045727590179044195,
      "loss": 5.8196,
      "mean_token_accuracy": 0.1831447511911392,
      "num_tokens": 45845723.0,
      "step": 18080
    },
    {
      "entropy": 6.206961297988892,
      "epoch": 2.087132140796307,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004572517066391211,
      "loss": 5.8924,
      "mean_token_accuracy": 0.18496408313512802,
      "num_tokens": 45857779.0,
      "step": 18085
    },
    {
      "entropy": 6.088419771194458,
      "epoch": 2.0877091748413155,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045722750535785484,
      "loss": 5.8481,
      "mean_token_accuracy": 0.17633532136678695,
      "num_tokens": 45871321.0,
      "step": 18090
    },
    {
      "entropy": 6.158281707763672,
      "epoch": 2.0882862088863243,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045720329794745685,
      "loss": 5.8571,
      "mean_token_accuracy": 0.18947115540504456,
      "num_tokens": 45884048.0,
      "step": 18095
    },
    {
      "entropy": 6.138871192932129,
      "epoch": 2.0888632429313327,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004571790844087415,
      "loss": 5.8864,
      "mean_token_accuracy": 0.17597791254520417,
      "num_tokens": 45895986.0,
      "step": 18100
    },
    {
      "entropy": 6.126863193511963,
      "epoch": 2.0894402769763416,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004571548647425231,
      "loss": 5.9077,
      "mean_token_accuracy": 0.18134653121232985,
      "num_tokens": 45909120.0,
      "step": 18105
    },
    {
      "entropy": 6.11176381111145,
      "epoch": 2.0900173110213505,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004571306389496164,
      "loss": 5.7977,
      "mean_token_accuracy": 0.18397730439901352,
      "num_tokens": 45921164.0,
      "step": 18110
    },
    {
      "entropy": 6.064571571350098,
      "epoch": 2.090594345066359,
      "grad_norm": 1.875,
      "learning_rate": 0.00045710640703083594,
      "loss": 5.7138,
      "mean_token_accuracy": 0.1884634479880333,
      "num_tokens": 45933037.0,
      "step": 18115
    },
    {
      "entropy": 6.081005954742432,
      "epoch": 2.0911713791113677,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045708216898699707,
      "loss": 5.8057,
      "mean_token_accuracy": 0.18063925504684447,
      "num_tokens": 45945381.0,
      "step": 18120
    },
    {
      "entropy": 6.110395193099976,
      "epoch": 2.091748413156376,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045705792481891483,
      "loss": 5.721,
      "mean_token_accuracy": 0.18918657004833223,
      "num_tokens": 45959446.0,
      "step": 18125
    },
    {
      "entropy": 6.229499673843383,
      "epoch": 2.092325447201385,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045703367452740477,
      "loss": 5.947,
      "mean_token_accuracy": 0.1709427833557129,
      "num_tokens": 45971773.0,
      "step": 18130
    },
    {
      "entropy": 6.071350765228272,
      "epoch": 2.0929024812463934,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045700941811328247,
      "loss": 5.7966,
      "mean_token_accuracy": 0.17864670604467392,
      "num_tokens": 45983543.0,
      "step": 18135
    },
    {
      "entropy": 6.137739610671997,
      "epoch": 2.0934795152914023,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045698515557736374,
      "loss": 5.7935,
      "mean_token_accuracy": 0.1798617199063301,
      "num_tokens": 45995231.0,
      "step": 18140
    },
    {
      "entropy": 6.158017969131469,
      "epoch": 2.0940565493364107,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00045696088692046477,
      "loss": 5.8582,
      "mean_token_accuracy": 0.17394719570875167,
      "num_tokens": 46009414.0,
      "step": 18145
    },
    {
      "entropy": 6.09553632736206,
      "epoch": 2.0946335833814196,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045693661214340174,
      "loss": 5.8093,
      "mean_token_accuracy": 0.1791607216000557,
      "num_tokens": 46020781.0,
      "step": 18150
    },
    {
      "entropy": 6.157679700851441,
      "epoch": 2.095210617426428,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045691233124699117,
      "loss": 5.8261,
      "mean_token_accuracy": 0.18285781145095825,
      "num_tokens": 46032479.0,
      "step": 18155
    },
    {
      "entropy": 6.086474943161011,
      "epoch": 2.095787651471437,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004568880442320497,
      "loss": 5.7969,
      "mean_token_accuracy": 0.18403210639953613,
      "num_tokens": 46044837.0,
      "step": 18160
    },
    {
      "entropy": 6.113948678970337,
      "epoch": 2.0963646855164453,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00045686375109939417,
      "loss": 5.8961,
      "mean_token_accuracy": 0.17929877787828447,
      "num_tokens": 46056194.0,
      "step": 18165
    },
    {
      "entropy": 6.099827527999878,
      "epoch": 2.096941719561454,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004568394518498417,
      "loss": 5.8218,
      "mean_token_accuracy": 0.1776571527123451,
      "num_tokens": 46069536.0,
      "step": 18170
    },
    {
      "entropy": 6.004854345321656,
      "epoch": 2.0975187536064626,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045681514648420965,
      "loss": 5.7076,
      "mean_token_accuracy": 0.19538643956184387,
      "num_tokens": 46082616.0,
      "step": 18175
    },
    {
      "entropy": 6.078830051422119,
      "epoch": 2.0980957876514714,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004567908350033154,
      "loss": 5.842,
      "mean_token_accuracy": 0.17523629367351531,
      "num_tokens": 46094839.0,
      "step": 18180
    },
    {
      "entropy": 6.09807243347168,
      "epoch": 2.0986728216964803,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004567665174079767,
      "loss": 5.8438,
      "mean_token_accuracy": 0.1792247176170349,
      "num_tokens": 46106475.0,
      "step": 18185
    },
    {
      "entropy": 6.063588714599609,
      "epoch": 2.0992498557414887,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045674219369901153,
      "loss": 5.7783,
      "mean_token_accuracy": 0.1812007173895836,
      "num_tokens": 46118922.0,
      "step": 18190
    },
    {
      "entropy": 6.185597896575928,
      "epoch": 2.0998268897864976,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004567178638772379,
      "loss": 5.9055,
      "mean_token_accuracy": 0.1734337881207466,
      "num_tokens": 46131848.0,
      "step": 18195
    },
    {
      "entropy": 6.07983021736145,
      "epoch": 2.100403923831506,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004566935279434742,
      "loss": 5.818,
      "mean_token_accuracy": 0.17829743027687073,
      "num_tokens": 46145130.0,
      "step": 18200
    },
    {
      "entropy": 6.109281873703003,
      "epoch": 2.100980957876515,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00045666918589853895,
      "loss": 5.7722,
      "mean_token_accuracy": 0.1829168826341629,
      "num_tokens": 46160301.0,
      "step": 18205
    },
    {
      "entropy": 6.191259813308716,
      "epoch": 2.1015579919215233,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045664483774325094,
      "loss": 5.8475,
      "mean_token_accuracy": 0.18255991637706756,
      "num_tokens": 46171650.0,
      "step": 18210
    },
    {
      "entropy": 6.128474855422974,
      "epoch": 2.102135025966532,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004566204834784289,
      "loss": 5.8523,
      "mean_token_accuracy": 0.18033778071403503,
      "num_tokens": 46183302.0,
      "step": 18215
    },
    {
      "entropy": 6.095445251464843,
      "epoch": 2.1027120600115405,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045659612310489225,
      "loss": 5.814,
      "mean_token_accuracy": 0.18037196397781372,
      "num_tokens": 46195459.0,
      "step": 18220
    },
    {
      "entropy": 6.056231594085693,
      "epoch": 2.1032890940565494,
      "grad_norm": 1.390625,
      "learning_rate": 0.00045657175662346014,
      "loss": 5.7611,
      "mean_token_accuracy": 0.18142868131399154,
      "num_tokens": 46207969.0,
      "step": 18225
    },
    {
      "entropy": 6.185912609100342,
      "epoch": 2.103866128101558,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004565473840349522,
      "loss": 5.8783,
      "mean_token_accuracy": 0.17642370164394378,
      "num_tokens": 46219976.0,
      "step": 18230
    },
    {
      "entropy": 6.098413181304932,
      "epoch": 2.1044431621465667,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045652300534018816,
      "loss": 5.7659,
      "mean_token_accuracy": 0.1775495797395706,
      "num_tokens": 46232172.0,
      "step": 18235
    },
    {
      "entropy": 6.104469585418701,
      "epoch": 2.105020196191575,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000456498620539988,
      "loss": 5.9136,
      "mean_token_accuracy": 0.17705566734075545,
      "num_tokens": 46244288.0,
      "step": 18240
    },
    {
      "entropy": 6.18358850479126,
      "epoch": 2.105597230236584,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004564742296351719,
      "loss": 5.8625,
      "mean_token_accuracy": 0.18261765092611312,
      "num_tokens": 46256464.0,
      "step": 18245
    },
    {
      "entropy": 6.237294387817383,
      "epoch": 2.106174264281593,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045644983262656014,
      "loss": 5.9542,
      "mean_token_accuracy": 0.17416095733642578,
      "num_tokens": 46270356.0,
      "step": 18250
    },
    {
      "entropy": 6.074225234985351,
      "epoch": 2.1067512983266012,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004564254295149735,
      "loss": 5.8055,
      "mean_token_accuracy": 0.18673934489488603,
      "num_tokens": 46282990.0,
      "step": 18255
    },
    {
      "entropy": 6.113529396057129,
      "epoch": 2.10732833237161,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045640102030123264,
      "loss": 5.7685,
      "mean_token_accuracy": 0.17842842340469361,
      "num_tokens": 46294665.0,
      "step": 18260
    },
    {
      "entropy": 6.142454195022583,
      "epoch": 2.1079053664166185,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045637660498615865,
      "loss": 5.7896,
      "mean_token_accuracy": 0.17992257624864577,
      "num_tokens": 46306950.0,
      "step": 18265
    },
    {
      "entropy": 6.1777101993560795,
      "epoch": 2.1084824004616274,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004563521835705725,
      "loss": 5.916,
      "mean_token_accuracy": 0.17760988026857377,
      "num_tokens": 46320475.0,
      "step": 18270
    },
    {
      "entropy": 6.155001401901245,
      "epoch": 2.109059434506636,
      "grad_norm": 0.9375,
      "learning_rate": 0.00045632775605529587,
      "loss": 5.8271,
      "mean_token_accuracy": 0.18418175727128983,
      "num_tokens": 46332624.0,
      "step": 18275
    },
    {
      "entropy": 6.158223724365234,
      "epoch": 2.1096364685516447,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004563033224411501,
      "loss": 5.8076,
      "mean_token_accuracy": 0.1860724151134491,
      "num_tokens": 46344257.0,
      "step": 18280
    },
    {
      "entropy": 6.077128076553345,
      "epoch": 2.110213502596653,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004562788827289572,
      "loss": 5.792,
      "mean_token_accuracy": 0.17955469340085983,
      "num_tokens": 46356526.0,
      "step": 18285
    },
    {
      "entropy": 6.1878846168518065,
      "epoch": 2.110790536641662,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045625443691953914,
      "loss": 5.858,
      "mean_token_accuracy": 0.17952210009098052,
      "num_tokens": 46368439.0,
      "step": 18290
    },
    {
      "entropy": 6.114879989624024,
      "epoch": 2.1113675706866704,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004562299850137181,
      "loss": 5.8792,
      "mean_token_accuracy": 0.17571465373039247,
      "num_tokens": 46381627.0,
      "step": 18295
    },
    {
      "entropy": 6.138954210281372,
      "epoch": 2.111944604731679,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004562055270123166,
      "loss": 5.8291,
      "mean_token_accuracy": 0.17781825959682465,
      "num_tokens": 46393869.0,
      "step": 18300
    },
    {
      "entropy": 6.126654767990113,
      "epoch": 2.1125216387766876,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045618106291615715,
      "loss": 5.7993,
      "mean_token_accuracy": 0.18073048293590546,
      "num_tokens": 46405911.0,
      "step": 18305
    },
    {
      "entropy": 6.114129972457886,
      "epoch": 2.1130986728216965,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004561565927260627,
      "loss": 5.8574,
      "mean_token_accuracy": 0.18035200238227844,
      "num_tokens": 46418258.0,
      "step": 18310
    },
    {
      "entropy": 6.280330514907837,
      "epoch": 2.1136757068667054,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004561321164428561,
      "loss": 5.9502,
      "mean_token_accuracy": 0.17468089908361434,
      "num_tokens": 46431211.0,
      "step": 18315
    },
    {
      "entropy": 6.289624214172363,
      "epoch": 2.114252740911714,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004561076340673609,
      "loss": 5.9833,
      "mean_token_accuracy": 0.16389265060424804,
      "num_tokens": 46445062.0,
      "step": 18320
    },
    {
      "entropy": 6.149607133865357,
      "epoch": 2.1148297749567226,
      "grad_norm": 1.0,
      "learning_rate": 0.0004560831456004003,
      "loss": 5.9177,
      "mean_token_accuracy": 0.1738544449210167,
      "num_tokens": 46457251.0,
      "step": 18325
    },
    {
      "entropy": 6.105461931228637,
      "epoch": 2.115406809001731,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004560586510427981,
      "loss": 5.7471,
      "mean_token_accuracy": 0.18489423841238023,
      "num_tokens": 46471973.0,
      "step": 18330
    },
    {
      "entropy": 6.1815605640411375,
      "epoch": 2.11598384304674,
      "grad_norm": 1.0,
      "learning_rate": 0.0004560341503953781,
      "loss": 5.8493,
      "mean_token_accuracy": 0.17299876511096954,
      "num_tokens": 46484363.0,
      "step": 18335
    },
    {
      "entropy": 6.1645143032073975,
      "epoch": 2.1165608770917483,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004560096436589643,
      "loss": 5.8397,
      "mean_token_accuracy": 0.18262559175491333,
      "num_tokens": 46496033.0,
      "step": 18340
    },
    {
      "entropy": 6.08131103515625,
      "epoch": 2.117137911136757,
      "grad_norm": 0.96875,
      "learning_rate": 0.00045598513083438115,
      "loss": 5.7415,
      "mean_token_accuracy": 0.18414081037044525,
      "num_tokens": 46507826.0,
      "step": 18345
    },
    {
      "entropy": 6.064138746261596,
      "epoch": 2.1177149451817656,
      "grad_norm": 1.125,
      "learning_rate": 0.00045596061192245297,
      "loss": 5.8074,
      "mean_token_accuracy": 0.181829634308815,
      "num_tokens": 46519697.0,
      "step": 18350
    },
    {
      "entropy": 6.140630292892456,
      "epoch": 2.1182919792267745,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004559360869240045,
      "loss": 5.8625,
      "mean_token_accuracy": 0.1774944022297859,
      "num_tokens": 46532486.0,
      "step": 18355
    },
    {
      "entropy": 6.1967394828796385,
      "epoch": 2.118869013271783,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004559115558398606,
      "loss": 5.8337,
      "mean_token_accuracy": 0.18263743668794633,
      "num_tokens": 46543985.0,
      "step": 18360
    },
    {
      "entropy": 6.107927083969116,
      "epoch": 2.1194460473167918,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004558870186708465,
      "loss": 5.7908,
      "mean_token_accuracy": 0.1891636148095131,
      "num_tokens": 46556326.0,
      "step": 18365
    },
    {
      "entropy": 6.101717758178711,
      "epoch": 2.1200230813618,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045586247541778724,
      "loss": 5.8625,
      "mean_token_accuracy": 0.17593564242124557,
      "num_tokens": 46567955.0,
      "step": 18370
    },
    {
      "entropy": 6.120845317840576,
      "epoch": 2.120600115406809,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004558379260815085,
      "loss": 5.7953,
      "mean_token_accuracy": 0.1774279549717903,
      "num_tokens": 46580581.0,
      "step": 18375
    },
    {
      "entropy": 6.136447286605835,
      "epoch": 2.1211771494518175,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004558133706628359,
      "loss": 5.8537,
      "mean_token_accuracy": 0.18103795647621154,
      "num_tokens": 46592503.0,
      "step": 18380
    },
    {
      "entropy": 6.128790664672851,
      "epoch": 2.1217541834968263,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00045578880916259554,
      "loss": 5.7812,
      "mean_token_accuracy": 0.18831520676612853,
      "num_tokens": 46605542.0,
      "step": 18385
    },
    {
      "entropy": 6.12329626083374,
      "epoch": 2.122331217541835,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004557642415816132,
      "loss": 5.7553,
      "mean_token_accuracy": 0.1863365352153778,
      "num_tokens": 46618826.0,
      "step": 18390
    },
    {
      "entropy": 6.06908164024353,
      "epoch": 2.1229082515868436,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004557396679207155,
      "loss": 5.7862,
      "mean_token_accuracy": 0.17902138829231262,
      "num_tokens": 46630295.0,
      "step": 18395
    },
    {
      "entropy": 6.147324943542481,
      "epoch": 2.1234852856318525,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045571508818072887,
      "loss": 5.8815,
      "mean_token_accuracy": 0.17316668331623078,
      "num_tokens": 46642605.0,
      "step": 18400
    },
    {
      "entropy": 6.199816370010376,
      "epoch": 2.124062319676861,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004556905023624799,
      "loss": 5.9545,
      "mean_token_accuracy": 0.1660635530948639,
      "num_tokens": 46656411.0,
      "step": 18405
    },
    {
      "entropy": 6.225232028961182,
      "epoch": 2.1246393537218697,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045566591046679577,
      "loss": 5.8855,
      "mean_token_accuracy": 0.18062201887369156,
      "num_tokens": 46668231.0,
      "step": 18410
    },
    {
      "entropy": 6.202814102172852,
      "epoch": 2.125216387766878,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00045564131249450343,
      "loss": 5.8812,
      "mean_token_accuracy": 0.18052396476268767,
      "num_tokens": 46681448.0,
      "step": 18415
    },
    {
      "entropy": 6.076045179367066,
      "epoch": 2.125793421811887,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004556167084464302,
      "loss": 5.8125,
      "mean_token_accuracy": 0.17714585065841676,
      "num_tokens": 46695043.0,
      "step": 18420
    },
    {
      "entropy": 6.201786851882934,
      "epoch": 2.1263704558568954,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004555920983234038,
      "loss": 5.8674,
      "mean_token_accuracy": 0.1720680058002472,
      "num_tokens": 46707099.0,
      "step": 18425
    },
    {
      "entropy": 6.109007930755615,
      "epoch": 2.1269474899019043,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045556748212625183,
      "loss": 5.7831,
      "mean_token_accuracy": 0.1828474223613739,
      "num_tokens": 46720913.0,
      "step": 18430
    },
    {
      "entropy": 6.171474409103394,
      "epoch": 2.1275245239469127,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004555428598558023,
      "loss": 5.8811,
      "mean_token_accuracy": 0.17357461303472518,
      "num_tokens": 46733405.0,
      "step": 18435
    },
    {
      "entropy": 6.217512607574463,
      "epoch": 2.1281015579919216,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004555182315128833,
      "loss": 5.8796,
      "mean_token_accuracy": 0.17767830789089203,
      "num_tokens": 46745588.0,
      "step": 18440
    },
    {
      "entropy": 6.161148357391357,
      "epoch": 2.12867859203693,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004554935970983234,
      "loss": 5.9146,
      "mean_token_accuracy": 0.17110783606767654,
      "num_tokens": 46758091.0,
      "step": 18445
    },
    {
      "entropy": 6.142706441879272,
      "epoch": 2.129255626081939,
      "grad_norm": 0.875,
      "learning_rate": 0.0004554689566129509,
      "loss": 5.8611,
      "mean_token_accuracy": 0.17835298031568528,
      "num_tokens": 46770966.0,
      "step": 18450
    },
    {
      "entropy": 6.182175874710083,
      "epoch": 2.1298326601269473,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045544431005759467,
      "loss": 5.8886,
      "mean_token_accuracy": 0.17507684230804443,
      "num_tokens": 46783947.0,
      "step": 18455
    },
    {
      "entropy": 6.066798353195191,
      "epoch": 2.130409694171956,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045541965743308376,
      "loss": 5.7219,
      "mean_token_accuracy": 0.1836577132344246,
      "num_tokens": 46796985.0,
      "step": 18460
    },
    {
      "entropy": 6.095421028137207,
      "epoch": 2.130986728216965,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004553949987402473,
      "loss": 5.7771,
      "mean_token_accuracy": 0.18556680530309677,
      "num_tokens": 46809397.0,
      "step": 18465
    },
    {
      "entropy": 6.056061267852783,
      "epoch": 2.1315637622619734,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004553703339799146,
      "loss": 5.7784,
      "mean_token_accuracy": 0.190565824508667,
      "num_tokens": 46822336.0,
      "step": 18470
    },
    {
      "entropy": 6.067279481887818,
      "epoch": 2.1321407963069823,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004553456631529154,
      "loss": 5.791,
      "mean_token_accuracy": 0.19106384813785554,
      "num_tokens": 46835048.0,
      "step": 18475
    },
    {
      "entropy": 6.140585851669312,
      "epoch": 2.1327178303519907,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004553209862600794,
      "loss": 5.9148,
      "mean_token_accuracy": 0.17574238479137422,
      "num_tokens": 46847295.0,
      "step": 18480
    },
    {
      "entropy": 6.133851099014282,
      "epoch": 2.1332948643969996,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004552963033022365,
      "loss": 5.7609,
      "mean_token_accuracy": 0.1868451252579689,
      "num_tokens": 46859373.0,
      "step": 18485
    },
    {
      "entropy": 6.149752521514893,
      "epoch": 2.133871898442008,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00045527161428021706,
      "loss": 5.7814,
      "mean_token_accuracy": 0.1881526991724968,
      "num_tokens": 46871307.0,
      "step": 18490
    },
    {
      "entropy": 6.084883975982666,
      "epoch": 2.134448932487017,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004552469191948514,
      "loss": 5.7869,
      "mean_token_accuracy": 0.18610908687114716,
      "num_tokens": 46883577.0,
      "step": 18495
    },
    {
      "entropy": 6.102695083618164,
      "epoch": 2.1350259665320253,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004552222180469702,
      "loss": 5.7902,
      "mean_token_accuracy": 0.18510093539953232,
      "num_tokens": 46896413.0,
      "step": 18500
    },
    {
      "entropy": 6.154614543914795,
      "epoch": 2.135603000577034,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045519751083740413,
      "loss": 5.8465,
      "mean_token_accuracy": 0.17691437602043153,
      "num_tokens": 46907225.0,
      "step": 18505
    },
    {
      "entropy": 6.076205921173096,
      "epoch": 2.1361800346220425,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045517279756698435,
      "loss": 5.7338,
      "mean_token_accuracy": 0.18658654540777206,
      "num_tokens": 46919852.0,
      "step": 18510
    },
    {
      "entropy": 6.142713260650635,
      "epoch": 2.1367570686670514,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00045514807823654203,
      "loss": 5.8741,
      "mean_token_accuracy": 0.17789843529462815,
      "num_tokens": 46932803.0,
      "step": 18515
    },
    {
      "entropy": 6.106948804855347,
      "epoch": 2.13733410271206,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004551233528469086,
      "loss": 5.8577,
      "mean_token_accuracy": 0.17780109643936157,
      "num_tokens": 46945340.0,
      "step": 18520
    },
    {
      "entropy": 6.127005195617675,
      "epoch": 2.1379111367570687,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045509862139891553,
      "loss": 5.841,
      "mean_token_accuracy": 0.1787581264972687,
      "num_tokens": 46957837.0,
      "step": 18525
    },
    {
      "entropy": 6.132759141921997,
      "epoch": 2.138488170802077,
      "grad_norm": 0.96875,
      "learning_rate": 0.00045507388389339496,
      "loss": 5.8152,
      "mean_token_accuracy": 0.17835501581430435,
      "num_tokens": 46971106.0,
      "step": 18530
    },
    {
      "entropy": 6.160295677185059,
      "epoch": 2.139065204847086,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045504914033117866,
      "loss": 5.8379,
      "mean_token_accuracy": 0.17522946000099182,
      "num_tokens": 46982897.0,
      "step": 18535
    },
    {
      "entropy": 6.126948642730713,
      "epoch": 2.139642238892095,
      "grad_norm": 1.4375,
      "learning_rate": 0.00045502439071309897,
      "loss": 5.8566,
      "mean_token_accuracy": 0.1783952981233597,
      "num_tokens": 46995812.0,
      "step": 18540
    },
    {
      "entropy": 6.175791072845459,
      "epoch": 2.1402192729371032,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00045499963503998835,
      "loss": 5.8892,
      "mean_token_accuracy": 0.17423874586820604,
      "num_tokens": 47008092.0,
      "step": 18545
    },
    {
      "entropy": 6.06117730140686,
      "epoch": 2.140796306982112,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004549748733126794,
      "loss": 5.6681,
      "mean_token_accuracy": 0.18998509347438813,
      "num_tokens": 47020870.0,
      "step": 18550
    },
    {
      "entropy": 6.056320524215698,
      "epoch": 2.1413733410271205,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000454950105532005,
      "loss": 5.7433,
      "mean_token_accuracy": 0.187506827712059,
      "num_tokens": 47033051.0,
      "step": 18555
    },
    {
      "entropy": 6.107619714736939,
      "epoch": 2.1419503750721294,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00045492533169879816,
      "loss": 5.8118,
      "mean_token_accuracy": 0.1814291924238205,
      "num_tokens": 47044868.0,
      "step": 18560
    },
    {
      "entropy": 6.096580648422242,
      "epoch": 2.142527409117138,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00045490055181389216,
      "loss": 5.7811,
      "mean_token_accuracy": 0.187594373524189,
      "num_tokens": 47057050.0,
      "step": 18565
    },
    {
      "entropy": 6.091726398468017,
      "epoch": 2.1431044431621467,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045487576587812046,
      "loss": 5.7592,
      "mean_token_accuracy": 0.18470898568630217,
      "num_tokens": 47070338.0,
      "step": 18570
    },
    {
      "entropy": 5.996621608734131,
      "epoch": 2.143681477207155,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045485097389231675,
      "loss": 5.6982,
      "mean_token_accuracy": 0.19766385704278946,
      "num_tokens": 47082958.0,
      "step": 18575
    },
    {
      "entropy": 6.08142728805542,
      "epoch": 2.144258511252164,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004548261758573149,
      "loss": 5.8191,
      "mean_token_accuracy": 0.1848795458674431,
      "num_tokens": 47096803.0,
      "step": 18580
    },
    {
      "entropy": 6.176984643936157,
      "epoch": 2.1448355452971724,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004548013717739489,
      "loss": 5.8341,
      "mean_token_accuracy": 0.17580230236053468,
      "num_tokens": 47109858.0,
      "step": 18585
    },
    {
      "entropy": 6.160211181640625,
      "epoch": 2.145412579342181,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004547765616430531,
      "loss": 5.8464,
      "mean_token_accuracy": 0.18433965295553206,
      "num_tokens": 47123664.0,
      "step": 18590
    },
    {
      "entropy": 6.056757831573487,
      "epoch": 2.14598961338719,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004547517454654619,
      "loss": 5.8083,
      "mean_token_accuracy": 0.178685262799263,
      "num_tokens": 47135860.0,
      "step": 18595
    },
    {
      "entropy": 6.108637952804566,
      "epoch": 2.1465666474321985,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00045472692324201005,
      "loss": 5.8288,
      "mean_token_accuracy": 0.17702032774686813,
      "num_tokens": 47147571.0,
      "step": 18600
    },
    {
      "entropy": 6.1297613143920895,
      "epoch": 2.1471436814772074,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045470209497353243,
      "loss": 5.7777,
      "mean_token_accuracy": 0.18645027577877044,
      "num_tokens": 47158947.0,
      "step": 18605
    },
    {
      "entropy": 6.0609911441802975,
      "epoch": 2.1477207155222158,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004546772606608641,
      "loss": 5.7851,
      "mean_token_accuracy": 0.18341971933841705,
      "num_tokens": 47172666.0,
      "step": 18610
    },
    {
      "entropy": 6.10962986946106,
      "epoch": 2.1482977495672246,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004546524203048404,
      "loss": 5.77,
      "mean_token_accuracy": 0.19046162962913513,
      "num_tokens": 47185664.0,
      "step": 18615
    },
    {
      "entropy": 6.1882795810699465,
      "epoch": 2.148874783612233,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004546275739062967,
      "loss": 5.8841,
      "mean_token_accuracy": 0.17551183998584746,
      "num_tokens": 47198325.0,
      "step": 18620
    },
    {
      "entropy": 6.084166479110718,
      "epoch": 2.149451817657242,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004546027214660688,
      "loss": 5.7505,
      "mean_token_accuracy": 0.18785581439733506,
      "num_tokens": 47211676.0,
      "step": 18625
    },
    {
      "entropy": 6.1577411651611325,
      "epoch": 2.1500288517022503,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004545778629849926,
      "loss": 5.8591,
      "mean_token_accuracy": 0.17159380316734313,
      "num_tokens": 47223956.0,
      "step": 18630
    },
    {
      "entropy": 6.118579149246216,
      "epoch": 2.150605885747259,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004545529984639042,
      "loss": 5.7961,
      "mean_token_accuracy": 0.18656814694404603,
      "num_tokens": 47236806.0,
      "step": 18635
    },
    {
      "entropy": 6.086405181884766,
      "epoch": 2.1511829197922676,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004545281279036398,
      "loss": 5.796,
      "mean_token_accuracy": 0.1826661452651024,
      "num_tokens": 47249750.0,
      "step": 18640
    },
    {
      "entropy": 6.074776649475098,
      "epoch": 2.1517599538372765,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004545032513050361,
      "loss": 5.8,
      "mean_token_accuracy": 0.18613702654838563,
      "num_tokens": 47261480.0,
      "step": 18645
    },
    {
      "entropy": 6.162259864807129,
      "epoch": 2.152336987882285,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004544783686689296,
      "loss": 5.8891,
      "mean_token_accuracy": 0.17685239166021346,
      "num_tokens": 47275025.0,
      "step": 18650
    },
    {
      "entropy": 6.121089601516724,
      "epoch": 2.1529140219272938,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00045445347999615736,
      "loss": 5.8213,
      "mean_token_accuracy": 0.18394801914691924,
      "num_tokens": 47287966.0,
      "step": 18655
    },
    {
      "entropy": 6.1640623092651365,
      "epoch": 2.153491055972302,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045442858528755653,
      "loss": 5.8263,
      "mean_token_accuracy": 0.18183122873306273,
      "num_tokens": 47299883.0,
      "step": 18660
    },
    {
      "entropy": 6.099168491363526,
      "epoch": 2.154068090017311,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045440368454396435,
      "loss": 5.8478,
      "mean_token_accuracy": 0.1780385732650757,
      "num_tokens": 47311704.0,
      "step": 18665
    },
    {
      "entropy": 6.197096538543701,
      "epoch": 2.15464512406232,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004543787777662183,
      "loss": 5.9305,
      "mean_token_accuracy": 0.17666108757257462,
      "num_tokens": 47325663.0,
      "step": 18670
    },
    {
      "entropy": 6.182325029373169,
      "epoch": 2.1552221581073283,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00045435386495515617,
      "loss": 5.8993,
      "mean_token_accuracy": 0.17696262300014495,
      "num_tokens": 47339349.0,
      "step": 18675
    },
    {
      "entropy": 6.1271071434021,
      "epoch": 2.155799192152337,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004543289461116159,
      "loss": 5.956,
      "mean_token_accuracy": 0.1753370225429535,
      "num_tokens": 47352245.0,
      "step": 18680
    },
    {
      "entropy": 6.1677321434021,
      "epoch": 2.1563762261973456,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004543040212364356,
      "loss": 5.8268,
      "mean_token_accuracy": 0.1744371086359024,
      "num_tokens": 47364564.0,
      "step": 18685
    },
    {
      "entropy": 6.088117885589599,
      "epoch": 2.1569532602423545,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004542790903304536,
      "loss": 5.767,
      "mean_token_accuracy": 0.1808383896946907,
      "num_tokens": 47377156.0,
      "step": 18690
    },
    {
      "entropy": 6.130134391784668,
      "epoch": 2.157530294287363,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004542541533945085,
      "loss": 5.8399,
      "mean_token_accuracy": 0.18007858395576476,
      "num_tokens": 47390003.0,
      "step": 18695
    },
    {
      "entropy": 6.211007070541382,
      "epoch": 2.1581073283323717,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045422921042943885,
      "loss": 5.919,
      "mean_token_accuracy": 0.17603070437908172,
      "num_tokens": 47403651.0,
      "step": 18700
    },
    {
      "entropy": 6.101494884490966,
      "epoch": 2.15868436237738,
      "grad_norm": 1.078125,
      "learning_rate": 0.000454204261436084,
      "loss": 5.8079,
      "mean_token_accuracy": 0.17943190187215804,
      "num_tokens": 47415481.0,
      "step": 18705
    },
    {
      "entropy": 6.091569185256958,
      "epoch": 2.159261396422389,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045417930641528255,
      "loss": 5.8885,
      "mean_token_accuracy": 0.17819674611091613,
      "num_tokens": 47428303.0,
      "step": 18710
    },
    {
      "entropy": 6.13434796333313,
      "epoch": 2.1598384304673974,
      "grad_norm": 1.25,
      "learning_rate": 0.00045415434536787424,
      "loss": 5.7902,
      "mean_token_accuracy": 0.17965091466903688,
      "num_tokens": 47441270.0,
      "step": 18715
    },
    {
      "entropy": 6.1493792057037355,
      "epoch": 2.1604154645124063,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004541293782946985,
      "loss": 5.7796,
      "mean_token_accuracy": 0.18373124450445175,
      "num_tokens": 47454870.0,
      "step": 18720
    },
    {
      "entropy": 6.149935626983643,
      "epoch": 2.1609924985574147,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004541044051965952,
      "loss": 5.7973,
      "mean_token_accuracy": 0.18728828579187393,
      "num_tokens": 47466099.0,
      "step": 18725
    },
    {
      "entropy": 6.120571422576904,
      "epoch": 2.1615695326024236,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004540794260744041,
      "loss": 5.7958,
      "mean_token_accuracy": 0.1790065959095955,
      "num_tokens": 47477206.0,
      "step": 18730
    },
    {
      "entropy": 5.939454364776611,
      "epoch": 2.162146566647432,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004540544409289655,
      "loss": 5.669,
      "mean_token_accuracy": 0.1870392680168152,
      "num_tokens": 47489500.0,
      "step": 18735
    },
    {
      "entropy": 6.206492710113525,
      "epoch": 2.162723600692441,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004540294497611197,
      "loss": 5.8729,
      "mean_token_accuracy": 0.17840335220098497,
      "num_tokens": 47502090.0,
      "step": 18740
    },
    {
      "entropy": 6.285609006881714,
      "epoch": 2.1633006347374497,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045400445257170725,
      "loss": 5.9784,
      "mean_token_accuracy": 0.17141806483268737,
      "num_tokens": 47515496.0,
      "step": 18745
    },
    {
      "entropy": 6.1632373332977295,
      "epoch": 2.163877668782458,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004539794493615689,
      "loss": 5.822,
      "mean_token_accuracy": 0.17540449053049087,
      "num_tokens": 47529174.0,
      "step": 18750
    },
    {
      "entropy": 6.141179466247559,
      "epoch": 2.164454702827467,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004539544401315458,
      "loss": 5.8598,
      "mean_token_accuracy": 0.17870617806911468,
      "num_tokens": 47542495.0,
      "step": 18755
    },
    {
      "entropy": 6.137688255310058,
      "epoch": 2.1650317368724754,
      "grad_norm": 1.0625,
      "learning_rate": 0.000453929424882479,
      "loss": 5.8164,
      "mean_token_accuracy": 0.1838001102209091,
      "num_tokens": 47554936.0,
      "step": 18760
    },
    {
      "entropy": 6.252662944793701,
      "epoch": 2.1656087709174843,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045390440361520987,
      "loss": 5.9002,
      "mean_token_accuracy": 0.17206381112337113,
      "num_tokens": 47568168.0,
      "step": 18765
    },
    {
      "entropy": 6.132688236236572,
      "epoch": 2.1661858049624927,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004538793763305799,
      "loss": 5.7544,
      "mean_token_accuracy": 0.18806611150503158,
      "num_tokens": 47581944.0,
      "step": 18770
    },
    {
      "entropy": 6.094078063964844,
      "epoch": 2.1667628390075016,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045385434302943104,
      "loss": 5.8481,
      "mean_token_accuracy": 0.17577507793903352,
      "num_tokens": 47595554.0,
      "step": 18775
    },
    {
      "entropy": 6.208167028427124,
      "epoch": 2.16733987305251,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004538293037126052,
      "loss": 5.953,
      "mean_token_accuracy": 0.16940199285745622,
      "num_tokens": 47608516.0,
      "step": 18780
    },
    {
      "entropy": 6.112657451629639,
      "epoch": 2.167916907097519,
      "grad_norm": 0.96875,
      "learning_rate": 0.00045380425838094444,
      "loss": 5.8132,
      "mean_token_accuracy": 0.18307690173387528,
      "num_tokens": 47620100.0,
      "step": 18785
    },
    {
      "entropy": 6.128265619277954,
      "epoch": 2.1684939411425272,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045377920703529133,
      "loss": 5.774,
      "mean_token_accuracy": 0.1876115694642067,
      "num_tokens": 47632982.0,
      "step": 18790
    },
    {
      "entropy": 6.109172010421753,
      "epoch": 2.169070975187536,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004537541496764885,
      "loss": 5.7849,
      "mean_token_accuracy": 0.1780137836933136,
      "num_tokens": 47645967.0,
      "step": 18795
    },
    {
      "entropy": 6.048741674423217,
      "epoch": 2.1696480092325445,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004537290863053786,
      "loss": 5.7597,
      "mean_token_accuracy": 0.18909482061862945,
      "num_tokens": 47659392.0,
      "step": 18800
    },
    {
      "entropy": 6.1733301162719725,
      "epoch": 2.1702250432775534,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045370401692280455,
      "loss": 5.8423,
      "mean_token_accuracy": 0.1721069782972336,
      "num_tokens": 47672516.0,
      "step": 18805
    },
    {
      "entropy": 6.191349840164184,
      "epoch": 2.170802077322562,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045367894152960976,
      "loss": 5.8518,
      "mean_token_accuracy": 0.17733471989631652,
      "num_tokens": 47684638.0,
      "step": 18810
    },
    {
      "entropy": 6.112470579147339,
      "epoch": 2.1713791113675707,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00045365386012663744,
      "loss": 5.8048,
      "mean_token_accuracy": 0.18806953281164168,
      "num_tokens": 47697573.0,
      "step": 18815
    },
    {
      "entropy": 6.075803661346436,
      "epoch": 2.1719561454125795,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045362877271473137,
      "loss": 5.7973,
      "mean_token_accuracy": 0.18486253023147584,
      "num_tokens": 47709120.0,
      "step": 18820
    },
    {
      "entropy": 6.12700457572937,
      "epoch": 2.172533179457588,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045360367929473517,
      "loss": 5.8895,
      "mean_token_accuracy": 0.17622653245925904,
      "num_tokens": 47722279.0,
      "step": 18825
    },
    {
      "entropy": 6.103784513473511,
      "epoch": 2.173110213502597,
      "grad_norm": 1.0625,
      "learning_rate": 0.000453578579867493,
      "loss": 5.7514,
      "mean_token_accuracy": 0.19038948267698289,
      "num_tokens": 47735069.0,
      "step": 18830
    },
    {
      "entropy": 6.113170862197876,
      "epoch": 2.1736872475476052,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045355347443384896,
      "loss": 5.8396,
      "mean_token_accuracy": 0.1801852747797966,
      "num_tokens": 47748370.0,
      "step": 18835
    },
    {
      "entropy": 6.091566562652588,
      "epoch": 2.174264281592614,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045352836299464755,
      "loss": 5.7469,
      "mean_token_accuracy": 0.1890873372554779,
      "num_tokens": 47761509.0,
      "step": 18840
    },
    {
      "entropy": 6.020796823501587,
      "epoch": 2.1748413156376225,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004535032455507333,
      "loss": 5.6904,
      "mean_token_accuracy": 0.1975710168480873,
      "num_tokens": 47774551.0,
      "step": 18845
    },
    {
      "entropy": 6.185377168655395,
      "epoch": 2.1754183496826314,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045347812210295107,
      "loss": 5.8719,
      "mean_token_accuracy": 0.17438095211982726,
      "num_tokens": 47788063.0,
      "step": 18850
    },
    {
      "entropy": 6.188257026672363,
      "epoch": 2.17599538372764,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045345299265214583,
      "loss": 5.8603,
      "mean_token_accuracy": 0.1772001415491104,
      "num_tokens": 47799781.0,
      "step": 18855
    },
    {
      "entropy": 6.062232971191406,
      "epoch": 2.1765724177726486,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045342785719916284,
      "loss": 5.8154,
      "mean_token_accuracy": 0.17675431370735167,
      "num_tokens": 47812066.0,
      "step": 18860
    },
    {
      "entropy": 6.094439315795898,
      "epoch": 2.177149451817657,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045340271574484755,
      "loss": 5.918,
      "mean_token_accuracy": 0.17627026587724687,
      "num_tokens": 47824943.0,
      "step": 18865
    },
    {
      "entropy": 6.129074239730835,
      "epoch": 2.177726485862666,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004533775682900454,
      "loss": 5.8468,
      "mean_token_accuracy": 0.17919820845127105,
      "num_tokens": 47836942.0,
      "step": 18870
    },
    {
      "entropy": 6.215762090682984,
      "epoch": 2.1783035199076743,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004533524148356025,
      "loss": 5.8532,
      "mean_token_accuracy": 0.18112643659114838,
      "num_tokens": 47849895.0,
      "step": 18875
    },
    {
      "entropy": 6.115733814239502,
      "epoch": 2.178880553952683,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004533272553823646,
      "loss": 5.7512,
      "mean_token_accuracy": 0.18537905514240266,
      "num_tokens": 47861452.0,
      "step": 18880
    },
    {
      "entropy": 6.101626253128051,
      "epoch": 2.179457587997692,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045330208993117816,
      "loss": 5.8842,
      "mean_token_accuracy": 0.16991196274757386,
      "num_tokens": 47873913.0,
      "step": 18885
    },
    {
      "entropy": 6.142817068099975,
      "epoch": 2.1800346220427005,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004532769184828894,
      "loss": 5.8354,
      "mean_token_accuracy": 0.17991660833358764,
      "num_tokens": 47887986.0,
      "step": 18890
    },
    {
      "entropy": 6.149009418487549,
      "epoch": 2.1806116560877093,
      "grad_norm": 1.25,
      "learning_rate": 0.0004532517410383451,
      "loss": 5.8633,
      "mean_token_accuracy": 0.17761720567941666,
      "num_tokens": 47899272.0,
      "step": 18895
    },
    {
      "entropy": 6.115488433837891,
      "epoch": 2.1811886901327178,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000453226557598392,
      "loss": 5.7216,
      "mean_token_accuracy": 0.19611804634332658,
      "num_tokens": 47911097.0,
      "step": 18900
    },
    {
      "entropy": 6.099393606185913,
      "epoch": 2.1817657241777266,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004532013681638771,
      "loss": 5.8842,
      "mean_token_accuracy": 0.1749731719493866,
      "num_tokens": 47924034.0,
      "step": 18905
    },
    {
      "entropy": 6.10388035774231,
      "epoch": 2.182342758222735,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004531761727356478,
      "loss": 5.7763,
      "mean_token_accuracy": 0.1874849170446396,
      "num_tokens": 47937589.0,
      "step": 18910
    },
    {
      "entropy": 6.2070458889007565,
      "epoch": 2.182919792267744,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004531509713145514,
      "loss": 5.9086,
      "mean_token_accuracy": 0.1763848140835762,
      "num_tokens": 47951880.0,
      "step": 18915
    },
    {
      "entropy": 6.145628452301025,
      "epoch": 2.1834968263127523,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045312576390143557,
      "loss": 5.8121,
      "mean_token_accuracy": 0.18234873414039612,
      "num_tokens": 47964599.0,
      "step": 18920
    },
    {
      "entropy": 6.183667707443237,
      "epoch": 2.184073860357761,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00045310055049714815,
      "loss": 5.877,
      "mean_token_accuracy": 0.1756881892681122,
      "num_tokens": 47976879.0,
      "step": 18925
    },
    {
      "entropy": 6.173946571350098,
      "epoch": 2.1846508944027696,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045307533110253726,
      "loss": 5.8963,
      "mean_token_accuracy": 0.17364016473293303,
      "num_tokens": 47990462.0,
      "step": 18930
    },
    {
      "entropy": 6.1959531784057615,
      "epoch": 2.1852279284477785,
      "grad_norm": 1.125,
      "learning_rate": 0.00045305010571845096,
      "loss": 5.8086,
      "mean_token_accuracy": 0.18160933554172515,
      "num_tokens": 48002924.0,
      "step": 18935
    },
    {
      "entropy": 6.1512704372406,
      "epoch": 2.185804962492787,
      "grad_norm": 1.0,
      "learning_rate": 0.0004530248743457378,
      "loss": 5.8436,
      "mean_token_accuracy": 0.17881470173597336,
      "num_tokens": 48013826.0,
      "step": 18940
    },
    {
      "entropy": 6.144403600692749,
      "epoch": 2.1863819965377957,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004529996369852465,
      "loss": 5.8111,
      "mean_token_accuracy": 0.1831545427441597,
      "num_tokens": 48027087.0,
      "step": 18945
    },
    {
      "entropy": 6.115679121017456,
      "epoch": 2.1869590305828046,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00045297439363782576,
      "loss": 5.8539,
      "mean_token_accuracy": 0.1803695499897003,
      "num_tokens": 48041157.0,
      "step": 18950
    },
    {
      "entropy": 6.10780897140503,
      "epoch": 2.187536064627813,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004529491443043247,
      "loss": 5.7868,
      "mean_token_accuracy": 0.1803997650742531,
      "num_tokens": 48054168.0,
      "step": 18955
    },
    {
      "entropy": 6.119778728485107,
      "epoch": 2.188113098672822,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004529238889855926,
      "loss": 5.8081,
      "mean_token_accuracy": 0.18389812260866165,
      "num_tokens": 48066826.0,
      "step": 18960
    },
    {
      "entropy": 6.105274248123169,
      "epoch": 2.1886901327178303,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004528986276824789,
      "loss": 5.858,
      "mean_token_accuracy": 0.18109025359153746,
      "num_tokens": 48079788.0,
      "step": 18965
    },
    {
      "entropy": 6.165717124938965,
      "epoch": 2.189267166762839,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004528733603958331,
      "loss": 5.791,
      "mean_token_accuracy": 0.17625951170921325,
      "num_tokens": 48091517.0,
      "step": 18970
    },
    {
      "entropy": 6.079336261749267,
      "epoch": 2.1898442008078476,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004528480871265053,
      "loss": 5.7784,
      "mean_token_accuracy": 0.18088893294334413,
      "num_tokens": 48104964.0,
      "step": 18975
    },
    {
      "entropy": 6.12470760345459,
      "epoch": 2.1904212348528564,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045282280787534537,
      "loss": 5.834,
      "mean_token_accuracy": 0.18543060570955278,
      "num_tokens": 48118232.0,
      "step": 18980
    },
    {
      "entropy": 6.1677028179168705,
      "epoch": 2.190998268897865,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004527975226432036,
      "loss": 5.8834,
      "mean_token_accuracy": 0.1806299716234207,
      "num_tokens": 48131320.0,
      "step": 18985
    },
    {
      "entropy": 6.13905439376831,
      "epoch": 2.1915753029428737,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045277223143093057,
      "loss": 5.8848,
      "mean_token_accuracy": 0.17591743618249894,
      "num_tokens": 48143432.0,
      "step": 18990
    },
    {
      "entropy": 6.100077056884766,
      "epoch": 2.192152336987882,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045274693423937674,
      "loss": 5.8294,
      "mean_token_accuracy": 0.18974421173334122,
      "num_tokens": 48158205.0,
      "step": 18995
    },
    {
      "entropy": 6.128110265731811,
      "epoch": 2.192729371032891,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045272163106939314,
      "loss": 5.8039,
      "mean_token_accuracy": 0.18135794699192048,
      "num_tokens": 48169225.0,
      "step": 19000
    },
    {
      "entropy": 6.198502826690674,
      "epoch": 2.1933064050778994,
      "grad_norm": 0.984375,
      "learning_rate": 0.00045269632192183076,
      "loss": 5.9413,
      "mean_token_accuracy": 0.16954950392246246,
      "num_tokens": 48182715.0,
      "step": 19005
    },
    {
      "entropy": 6.103990507125855,
      "epoch": 2.1938834391229083,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045267100679754075,
      "loss": 5.7727,
      "mean_token_accuracy": 0.18646319061517716,
      "num_tokens": 48195817.0,
      "step": 19010
    },
    {
      "entropy": 6.080235862731934,
      "epoch": 2.1944604731679167,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004526456856973748,
      "loss": 5.8017,
      "mean_token_accuracy": 0.18107461780309678,
      "num_tokens": 48208347.0,
      "step": 19015
    },
    {
      "entropy": 6.112344598770141,
      "epoch": 2.1950375072129256,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004526203586221844,
      "loss": 5.8006,
      "mean_token_accuracy": 0.18088992238044738,
      "num_tokens": 48221202.0,
      "step": 19020
    },
    {
      "entropy": 6.043634986877441,
      "epoch": 2.1956145412579344,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045259502557282145,
      "loss": 5.7344,
      "mean_token_accuracy": 0.18655781596899032,
      "num_tokens": 48233471.0,
      "step": 19025
    },
    {
      "entropy": 6.07668833732605,
      "epoch": 2.196191575302943,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004525696865501381,
      "loss": 5.7858,
      "mean_token_accuracy": 0.18457195907831192,
      "num_tokens": 48246685.0,
      "step": 19030
    },
    {
      "entropy": 6.143138456344604,
      "epoch": 2.1967686093479517,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004525443415549865,
      "loss": 5.7945,
      "mean_token_accuracy": 0.18536327183246612,
      "num_tokens": 48259942.0,
      "step": 19035
    },
    {
      "entropy": 6.163250589370728,
      "epoch": 2.19734564339296,
      "grad_norm": 1.125,
      "learning_rate": 0.00045251899058821915,
      "loss": 5.8058,
      "mean_token_accuracy": 0.18221941590309143,
      "num_tokens": 48272628.0,
      "step": 19040
    },
    {
      "entropy": 6.104145812988281,
      "epoch": 2.197922677437969,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004524936336506887,
      "loss": 5.7422,
      "mean_token_accuracy": 0.19140811860561371,
      "num_tokens": 48286463.0,
      "step": 19045
    },
    {
      "entropy": 6.12087082862854,
      "epoch": 2.1984997114829774,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004524682707432482,
      "loss": 5.8517,
      "mean_token_accuracy": 0.18094166815280915,
      "num_tokens": 48299074.0,
      "step": 19050
    },
    {
      "entropy": 6.180783891677857,
      "epoch": 2.1990767455279863,
      "grad_norm": 1.125,
      "learning_rate": 0.00045244290186675034,
      "loss": 5.9229,
      "mean_token_accuracy": 0.18032517284154892,
      "num_tokens": 48312419.0,
      "step": 19055
    },
    {
      "entropy": 6.183932495117188,
      "epoch": 2.1996537795729947,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004524175270220489,
      "loss": 5.8424,
      "mean_token_accuracy": 0.17773233056068422,
      "num_tokens": 48325085.0,
      "step": 19060
    },
    {
      "entropy": 6.229330396652221,
      "epoch": 2.2002308136180035,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00045239214620999683,
      "loss": 5.8795,
      "mean_token_accuracy": 0.1801878347992897,
      "num_tokens": 48337461.0,
      "step": 19065
    },
    {
      "entropy": 6.070000791549683,
      "epoch": 2.200807847663012,
      "grad_norm": 1.125,
      "learning_rate": 0.0004523667594314481,
      "loss": 5.8449,
      "mean_token_accuracy": 0.18882401138544083,
      "num_tokens": 48351851.0,
      "step": 19070
    },
    {
      "entropy": 6.101251173019409,
      "epoch": 2.201384881708021,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045234136668725655,
      "loss": 5.8336,
      "mean_token_accuracy": 0.18505766987800598,
      "num_tokens": 48363443.0,
      "step": 19075
    },
    {
      "entropy": 6.065783405303955,
      "epoch": 2.2019619157530292,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045231596797827616,
      "loss": 5.7764,
      "mean_token_accuracy": 0.1769060656428337,
      "num_tokens": 48376757.0,
      "step": 19080
    },
    {
      "entropy": 6.066257667541504,
      "epoch": 2.202538949798038,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00045229056330536134,
      "loss": 5.729,
      "mean_token_accuracy": 0.18573248833417894,
      "num_tokens": 48390127.0,
      "step": 19085
    },
    {
      "entropy": 6.213972330093384,
      "epoch": 2.2031159838430465,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045226515266936644,
      "loss": 5.8448,
      "mean_token_accuracy": 0.1765856370329857,
      "num_tokens": 48401760.0,
      "step": 19090
    },
    {
      "entropy": 6.149196147918701,
      "epoch": 2.2036930178880554,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004522397360711462,
      "loss": 5.8852,
      "mean_token_accuracy": 0.17650120109319686,
      "num_tokens": 48414291.0,
      "step": 19095
    },
    {
      "entropy": 6.119310569763184,
      "epoch": 2.2042700519330642,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004522143135115555,
      "loss": 5.8287,
      "mean_token_accuracy": 0.18463417440652846,
      "num_tokens": 48428818.0,
      "step": 19100
    },
    {
      "entropy": 6.107091665267944,
      "epoch": 2.2048470859780727,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045218888499144933,
      "loss": 5.8652,
      "mean_token_accuracy": 0.1763901188969612,
      "num_tokens": 48442425.0,
      "step": 19105
    },
    {
      "entropy": 6.045023965835571,
      "epoch": 2.2054241200230815,
      "grad_norm": 2.109375,
      "learning_rate": 0.00045216345051168314,
      "loss": 5.7402,
      "mean_token_accuracy": 0.18459443747997284,
      "num_tokens": 48454814.0,
      "step": 19110
    },
    {
      "entropy": 6.163419151306153,
      "epoch": 2.20600115406809,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004521380100731122,
      "loss": 5.7499,
      "mean_token_accuracy": 0.1823705717921257,
      "num_tokens": 48467211.0,
      "step": 19115
    },
    {
      "entropy": 6.122723007202149,
      "epoch": 2.206578188113099,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045211256367659234,
      "loss": 5.7407,
      "mean_token_accuracy": 0.1881166934967041,
      "num_tokens": 48479252.0,
      "step": 19120
    },
    {
      "entropy": 6.069229602813721,
      "epoch": 2.207155222158107,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004520871113229793,
      "loss": 5.7899,
      "mean_token_accuracy": 0.18455973118543625,
      "num_tokens": 48491227.0,
      "step": 19125
    },
    {
      "entropy": 6.163376665115356,
      "epoch": 2.207732256203116,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045206165301312927,
      "loss": 5.8273,
      "mean_token_accuracy": 0.18483382612466812,
      "num_tokens": 48503374.0,
      "step": 19130
    },
    {
      "entropy": 6.072341442108154,
      "epoch": 2.2083092902481245,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004520361887478985,
      "loss": 5.7897,
      "mean_token_accuracy": 0.184389565885067,
      "num_tokens": 48515294.0,
      "step": 19135
    },
    {
      "entropy": 6.140119934082032,
      "epoch": 2.2088863242931334,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004520107185281435,
      "loss": 5.8646,
      "mean_token_accuracy": 0.1763157442212105,
      "num_tokens": 48528429.0,
      "step": 19140
    },
    {
      "entropy": 6.166807317733765,
      "epoch": 2.209463358338142,
      "grad_norm": 1.125,
      "learning_rate": 0.0004519852423547208,
      "loss": 5.8091,
      "mean_token_accuracy": 0.17584212273359298,
      "num_tokens": 48541077.0,
      "step": 19145
    },
    {
      "entropy": 6.0820800304412845,
      "epoch": 2.2100403923831506,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004519597602284875,
      "loss": 5.7768,
      "mean_token_accuracy": 0.19034706503152848,
      "num_tokens": 48555338.0,
      "step": 19150
    },
    {
      "entropy": 6.120675659179687,
      "epoch": 2.210617426428159,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004519342721503005,
      "loss": 5.884,
      "mean_token_accuracy": 0.17891746908426284,
      "num_tokens": 48568374.0,
      "step": 19155
    },
    {
      "entropy": 6.140750169754028,
      "epoch": 2.211194460473168,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004519087781210171,
      "loss": 5.8079,
      "mean_token_accuracy": 0.17806962728500367,
      "num_tokens": 48582412.0,
      "step": 19160
    },
    {
      "entropy": 6.179685020446778,
      "epoch": 2.2117714945181763,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000451883278141495,
      "loss": 5.8126,
      "mean_token_accuracy": 0.17811158895492554,
      "num_tokens": 48594942.0,
      "step": 19165
    },
    {
      "entropy": 6.010459136962891,
      "epoch": 2.212348528563185,
      "grad_norm": 0.921875,
      "learning_rate": 0.00045185777221259157,
      "loss": 5.635,
      "mean_token_accuracy": 0.20024493038654329,
      "num_tokens": 48605748.0,
      "step": 19170
    },
    {
      "entropy": 6.104791212081909,
      "epoch": 2.212925562608194,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004518322603351648,
      "loss": 5.9389,
      "mean_token_accuracy": 0.17579580396413802,
      "num_tokens": 48617894.0,
      "step": 19175
    },
    {
      "entropy": 6.1805943012237545,
      "epoch": 2.2135025966532025,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045180674251007287,
      "loss": 5.8147,
      "mean_token_accuracy": 0.17808202505111695,
      "num_tokens": 48631237.0,
      "step": 19180
    },
    {
      "entropy": 6.097504329681397,
      "epoch": 2.2140796306982113,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045178121873817395,
      "loss": 5.8374,
      "mean_token_accuracy": 0.17988955080509186,
      "num_tokens": 48643218.0,
      "step": 19185
    },
    {
      "entropy": 6.114158248901367,
      "epoch": 2.2146566647432198,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004517556890203265,
      "loss": 5.8528,
      "mean_token_accuracy": 0.1775271937251091,
      "num_tokens": 48655281.0,
      "step": 19190
    },
    {
      "entropy": 6.204569149017334,
      "epoch": 2.2152336987882286,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045173015335738925,
      "loss": 5.9747,
      "mean_token_accuracy": 0.17318729013204576,
      "num_tokens": 48669301.0,
      "step": 19195
    },
    {
      "entropy": 6.143696212768555,
      "epoch": 2.215810732833237,
      "grad_norm": 0.96875,
      "learning_rate": 0.00045170461175022116,
      "loss": 5.8055,
      "mean_token_accuracy": 0.1817542150616646,
      "num_tokens": 48681966.0,
      "step": 19200
    },
    {
      "entropy": 6.105476713180542,
      "epoch": 2.216387766878246,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004516790641996812,
      "loss": 5.8437,
      "mean_token_accuracy": 0.17949002087116242,
      "num_tokens": 48695414.0,
      "step": 19205
    },
    {
      "entropy": 6.167020702362061,
      "epoch": 2.2169648009232543,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004516535107066286,
      "loss": 5.8715,
      "mean_token_accuracy": 0.17665734589099885,
      "num_tokens": 48707953.0,
      "step": 19210
    },
    {
      "entropy": 6.143227672576904,
      "epoch": 2.217541834968263,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045162795127192296,
      "loss": 5.807,
      "mean_token_accuracy": 0.1873381718993187,
      "num_tokens": 48720524.0,
      "step": 19215
    },
    {
      "entropy": 6.141962099075317,
      "epoch": 2.2181188690132716,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000451602385896424,
      "loss": 5.8588,
      "mean_token_accuracy": 0.18284614086151124,
      "num_tokens": 48732544.0,
      "step": 19220
    },
    {
      "entropy": 6.220880031585693,
      "epoch": 2.2186959030582805,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00045157681458099145,
      "loss": 5.9216,
      "mean_token_accuracy": 0.17523401081562043,
      "num_tokens": 48745520.0,
      "step": 19225
    },
    {
      "entropy": 6.206246709823608,
      "epoch": 2.2192729371032893,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004515512373264854,
      "loss": 5.8868,
      "mean_token_accuracy": 0.17550836503505707,
      "num_tokens": 48757442.0,
      "step": 19230
    },
    {
      "entropy": 6.116058731079102,
      "epoch": 2.2198499711482977,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045152565413376623,
      "loss": 5.8352,
      "mean_token_accuracy": 0.17994634956121444,
      "num_tokens": 48771351.0,
      "step": 19235
    },
    {
      "entropy": 6.1099464893341064,
      "epoch": 2.2204270051933066,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004515000650036944,
      "loss": 5.8564,
      "mean_token_accuracy": 0.18286526799201966,
      "num_tokens": 48783897.0,
      "step": 19240
    },
    {
      "entropy": 6.104422235488892,
      "epoch": 2.221004039238315,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004514744699371305,
      "loss": 5.7265,
      "mean_token_accuracy": 0.19128636121749878,
      "num_tokens": 48796334.0,
      "step": 19245
    },
    {
      "entropy": 6.15315465927124,
      "epoch": 2.221581073283324,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045144886893493547,
      "loss": 5.7916,
      "mean_token_accuracy": 0.1839529037475586,
      "num_tokens": 48809762.0,
      "step": 19250
    },
    {
      "entropy": 6.092352199554443,
      "epoch": 2.2221581073283323,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045142326199797047,
      "loss": 5.7674,
      "mean_token_accuracy": 0.18393055200576783,
      "num_tokens": 48822038.0,
      "step": 19255
    },
    {
      "entropy": 6.122200202941895,
      "epoch": 2.222735141373341,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045139764912709656,
      "loss": 5.8374,
      "mean_token_accuracy": 0.18118378669023513,
      "num_tokens": 48835284.0,
      "step": 19260
    },
    {
      "entropy": 6.041936826705933,
      "epoch": 2.2233121754183496,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004513720303231754,
      "loss": 5.7538,
      "mean_token_accuracy": 0.1877487674355507,
      "num_tokens": 48848673.0,
      "step": 19265
    },
    {
      "entropy": 6.128367900848389,
      "epoch": 2.2238892094633584,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045134640558706864,
      "loss": 5.8718,
      "mean_token_accuracy": 0.17527904659509658,
      "num_tokens": 48861071.0,
      "step": 19270
    },
    {
      "entropy": 6.193828344345093,
      "epoch": 2.224466243508367,
      "grad_norm": 1.015625,
      "learning_rate": 0.000451320774919638,
      "loss": 5.9185,
      "mean_token_accuracy": 0.18022517263889312,
      "num_tokens": 48873471.0,
      "step": 19275
    },
    {
      "entropy": 6.074729537963867,
      "epoch": 2.2250432775533757,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045129513832174587,
      "loss": 5.8137,
      "mean_token_accuracy": 0.18527948260307311,
      "num_tokens": 48885565.0,
      "step": 19280
    },
    {
      "entropy": 6.155221748352051,
      "epoch": 2.225620311598384,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045126949579425414,
      "loss": 5.8322,
      "mean_token_accuracy": 0.17720213234424592,
      "num_tokens": 48897785.0,
      "step": 19285
    },
    {
      "entropy": 6.225070667266846,
      "epoch": 2.226197345643393,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045124384733802563,
      "loss": 5.8702,
      "mean_token_accuracy": 0.17325304299592972,
      "num_tokens": 48910202.0,
      "step": 19290
    },
    {
      "entropy": 6.061574983596802,
      "epoch": 2.2267743796884014,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004512181929539228,
      "loss": 5.7141,
      "mean_token_accuracy": 0.1902824103832245,
      "num_tokens": 48923643.0,
      "step": 19295
    },
    {
      "entropy": 6.173430252075195,
      "epoch": 2.2273514137334103,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045119253264280855,
      "loss": 5.8264,
      "mean_token_accuracy": 0.1795138657093048,
      "num_tokens": 48936513.0,
      "step": 19300
    },
    {
      "entropy": 6.091540431976318,
      "epoch": 2.227928447778419,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045116686640554607,
      "loss": 5.8125,
      "mean_token_accuracy": 0.18175228387117387,
      "num_tokens": 48950082.0,
      "step": 19305
    },
    {
      "entropy": 6.085782861709594,
      "epoch": 2.2285054818234276,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045114119424299843,
      "loss": 5.7714,
      "mean_token_accuracy": 0.18514148741960526,
      "num_tokens": 48961773.0,
      "step": 19310
    },
    {
      "entropy": 6.090428590774536,
      "epoch": 2.2290825158684364,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004511155161560293,
      "loss": 5.8191,
      "mean_token_accuracy": 0.18495211750268936,
      "num_tokens": 48974171.0,
      "step": 19315
    },
    {
      "entropy": 6.191088724136352,
      "epoch": 2.229659549913445,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00045108983214550225,
      "loss": 5.8463,
      "mean_token_accuracy": 0.17630040943622588,
      "num_tokens": 48987913.0,
      "step": 19320
    },
    {
      "entropy": 6.100524950027466,
      "epoch": 2.2302365839584537,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004510641422122811,
      "loss": 5.7495,
      "mean_token_accuracy": 0.18056867122650147,
      "num_tokens": 49001189.0,
      "step": 19325
    },
    {
      "entropy": 6.053651618957519,
      "epoch": 2.230813618003462,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00045103844635723005,
      "loss": 5.736,
      "mean_token_accuracy": 0.18226418793201446,
      "num_tokens": 49013827.0,
      "step": 19330
    },
    {
      "entropy": 6.092554807662964,
      "epoch": 2.231390652048471,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045101274458121335,
      "loss": 5.7897,
      "mean_token_accuracy": 0.1840750440955162,
      "num_tokens": 49024939.0,
      "step": 19335
    },
    {
      "entropy": 6.097409057617187,
      "epoch": 2.2319676860934794,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045098703688509537,
      "loss": 5.8558,
      "mean_token_accuracy": 0.17802898585796356,
      "num_tokens": 49037707.0,
      "step": 19340
    },
    {
      "entropy": 6.145604467391967,
      "epoch": 2.2325447201384883,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004509613232697408,
      "loss": 5.7687,
      "mean_token_accuracy": 0.1886694461107254,
      "num_tokens": 49051408.0,
      "step": 19345
    },
    {
      "entropy": 6.058500146865844,
      "epoch": 2.2331217541834967,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004509356037360145,
      "loss": 5.735,
      "mean_token_accuracy": 0.1868857592344284,
      "num_tokens": 49063664.0,
      "step": 19350
    },
    {
      "entropy": 6.053864002227783,
      "epoch": 2.2336987882285055,
      "grad_norm": 1.125,
      "learning_rate": 0.0004509098782847816,
      "loss": 5.798,
      "mean_token_accuracy": 0.1825099840760231,
      "num_tokens": 49076321.0,
      "step": 19355
    },
    {
      "entropy": 6.1475207805633545,
      "epoch": 2.234275822273514,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004508841469169073,
      "loss": 5.8352,
      "mean_token_accuracy": 0.17924839854240418,
      "num_tokens": 49088794.0,
      "step": 19360
    },
    {
      "entropy": 6.182398319244385,
      "epoch": 2.234852856318523,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045085840963325716,
      "loss": 5.8519,
      "mean_token_accuracy": 0.1822988510131836,
      "num_tokens": 49099986.0,
      "step": 19365
    },
    {
      "entropy": 6.066998052597046,
      "epoch": 2.2354298903635312,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004508326664346967,
      "loss": 5.7399,
      "mean_token_accuracy": 0.18655147701501845,
      "num_tokens": 49112811.0,
      "step": 19370
    },
    {
      "entropy": 6.048188018798828,
      "epoch": 2.23600692440854,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045080691732209176,
      "loss": 5.7911,
      "mean_token_accuracy": 0.1894754260778427,
      "num_tokens": 49124893.0,
      "step": 19375
    },
    {
      "entropy": 6.022425127029419,
      "epoch": 2.236583958453549,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045078116229630864,
      "loss": 5.8074,
      "mean_token_accuracy": 0.1799091085791588,
      "num_tokens": 49137239.0,
      "step": 19380
    },
    {
      "entropy": 6.150869274139405,
      "epoch": 2.2371609924985574,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045075540135821343,
      "loss": 5.8507,
      "mean_token_accuracy": 0.18135159462690353,
      "num_tokens": 49149876.0,
      "step": 19385
    },
    {
      "entropy": 6.177823257446289,
      "epoch": 2.2377380265435662,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004507296345086725,
      "loss": 5.8008,
      "mean_token_accuracy": 0.18128003180027008,
      "num_tokens": 49161599.0,
      "step": 19390
    },
    {
      "entropy": 6.124496793746948,
      "epoch": 2.2383150605885747,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004507038617485526,
      "loss": 5.8714,
      "mean_token_accuracy": 0.1766341209411621,
      "num_tokens": 49174278.0,
      "step": 19395
    },
    {
      "entropy": 6.156163549423217,
      "epoch": 2.2388920946335835,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00045067808307872064,
      "loss": 5.8449,
      "mean_token_accuracy": 0.18504138439893722,
      "num_tokens": 49187922.0,
      "step": 19400
    },
    {
      "entropy": 6.169332361221313,
      "epoch": 2.239469128678592,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045065229850004365,
      "loss": 5.8419,
      "mean_token_accuracy": 0.174832259118557,
      "num_tokens": 49200109.0,
      "step": 19405
    },
    {
      "entropy": 6.07872428894043,
      "epoch": 2.240046162723601,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004506265080133888,
      "loss": 5.818,
      "mean_token_accuracy": 0.1815360352396965,
      "num_tokens": 49213192.0,
      "step": 19410
    },
    {
      "entropy": 6.123006343841553,
      "epoch": 2.240623196768609,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00045060071161962364,
      "loss": 5.882,
      "mean_token_accuracy": 0.1734763726592064,
      "num_tokens": 49225439.0,
      "step": 19415
    },
    {
      "entropy": 6.080710506439209,
      "epoch": 2.241200230813618,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004505749093196158,
      "loss": 5.8424,
      "mean_token_accuracy": 0.1825777694582939,
      "num_tokens": 49240445.0,
      "step": 19420
    },
    {
      "entropy": 6.09651894569397,
      "epoch": 2.2417772648586265,
      "grad_norm": 1.15625,
      "learning_rate": 0.000450549101114233,
      "loss": 5.8183,
      "mean_token_accuracy": 0.176598259806633,
      "num_tokens": 49253909.0,
      "step": 19425
    },
    {
      "entropy": 6.096200704574585,
      "epoch": 2.2423542989036354,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004505232870043434,
      "loss": 5.727,
      "mean_token_accuracy": 0.18663746863603592,
      "num_tokens": 49266816.0,
      "step": 19430
    },
    {
      "entropy": 6.12756576538086,
      "epoch": 2.2429313329486438,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004504974669908152,
      "loss": 5.8103,
      "mean_token_accuracy": 0.18294149041175842,
      "num_tokens": 49279561.0,
      "step": 19435
    },
    {
      "entropy": 6.116925001144409,
      "epoch": 2.2435083669936526,
      "grad_norm": 1.484375,
      "learning_rate": 0.00045047164107451696,
      "loss": 5.9311,
      "mean_token_accuracy": 0.17216714322566987,
      "num_tokens": 49292448.0,
      "step": 19440
    },
    {
      "entropy": 6.189856004714966,
      "epoch": 2.244085401038661,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004504458092563172,
      "loss": 5.835,
      "mean_token_accuracy": 0.1812010243535042,
      "num_tokens": 49305078.0,
      "step": 19445
    },
    {
      "entropy": 6.093616533279419,
      "epoch": 2.24466243508367,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00045041997153708475,
      "loss": 5.7381,
      "mean_token_accuracy": 0.19003266990184783,
      "num_tokens": 49317353.0,
      "step": 19450
    },
    {
      "entropy": 6.025535440444946,
      "epoch": 2.2452394691286788,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00045039412791768877,
      "loss": 5.8018,
      "mean_token_accuracy": 0.18812852054834367,
      "num_tokens": 49329730.0,
      "step": 19455
    },
    {
      "entropy": 6.125547599792481,
      "epoch": 2.245816503173687,
      "grad_norm": 1.0,
      "learning_rate": 0.00045036827839899833,
      "loss": 5.8147,
      "mean_token_accuracy": 0.17289612889289857,
      "num_tokens": 49342656.0,
      "step": 19460
    },
    {
      "entropy": 6.026174592971802,
      "epoch": 2.246393537218696,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00045034242298188303,
      "loss": 5.7082,
      "mean_token_accuracy": 0.19131626486778258,
      "num_tokens": 49355216.0,
      "step": 19465
    },
    {
      "entropy": 6.085362815856934,
      "epoch": 2.2469705712637045,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004503165616672124,
      "loss": 5.7566,
      "mean_token_accuracy": 0.18321236371994018,
      "num_tokens": 49367897.0,
      "step": 19470
    },
    {
      "entropy": 6.078060865402222,
      "epoch": 2.2475476053087133,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004502906944558563,
      "loss": 5.7678,
      "mean_token_accuracy": 0.18487814217805862,
      "num_tokens": 49380509.0,
      "step": 19475
    },
    {
      "entropy": 6.185996627807617,
      "epoch": 2.2481246393537218,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004502648213486848,
      "loss": 5.9404,
      "mean_token_accuracy": 0.17641634494066238,
      "num_tokens": 49393832.0,
      "step": 19480
    },
    {
      "entropy": 6.188489198684692,
      "epoch": 2.2487016733987306,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045023894234656807,
      "loss": 5.8716,
      "mean_token_accuracy": 0.181330144405365,
      "num_tokens": 49406568.0,
      "step": 19485
    },
    {
      "entropy": 6.111005878448486,
      "epoch": 2.249278707443739,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004502130574503766,
      "loss": 5.8119,
      "mean_token_accuracy": 0.1793781876564026,
      "num_tokens": 49419171.0,
      "step": 19490
    },
    {
      "entropy": 6.025286674499512,
      "epoch": 2.249855741488748,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004501871666609809,
      "loss": 5.7597,
      "mean_token_accuracy": 0.19144954085350036,
      "num_tokens": 49431177.0,
      "step": 19495
    },
    {
      "entropy": 6.009071111679077,
      "epoch": 2.2504327755337563,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004501612699792519,
      "loss": 5.7038,
      "mean_token_accuracy": 0.1906972736120224,
      "num_tokens": 49443403.0,
      "step": 19500
    },
    {
      "entropy": 6.088020515441895,
      "epoch": 2.251009809578765,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004501353674060606,
      "loss": 5.8691,
      "mean_token_accuracy": 0.17330747246742248,
      "num_tokens": 49456338.0,
      "step": 19505
    },
    {
      "entropy": 6.062865829467773,
      "epoch": 2.251586843623774,
      "grad_norm": 1.25,
      "learning_rate": 0.0004501094589422782,
      "loss": 5.7333,
      "mean_token_accuracy": 0.19372376799583435,
      "num_tokens": 49470946.0,
      "step": 19510
    },
    {
      "entropy": 6.091143274307251,
      "epoch": 2.2521638776687825,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045008354458877614,
      "loss": 5.7733,
      "mean_token_accuracy": 0.18841186463832854,
      "num_tokens": 49483174.0,
      "step": 19515
    },
    {
      "entropy": 6.0468419075012205,
      "epoch": 2.252740911713791,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000450057624346426,
      "loss": 5.8043,
      "mean_token_accuracy": 0.18490788638591765,
      "num_tokens": 49495099.0,
      "step": 19520
    },
    {
      "entropy": 6.052328157424927,
      "epoch": 2.2533179457587997,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045003169821609967,
      "loss": 5.7743,
      "mean_token_accuracy": 0.17967904806137086,
      "num_tokens": 49507200.0,
      "step": 19525
    },
    {
      "entropy": 6.126697826385498,
      "epoch": 2.2538949798038086,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00045000576619866914,
      "loss": 5.8616,
      "mean_token_accuracy": 0.18034757375717164,
      "num_tokens": 49520093.0,
      "step": 19530
    },
    {
      "entropy": 6.077826976776123,
      "epoch": 2.254472013848817,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044997982829500657,
      "loss": 5.7995,
      "mean_token_accuracy": 0.18329958766698837,
      "num_tokens": 49532190.0,
      "step": 19535
    },
    {
      "entropy": 6.1029016971588135,
      "epoch": 2.255049047893826,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00044995388450598436,
      "loss": 5.7952,
      "mean_token_accuracy": 0.18855469971895217,
      "num_tokens": 49544468.0,
      "step": 19540
    },
    {
      "entropy": 6.0529341220855715,
      "epoch": 2.2556260819388343,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004499279348324751,
      "loss": 5.7516,
      "mean_token_accuracy": 0.18441680818796158,
      "num_tokens": 49556753.0,
      "step": 19545
    },
    {
      "entropy": 6.166997623443604,
      "epoch": 2.256203115983843,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044990197927535173,
      "loss": 5.7981,
      "mean_token_accuracy": 0.18828649818897247,
      "num_tokens": 49567756.0,
      "step": 19550
    },
    {
      "entropy": 6.072061395645141,
      "epoch": 2.2567801500288516,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044987601783548703,
      "loss": 5.8158,
      "mean_token_accuracy": 0.18020875751972198,
      "num_tokens": 49579996.0,
      "step": 19555
    },
    {
      "entropy": 6.13876576423645,
      "epoch": 2.2573571840738604,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004498500505137545,
      "loss": 5.8572,
      "mean_token_accuracy": 0.1764445498585701,
      "num_tokens": 49593407.0,
      "step": 19560
    },
    {
      "entropy": 6.233310174942017,
      "epoch": 2.257934218118869,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004498240773110273,
      "loss": 5.8834,
      "mean_token_accuracy": 0.17397699356079102,
      "num_tokens": 49606231.0,
      "step": 19565
    },
    {
      "entropy": 6.178919744491577,
      "epoch": 2.2585112521638777,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004497980982281791,
      "loss": 5.8194,
      "mean_token_accuracy": 0.17556500136852266,
      "num_tokens": 49616860.0,
      "step": 19570
    },
    {
      "entropy": 6.013501834869385,
      "epoch": 2.259088286208886,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004497721132660837,
      "loss": 5.7406,
      "mean_token_accuracy": 0.18569550067186355,
      "num_tokens": 49630415.0,
      "step": 19575
    },
    {
      "entropy": 6.047288131713867,
      "epoch": 2.259665320253895,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00044974612242561516,
      "loss": 5.8116,
      "mean_token_accuracy": 0.17592115253210067,
      "num_tokens": 49643173.0,
      "step": 19580
    },
    {
      "entropy": 6.217005062103271,
      "epoch": 2.260242354298904,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004497201257076475,
      "loss": 5.8777,
      "mean_token_accuracy": 0.17529686987400056,
      "num_tokens": 49656670.0,
      "step": 19585
    },
    {
      "entropy": 6.119595527648926,
      "epoch": 2.2608193883439123,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004496941231130552,
      "loss": 5.8102,
      "mean_token_accuracy": 0.186619570851326,
      "num_tokens": 49669572.0,
      "step": 19590
    },
    {
      "entropy": 6.1548559188842775,
      "epoch": 2.261396422388921,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004496681146427129,
      "loss": 5.8275,
      "mean_token_accuracy": 0.18289176523685455,
      "num_tokens": 49682253.0,
      "step": 19595
    },
    {
      "entropy": 6.118901491165161,
      "epoch": 2.2619734564339296,
      "grad_norm": 1.484375,
      "learning_rate": 0.00044964210029749523,
      "loss": 5.743,
      "mean_token_accuracy": 0.19030014425516129,
      "num_tokens": 49695284.0,
      "step": 19600
    },
    {
      "entropy": 6.073218727111817,
      "epoch": 2.2625504904789384,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044961608007827736,
      "loss": 5.7959,
      "mean_token_accuracy": 0.18105938881635666,
      "num_tokens": 49708039.0,
      "step": 19605
    },
    {
      "entropy": 6.076639842987061,
      "epoch": 2.263127524523947,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004495900539859344,
      "loss": 5.7957,
      "mean_token_accuracy": 0.18886059522628784,
      "num_tokens": 49722182.0,
      "step": 19610
    },
    {
      "entropy": 6.119974756240845,
      "epoch": 2.2637045585689557,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044956402202134157,
      "loss": 5.8025,
      "mean_token_accuracy": 0.18158803284168243,
      "num_tokens": 49733875.0,
      "step": 19615
    },
    {
      "entropy": 5.935592555999756,
      "epoch": 2.264281592613964,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044953798418537465,
      "loss": 5.6036,
      "mean_token_accuracy": 0.2064524844288826,
      "num_tokens": 49746701.0,
      "step": 19620
    },
    {
      "entropy": 6.05517086982727,
      "epoch": 2.264858626658973,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004495119404789093,
      "loss": 5.7259,
      "mean_token_accuracy": 0.18063586056232453,
      "num_tokens": 49760636.0,
      "step": 19625
    },
    {
      "entropy": 6.180433464050293,
      "epoch": 2.2654356607039814,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004494858909028216,
      "loss": 5.8432,
      "mean_token_accuracy": 0.17799864560365677,
      "num_tokens": 49773655.0,
      "step": 19630
    },
    {
      "entropy": 6.068809175491333,
      "epoch": 2.2660126947489903,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004494598354579875,
      "loss": 5.8789,
      "mean_token_accuracy": 0.1755536064505577,
      "num_tokens": 49787575.0,
      "step": 19635
    },
    {
      "entropy": 6.077509927749634,
      "epoch": 2.2665897287939987,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004494337741452836,
      "loss": 5.8322,
      "mean_token_accuracy": 0.1799464166164398,
      "num_tokens": 49802021.0,
      "step": 19640
    },
    {
      "entropy": 6.182551670074463,
      "epoch": 2.2671667628390075,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004494077069655863,
      "loss": 5.7489,
      "mean_token_accuracy": 0.1883100837469101,
      "num_tokens": 49813640.0,
      "step": 19645
    },
    {
      "entropy": 6.183745861053467,
      "epoch": 2.267743796884016,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004493816339197724,
      "loss": 5.8838,
      "mean_token_accuracy": 0.17708800584077836,
      "num_tokens": 49826318.0,
      "step": 19650
    },
    {
      "entropy": 6.097109174728393,
      "epoch": 2.268320830929025,
      "grad_norm": 1.125,
      "learning_rate": 0.00044935555500871897,
      "loss": 5.8297,
      "mean_token_accuracy": 0.180152927339077,
      "num_tokens": 49839278.0,
      "step": 19655
    },
    {
      "entropy": 6.161598205566406,
      "epoch": 2.2688978649740337,
      "grad_norm": 1.046875,
      "learning_rate": 0.000449329470233303,
      "loss": 5.8566,
      "mean_token_accuracy": 0.17958616018295287,
      "num_tokens": 49851417.0,
      "step": 19660
    },
    {
      "entropy": 6.1300993919372555,
      "epoch": 2.269474899019042,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044930337959440183,
      "loss": 5.7666,
      "mean_token_accuracy": 0.17614748626947402,
      "num_tokens": 49863797.0,
      "step": 19665
    },
    {
      "entropy": 6.11533203125,
      "epoch": 2.270051933064051,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004492772830928931,
      "loss": 5.8146,
      "mean_token_accuracy": 0.17867905646562576,
      "num_tokens": 49877767.0,
      "step": 19670
    },
    {
      "entropy": 6.128215360641479,
      "epoch": 2.2706289671090594,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044925118072965456,
      "loss": 5.8553,
      "mean_token_accuracy": 0.18150849491357804,
      "num_tokens": 49890301.0,
      "step": 19675
    },
    {
      "entropy": 6.098849725723267,
      "epoch": 2.2712060011540682,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000449225072505564,
      "loss": 5.8187,
      "mean_token_accuracy": 0.18519050925970076,
      "num_tokens": 49902095.0,
      "step": 19680
    },
    {
      "entropy": 6.186308193206787,
      "epoch": 2.2717830351990767,
      "grad_norm": 1.125,
      "learning_rate": 0.00044919895842149976,
      "loss": 5.8364,
      "mean_token_accuracy": 0.1799546256661415,
      "num_tokens": 49914764.0,
      "step": 19685
    },
    {
      "entropy": 6.127132987976074,
      "epoch": 2.2723600692440855,
      "grad_norm": 1.0,
      "learning_rate": 0.00044917283847834005,
      "loss": 5.7983,
      "mean_token_accuracy": 0.18316694498062133,
      "num_tokens": 49927729.0,
      "step": 19690
    },
    {
      "entropy": 6.128943061828613,
      "epoch": 2.272937103289094,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004491467126769635,
      "loss": 5.7986,
      "mean_token_accuracy": 0.18748575896024705,
      "num_tokens": 49940017.0,
      "step": 19695
    },
    {
      "entropy": 6.15049467086792,
      "epoch": 2.273514137334103,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044912058101824866,
      "loss": 5.8853,
      "mean_token_accuracy": 0.17876375019550322,
      "num_tokens": 49951894.0,
      "step": 19700
    },
    {
      "entropy": 6.073190307617187,
      "epoch": 2.274091171379111,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044909444350307463,
      "loss": 5.6492,
      "mean_token_accuracy": 0.19940230697393418,
      "num_tokens": 49965152.0,
      "step": 19705
    },
    {
      "entropy": 6.099264001846313,
      "epoch": 2.27466820542412,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004490683001323205,
      "loss": 5.8297,
      "mean_token_accuracy": 0.1818332254886627,
      "num_tokens": 49978713.0,
      "step": 19710
    },
    {
      "entropy": 6.114779281616211,
      "epoch": 2.2752452394691285,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044904215090686554,
      "loss": 5.8525,
      "mean_token_accuracy": 0.1718704342842102,
      "num_tokens": 49990090.0,
      "step": 19715
    },
    {
      "entropy": 6.162533330917358,
      "epoch": 2.2758222735141374,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044901599582758926,
      "loss": 5.8268,
      "mean_token_accuracy": 0.1829460307955742,
      "num_tokens": 50002205.0,
      "step": 19720
    },
    {
      "entropy": 6.061874341964722,
      "epoch": 2.2763993075591458,
      "grad_norm": 0.984375,
      "learning_rate": 0.00044898983489537143,
      "loss": 5.7108,
      "mean_token_accuracy": 0.18731240481138228,
      "num_tokens": 50014274.0,
      "step": 19725
    },
    {
      "entropy": 6.160400819778443,
      "epoch": 2.2769763416041546,
      "grad_norm": 1.046875,
      "learning_rate": 0.000448963668111092,
      "loss": 5.9261,
      "mean_token_accuracy": 0.17160774767398834,
      "num_tokens": 50026789.0,
      "step": 19730
    },
    {
      "entropy": 6.155254650115967,
      "epoch": 2.2775533756491635,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00044893749547563085,
      "loss": 5.8061,
      "mean_token_accuracy": 0.1880965054035187,
      "num_tokens": 50039615.0,
      "step": 19735
    },
    {
      "entropy": 6.07856936454773,
      "epoch": 2.278130409694172,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044891131698986846,
      "loss": 5.8195,
      "mean_token_accuracy": 0.18321505784988404,
      "num_tokens": 50051799.0,
      "step": 19740
    },
    {
      "entropy": 6.136298036575317,
      "epoch": 2.2787074437391808,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004488851326546854,
      "loss": 5.8193,
      "mean_token_accuracy": 0.18427640199661255,
      "num_tokens": 50064774.0,
      "step": 19745
    },
    {
      "entropy": 6.151190137863159,
      "epoch": 2.279284477784189,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004488589424709622,
      "loss": 5.8816,
      "mean_token_accuracy": 0.1748761311173439,
      "num_tokens": 50076631.0,
      "step": 19750
    },
    {
      "entropy": 6.0982747077941895,
      "epoch": 2.279861511829198,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004488327464395799,
      "loss": 5.7256,
      "mean_token_accuracy": 0.18104265481233597,
      "num_tokens": 50089493.0,
      "step": 19755
    },
    {
      "entropy": 6.102081155776977,
      "epoch": 2.2804385458742065,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004488065445614195,
      "loss": 5.7755,
      "mean_token_accuracy": 0.18780233711004257,
      "num_tokens": 50101412.0,
      "step": 19760
    },
    {
      "entropy": 6.106775617599487,
      "epoch": 2.2810155799192153,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004487803368373623,
      "loss": 5.7845,
      "mean_token_accuracy": 0.1838479071855545,
      "num_tokens": 50113634.0,
      "step": 19765
    },
    {
      "entropy": 6.090753030776978,
      "epoch": 2.2815926139642237,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004487541232682898,
      "loss": 5.8184,
      "mean_token_accuracy": 0.18962326198816298,
      "num_tokens": 50125771.0,
      "step": 19770
    },
    {
      "entropy": 6.1379376411437985,
      "epoch": 2.2821696480092326,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004487279038550836,
      "loss": 5.8522,
      "mean_token_accuracy": 0.17495402991771697,
      "num_tokens": 50138260.0,
      "step": 19775
    },
    {
      "entropy": 6.131827259063721,
      "epoch": 2.282746682054241,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004487016785986258,
      "loss": 5.8604,
      "mean_token_accuracy": 0.18200754672288894,
      "num_tokens": 50150813.0,
      "step": 19780
    },
    {
      "entropy": 6.096352243423462,
      "epoch": 2.28332371609925,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044867544749979817,
      "loss": 5.7374,
      "mean_token_accuracy": 0.18722015470266343,
      "num_tokens": 50163775.0,
      "step": 19785
    },
    {
      "entropy": 6.082024431228637,
      "epoch": 2.2839007501442588,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044864921055948317,
      "loss": 5.761,
      "mean_token_accuracy": 0.18260778337717057,
      "num_tokens": 50175697.0,
      "step": 19790
    },
    {
      "entropy": 6.105894136428833,
      "epoch": 2.284477784189267,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00044862296777856326,
      "loss": 5.8636,
      "mean_token_accuracy": 0.1824580177664757,
      "num_tokens": 50187826.0,
      "step": 19795
    },
    {
      "entropy": 6.159018611907959,
      "epoch": 2.2850548182342756,
      "grad_norm": 1.0,
      "learning_rate": 0.0004485967191579211,
      "loss": 5.8546,
      "mean_token_accuracy": 0.18229743987321853,
      "num_tokens": 50201611.0,
      "step": 19800
    },
    {
      "entropy": 6.118442487716675,
      "epoch": 2.2856318522792844,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004485704646984394,
      "loss": 5.7878,
      "mean_token_accuracy": 0.18008687049150468,
      "num_tokens": 50213341.0,
      "step": 19805
    },
    {
      "entropy": 6.068079233169556,
      "epoch": 2.2862088863242933,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004485442044010015,
      "loss": 5.8173,
      "mean_token_accuracy": 0.17838032990694047,
      "num_tokens": 50225165.0,
      "step": 19810
    },
    {
      "entropy": 6.082869148254394,
      "epoch": 2.2867859203693017,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004485179382664904,
      "loss": 5.702,
      "mean_token_accuracy": 0.18459280878305434,
      "num_tokens": 50238207.0,
      "step": 19815
    },
    {
      "entropy": 6.141095876693726,
      "epoch": 2.2873629544143106,
      "grad_norm": 3.84375,
      "learning_rate": 0.0004484916662957896,
      "loss": 5.7102,
      "mean_token_accuracy": 0.18362041115760802,
      "num_tokens": 50250347.0,
      "step": 19820
    },
    {
      "entropy": 6.111587190628052,
      "epoch": 2.287939988459319,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004484653884897829,
      "loss": 5.8035,
      "mean_token_accuracy": 0.1830605000257492,
      "num_tokens": 50263487.0,
      "step": 19825
    },
    {
      "entropy": 6.067152500152588,
      "epoch": 2.288517022504328,
      "grad_norm": 0.984375,
      "learning_rate": 0.00044843910484935394,
      "loss": 5.7396,
      "mean_token_accuracy": 0.18552889227867125,
      "num_tokens": 50275020.0,
      "step": 19830
    },
    {
      "entropy": 6.052919816970825,
      "epoch": 2.2890940565493363,
      "grad_norm": 1.0,
      "learning_rate": 0.0004484128153753868,
      "loss": 5.8325,
      "mean_token_accuracy": 0.18259522020816804,
      "num_tokens": 50287888.0,
      "step": 19835
    },
    {
      "entropy": 6.171370792388916,
      "epoch": 2.289671090594345,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044838652006876583,
      "loss": 5.832,
      "mean_token_accuracy": 0.17860480695962905,
      "num_tokens": 50301271.0,
      "step": 19840
    },
    {
      "entropy": 6.201105308532715,
      "epoch": 2.2902481246393536,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00044836021893037537,
      "loss": 5.852,
      "mean_token_accuracy": 0.18510336726903914,
      "num_tokens": 50312707.0,
      "step": 19845
    },
    {
      "entropy": 6.14098973274231,
      "epoch": 2.2908251586843624,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004483339119611001,
      "loss": 5.9157,
      "mean_token_accuracy": 0.17628826573491096,
      "num_tokens": 50325399.0,
      "step": 19850
    },
    {
      "entropy": 6.1501805782318115,
      "epoch": 2.291402192729371,
      "grad_norm": 1.796875,
      "learning_rate": 0.00044830759916182476,
      "loss": 5.787,
      "mean_token_accuracy": 0.18098410815000535,
      "num_tokens": 50338424.0,
      "step": 19855
    },
    {
      "entropy": 6.090538072586059,
      "epoch": 2.2919792267743797,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004482812805334344,
      "loss": 5.7863,
      "mean_token_accuracy": 0.18063820749521256,
      "num_tokens": 50351183.0,
      "step": 19860
    },
    {
      "entropy": 6.122755289077759,
      "epoch": 2.2925562608193886,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004482549560768142,
      "loss": 5.8086,
      "mean_token_accuracy": 0.1823497787117958,
      "num_tokens": 50364223.0,
      "step": 19865
    },
    {
      "entropy": 6.1533876895904545,
      "epoch": 2.293133294864397,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004482286257928497,
      "loss": 5.8448,
      "mean_token_accuracy": 0.17838067710399627,
      "num_tokens": 50376867.0,
      "step": 19870
    },
    {
      "entropy": 6.166789197921753,
      "epoch": 2.293710328909406,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004482022896824263,
      "loss": 5.7669,
      "mean_token_accuracy": 0.17984747141599655,
      "num_tokens": 50389180.0,
      "step": 19875
    },
    {
      "entropy": 6.1716163635253904,
      "epoch": 2.2942873629544143,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044817594774643004,
      "loss": 5.8174,
      "mean_token_accuracy": 0.1789398357272148,
      "num_tokens": 50401286.0,
      "step": 19880
    },
    {
      "entropy": 6.074377822875976,
      "epoch": 2.294864396999423,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044814959998574675,
      "loss": 5.8575,
      "mean_token_accuracy": 0.18021982461214064,
      "num_tokens": 50414215.0,
      "step": 19885
    },
    {
      "entropy": 6.110835123062134,
      "epoch": 2.2954414310444315,
      "grad_norm": 1.0,
      "learning_rate": 0.00044812324640126265,
      "loss": 5.7566,
      "mean_token_accuracy": 0.18258639723062514,
      "num_tokens": 50425672.0,
      "step": 19890
    },
    {
      "entropy": 6.166446161270142,
      "epoch": 2.2960184650894404,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004480968869938642,
      "loss": 5.8027,
      "mean_token_accuracy": 0.1866215795278549,
      "num_tokens": 50438578.0,
      "step": 19895
    },
    {
      "entropy": 6.085659980773926,
      "epoch": 2.296595499134449,
      "grad_norm": 1.453125,
      "learning_rate": 0.00044807052176443793,
      "loss": 5.749,
      "mean_token_accuracy": 0.18506508320569992,
      "num_tokens": 50451762.0,
      "step": 19900
    },
    {
      "entropy": 6.12231822013855,
      "epoch": 2.2971725331794577,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044804415071387067,
      "loss": 5.7874,
      "mean_token_accuracy": 0.17883945852518082,
      "num_tokens": 50464064.0,
      "step": 19905
    },
    {
      "entropy": 6.130839204788208,
      "epoch": 2.297749567224466,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004480177738430492,
      "loss": 5.8344,
      "mean_token_accuracy": 0.17545308023691178,
      "num_tokens": 50476843.0,
      "step": 19910
    },
    {
      "entropy": 6.1197669982910154,
      "epoch": 2.298326601269475,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044799139115286104,
      "loss": 5.767,
      "mean_token_accuracy": 0.1800383970141411,
      "num_tokens": 50488209.0,
      "step": 19915
    },
    {
      "entropy": 6.167664527893066,
      "epoch": 2.2989036353144834,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004479650026441933,
      "loss": 5.8427,
      "mean_token_accuracy": 0.17660924345254897,
      "num_tokens": 50500292.0,
      "step": 19920
    },
    {
      "entropy": 6.1808525085449215,
      "epoch": 2.2994806693594922,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044793860831793356,
      "loss": 5.8502,
      "mean_token_accuracy": 0.17803215682506562,
      "num_tokens": 50512563.0,
      "step": 19925
    },
    {
      "entropy": 6.177944135665894,
      "epoch": 2.3000577034045007,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00044791220817496963,
      "loss": 5.7966,
      "mean_token_accuracy": 0.1832399845123291,
      "num_tokens": 50525614.0,
      "step": 19930
    },
    {
      "entropy": 6.085981845855713,
      "epoch": 2.3006347374495095,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004478858022161895,
      "loss": 5.7192,
      "mean_token_accuracy": 0.18133516609668732,
      "num_tokens": 50537387.0,
      "step": 19935
    },
    {
      "entropy": 6.065998649597168,
      "epoch": 2.3012117714945184,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004478593904424813,
      "loss": 5.7575,
      "mean_token_accuracy": 0.18911363035440446,
      "num_tokens": 50549588.0,
      "step": 19940
    },
    {
      "entropy": 6.134940910339355,
      "epoch": 2.301788805539527,
      "grad_norm": 1.0,
      "learning_rate": 0.0004478329728547334,
      "loss": 5.9427,
      "mean_token_accuracy": 0.17148082405328752,
      "num_tokens": 50562276.0,
      "step": 19945
    },
    {
      "entropy": 6.155119848251343,
      "epoch": 2.3023658395845357,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00044780654945383424,
      "loss": 5.8444,
      "mean_token_accuracy": 0.17707214802503585,
      "num_tokens": 50575650.0,
      "step": 19950
    },
    {
      "entropy": 6.1881026268005375,
      "epoch": 2.302942873629544,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00044778012024067267,
      "loss": 5.8357,
      "mean_token_accuracy": 0.1766630232334137,
      "num_tokens": 50587611.0,
      "step": 19955
    },
    {
      "entropy": 6.1483336925506595,
      "epoch": 2.303519907674553,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004477536852161376,
      "loss": 5.8238,
      "mean_token_accuracy": 0.17419526726007462,
      "num_tokens": 50601001.0,
      "step": 19960
    },
    {
      "entropy": 6.172666740417481,
      "epoch": 2.3040969417195614,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004477272443811181,
      "loss": 5.8553,
      "mean_token_accuracy": 0.1727649450302124,
      "num_tokens": 50613539.0,
      "step": 19965
    },
    {
      "entropy": 6.062262535095215,
      "epoch": 2.3046739757645702,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004477007977365035,
      "loss": 5.7017,
      "mean_token_accuracy": 0.18586247861385347,
      "num_tokens": 50626536.0,
      "step": 19970
    },
    {
      "entropy": 6.131066513061524,
      "epoch": 2.3052510098095786,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004476743452831834,
      "loss": 5.8013,
      "mean_token_accuracy": 0.17664381265640258,
      "num_tokens": 50638617.0,
      "step": 19975
    },
    {
      "entropy": 6.184109210968018,
      "epoch": 2.3058280438545875,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044764788702204747,
      "loss": 5.8598,
      "mean_token_accuracy": 0.17941330075263978,
      "num_tokens": 50651779.0,
      "step": 19980
    },
    {
      "entropy": 6.067663621902466,
      "epoch": 2.306405077899596,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004476214229539856,
      "loss": 5.7454,
      "mean_token_accuracy": 0.18609073460102082,
      "num_tokens": 50665205.0,
      "step": 19985
    },
    {
      "entropy": 6.164358711242675,
      "epoch": 2.306982111944605,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004475949530798879,
      "loss": 5.8525,
      "mean_token_accuracy": 0.17694531381130219,
      "num_tokens": 50677621.0,
      "step": 19990
    },
    {
      "entropy": 6.185609245300293,
      "epoch": 2.307559145989613,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044756847740064475,
      "loss": 5.87,
      "mean_token_accuracy": 0.17396206110715867,
      "num_tokens": 50689544.0,
      "step": 19995
    },
    {
      "entropy": 6.226226949691773,
      "epoch": 2.308136180034622,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004475419959171465,
      "loss": 5.8469,
      "mean_token_accuracy": 0.177403062582016,
      "num_tokens": 50702733.0,
      "step": 20000
    },
    {
      "entropy": 6.06727819442749,
      "epoch": 2.3087132140796305,
      "grad_norm": 0.9375,
      "learning_rate": 0.000447515508630284,
      "loss": 5.7668,
      "mean_token_accuracy": 0.18631421029567719,
      "num_tokens": 50716296.0,
      "step": 20005
    },
    {
      "entropy": 6.0449587345123295,
      "epoch": 2.3092902481246393,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044748901554094806,
      "loss": 5.7248,
      "mean_token_accuracy": 0.18688036799430846,
      "num_tokens": 50729170.0,
      "step": 20010
    },
    {
      "entropy": 6.1286224842071535,
      "epoch": 2.309867282169648,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004474625166500297,
      "loss": 5.7861,
      "mean_token_accuracy": 0.1844197168946266,
      "num_tokens": 50741400.0,
      "step": 20015
    },
    {
      "entropy": 6.098959684371948,
      "epoch": 2.3104443162146566,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00044743601195842026,
      "loss": 5.8502,
      "mean_token_accuracy": 0.1793159067630768,
      "num_tokens": 50755231.0,
      "step": 20020
    },
    {
      "entropy": 6.13970456123352,
      "epoch": 2.3110213502596655,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044740950146701127,
      "loss": 5.8475,
      "mean_token_accuracy": 0.1720214605331421,
      "num_tokens": 50768044.0,
      "step": 20025
    },
    {
      "entropy": 6.188294410705566,
      "epoch": 2.311598384304674,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004473829851766943,
      "loss": 5.8842,
      "mean_token_accuracy": 0.17273385375738143,
      "num_tokens": 50780693.0,
      "step": 20030
    },
    {
      "entropy": 6.087898588180542,
      "epoch": 2.3121754183496828,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004473564630883612,
      "loss": 5.7449,
      "mean_token_accuracy": 0.18191283494234084,
      "num_tokens": 50792829.0,
      "step": 20035
    },
    {
      "entropy": 6.067259502410889,
      "epoch": 2.312752452394691,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004473299352029042,
      "loss": 5.713,
      "mean_token_accuracy": 0.18662918657064437,
      "num_tokens": 50806045.0,
      "step": 20040
    },
    {
      "entropy": 6.132836723327637,
      "epoch": 2.3133294864397,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004473034015212154,
      "loss": 5.8248,
      "mean_token_accuracy": 0.18288834244012833,
      "num_tokens": 50818789.0,
      "step": 20045
    },
    {
      "entropy": 6.194453191757202,
      "epoch": 2.3139065204847085,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004472768620441872,
      "loss": 5.8515,
      "mean_token_accuracy": 0.17958282679319382,
      "num_tokens": 50830681.0,
      "step": 20050
    },
    {
      "entropy": 6.06901330947876,
      "epoch": 2.3144835545297173,
      "grad_norm": 0.9375,
      "learning_rate": 0.00044725031677271234,
      "loss": 5.743,
      "mean_token_accuracy": 0.1875415861606598,
      "num_tokens": 50843162.0,
      "step": 20055
    },
    {
      "entropy": 6.060148859024048,
      "epoch": 2.3150605885747257,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004472237657076837,
      "loss": 5.7863,
      "mean_token_accuracy": 0.18670614063739777,
      "num_tokens": 50855339.0,
      "step": 20060
    },
    {
      "entropy": 6.178205585479736,
      "epoch": 2.3156376226197346,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004471972088499942,
      "loss": 5.8485,
      "mean_token_accuracy": 0.18234241008758545,
      "num_tokens": 50868885.0,
      "step": 20065
    },
    {
      "entropy": 6.119196462631225,
      "epoch": 2.3162146566647435,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004471706462005371,
      "loss": 5.7823,
      "mean_token_accuracy": 0.1797006294131279,
      "num_tokens": 50881027.0,
      "step": 20070
    },
    {
      "entropy": 6.075886964797974,
      "epoch": 2.316791690709752,
      "grad_norm": 1.125,
      "learning_rate": 0.0004471440777602059,
      "loss": 5.7448,
      "mean_token_accuracy": 0.1896383136510849,
      "num_tokens": 50893717.0,
      "step": 20075
    },
    {
      "entropy": 6.135770177841186,
      "epoch": 2.3173687247547603,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044711750352989407,
      "loss": 5.8451,
      "mean_token_accuracy": 0.1809339255094528,
      "num_tokens": 50904965.0,
      "step": 20080
    },
    {
      "entropy": 6.151015758514404,
      "epoch": 2.317945758799769,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004470909235104956,
      "loss": 5.87,
      "mean_token_accuracy": 0.18046355843544007,
      "num_tokens": 50917804.0,
      "step": 20085
    },
    {
      "entropy": 6.105581855773925,
      "epoch": 2.318522792844778,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004470643377029043,
      "loss": 5.8041,
      "mean_token_accuracy": 0.18260424882173537,
      "num_tokens": 50931124.0,
      "step": 20090
    },
    {
      "entropy": 6.10461688041687,
      "epoch": 2.3190998268897864,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004470377461080145,
      "loss": 5.7691,
      "mean_token_accuracy": 0.18184901475906373,
      "num_tokens": 50942983.0,
      "step": 20095
    },
    {
      "entropy": 6.146995735168457,
      "epoch": 2.3196768609347953,
      "grad_norm": 1.0,
      "learning_rate": 0.0004470111487267206,
      "loss": 5.8768,
      "mean_token_accuracy": 0.17984736561775208,
      "num_tokens": 50956282.0,
      "step": 20100
    },
    {
      "entropy": 6.184567022323608,
      "epoch": 2.3202538949798037,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004469845455599171,
      "loss": 5.8524,
      "mean_token_accuracy": 0.17381148785352707,
      "num_tokens": 50968112.0,
      "step": 20105
    },
    {
      "entropy": 6.05695219039917,
      "epoch": 2.3208309290248126,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004469579366084989,
      "loss": 5.6447,
      "mean_token_accuracy": 0.1921035349369049,
      "num_tokens": 50980377.0,
      "step": 20110
    },
    {
      "entropy": 6.150827646255493,
      "epoch": 2.321407963069821,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044693132187336094,
      "loss": 5.8347,
      "mean_token_accuracy": 0.17843090295791625,
      "num_tokens": 50993371.0,
      "step": 20115
    },
    {
      "entropy": 6.171275281906128,
      "epoch": 2.32198499711483,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004469047013553983,
      "loss": 5.8269,
      "mean_token_accuracy": 0.18262094706296922,
      "num_tokens": 51004409.0,
      "step": 20120
    },
    {
      "entropy": 6.065983772277832,
      "epoch": 2.3225620311598383,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00044687807505550646,
      "loss": 5.7138,
      "mean_token_accuracy": 0.19565078616142273,
      "num_tokens": 51017890.0,
      "step": 20125
    },
    {
      "entropy": 6.132612943649292,
      "epoch": 2.323139065204847,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044685144297458096,
      "loss": 5.8101,
      "mean_token_accuracy": 0.18011171072721482,
      "num_tokens": 51029734.0,
      "step": 20130
    },
    {
      "entropy": 6.125143194198609,
      "epoch": 2.3237160992498556,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044682480511351754,
      "loss": 5.8016,
      "mean_token_accuracy": 0.18687545508146286,
      "num_tokens": 51041876.0,
      "step": 20135
    },
    {
      "entropy": 6.121161937713623,
      "epoch": 2.3242931332948644,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004467981614732121,
      "loss": 5.8289,
      "mean_token_accuracy": 0.17989543378353118,
      "num_tokens": 51054603.0,
      "step": 20140
    },
    {
      "entropy": 6.1470324993133545,
      "epoch": 2.3248701673398733,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044677151205456086,
      "loss": 5.8044,
      "mean_token_accuracy": 0.1842400163412094,
      "num_tokens": 51067283.0,
      "step": 20145
    },
    {
      "entropy": 6.143877840042114,
      "epoch": 2.3254472013848817,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004467448568584601,
      "loss": 5.849,
      "mean_token_accuracy": 0.18063582628965377,
      "num_tokens": 51079841.0,
      "step": 20150
    },
    {
      "entropy": 6.048275327682495,
      "epoch": 2.32602423542989,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004467181958858064,
      "loss": 5.6743,
      "mean_token_accuracy": 0.1975342407822609,
      "num_tokens": 51091873.0,
      "step": 20155
    },
    {
      "entropy": 6.138991975784302,
      "epoch": 2.326601269474899,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004466915291374965,
      "loss": 5.8274,
      "mean_token_accuracy": 0.18261861354112624,
      "num_tokens": 51104955.0,
      "step": 20160
    },
    {
      "entropy": 6.155609655380249,
      "epoch": 2.327178303519908,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004466648566144273,
      "loss": 5.8375,
      "mean_token_accuracy": 0.1808705970644951,
      "num_tokens": 51117699.0,
      "step": 20165
    },
    {
      "entropy": 6.149796772003174,
      "epoch": 2.3277553375649163,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000446638178317496,
      "loss": 5.789,
      "mean_token_accuracy": 0.18843137472867966,
      "num_tokens": 51130322.0,
      "step": 20170
    },
    {
      "entropy": 6.186602830886841,
      "epoch": 2.328332371609925,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044661149424759974,
      "loss": 5.8664,
      "mean_token_accuracy": 0.18201289772987367,
      "num_tokens": 51144685.0,
      "step": 20175
    },
    {
      "entropy": 6.1225347995758055,
      "epoch": 2.3289094056549335,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004465848044056361,
      "loss": 5.8263,
      "mean_token_accuracy": 0.1751604363322258,
      "num_tokens": 51157653.0,
      "step": 20180
    },
    {
      "entropy": 6.180042695999146,
      "epoch": 2.3294864396999424,
      "grad_norm": 1.5,
      "learning_rate": 0.0004465581087925028,
      "loss": 5.8739,
      "mean_token_accuracy": 0.1779229000210762,
      "num_tokens": 51171314.0,
      "step": 20185
    },
    {
      "entropy": 6.0366394996643065,
      "epoch": 2.330063473744951,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004465314074090978,
      "loss": 5.7029,
      "mean_token_accuracy": 0.18800501823425292,
      "num_tokens": 51184261.0,
      "step": 20190
    },
    {
      "entropy": 6.170687961578369,
      "epoch": 2.3306405077899597,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044650470025631904,
      "loss": 5.9,
      "mean_token_accuracy": 0.1817521408200264,
      "num_tokens": 51195908.0,
      "step": 20195
    },
    {
      "entropy": 6.164652967453003,
      "epoch": 2.331217541834968,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004464779873350649,
      "loss": 5.8301,
      "mean_token_accuracy": 0.17846509963274002,
      "num_tokens": 51208275.0,
      "step": 20200
    },
    {
      "entropy": 6.159370756149292,
      "epoch": 2.331794575879977,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004464512686462339,
      "loss": 5.779,
      "mean_token_accuracy": 0.17974087744951248,
      "num_tokens": 51221331.0,
      "step": 20205
    },
    {
      "entropy": 6.110942649841308,
      "epoch": 2.3323716099249854,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044642454419072455,
      "loss": 5.817,
      "mean_token_accuracy": 0.18145658820867538,
      "num_tokens": 51233875.0,
      "step": 20210
    },
    {
      "entropy": 6.159333086013794,
      "epoch": 2.3329486439699942,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004463978139694358,
      "loss": 5.827,
      "mean_token_accuracy": 0.17795737385749816,
      "num_tokens": 51246397.0,
      "step": 20215
    },
    {
      "entropy": 6.122809267044067,
      "epoch": 2.333525678015003,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044637107798326675,
      "loss": 5.8155,
      "mean_token_accuracy": 0.17720494717359542,
      "num_tokens": 51260823.0,
      "step": 20220
    },
    {
      "entropy": 6.093409776687622,
      "epoch": 2.3341027120600115,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004463443362331166,
      "loss": 5.7433,
      "mean_token_accuracy": 0.18534752875566482,
      "num_tokens": 51273093.0,
      "step": 20225
    },
    {
      "entropy": 6.163913822174072,
      "epoch": 2.3346797461050204,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00044631758871988486,
      "loss": 5.7953,
      "mean_token_accuracy": 0.18886077404022217,
      "num_tokens": 51284971.0,
      "step": 20230
    },
    {
      "entropy": 6.1402663230896,
      "epoch": 2.335256780150029,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004462908354444711,
      "loss": 5.9128,
      "mean_token_accuracy": 0.18556461930274964,
      "num_tokens": 51296989.0,
      "step": 20235
    },
    {
      "entropy": 6.139644002914428,
      "epoch": 2.3358338141950377,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004462640764077751,
      "loss": 5.8326,
      "mean_token_accuracy": 0.18049248903989792,
      "num_tokens": 51309022.0,
      "step": 20240
    },
    {
      "entropy": 6.197623062133789,
      "epoch": 2.336410848240046,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004462373116106971,
      "loss": 5.937,
      "mean_token_accuracy": 0.17476667016744613,
      "num_tokens": 51322481.0,
      "step": 20245
    },
    {
      "entropy": 6.211710214614868,
      "epoch": 2.336987882285055,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044621054105413704,
      "loss": 5.8352,
      "mean_token_accuracy": 0.17905059605836868,
      "num_tokens": 51334499.0,
      "step": 20250
    },
    {
      "entropy": 6.134792327880859,
      "epoch": 2.3375649163300634,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044618376473899555,
      "loss": 5.7876,
      "mean_token_accuracy": 0.1791130542755127,
      "num_tokens": 51346489.0,
      "step": 20255
    },
    {
      "entropy": 6.220087194442749,
      "epoch": 2.338141950375072,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004461569826661732,
      "loss": 5.9089,
      "mean_token_accuracy": 0.17656247168779374,
      "num_tokens": 51359624.0,
      "step": 20260
    },
    {
      "entropy": 6.1995124340057375,
      "epoch": 2.3387189844200806,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004461301948365707,
      "loss": 5.8697,
      "mean_token_accuracy": 0.17985248267650605,
      "num_tokens": 51373768.0,
      "step": 20265
    },
    {
      "entropy": 6.08966646194458,
      "epoch": 2.3392960184650895,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004461034012510891,
      "loss": 5.8646,
      "mean_token_accuracy": 0.1691257730126381,
      "num_tokens": 51386160.0,
      "step": 20270
    },
    {
      "entropy": 6.198019075393677,
      "epoch": 2.339873052510098,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00044607660191062963,
      "loss": 5.8212,
      "mean_token_accuracy": 0.1847350165247917,
      "num_tokens": 51397579.0,
      "step": 20275
    },
    {
      "entropy": 6.07816801071167,
      "epoch": 2.340450086555107,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00044604979681609364,
      "loss": 5.7736,
      "mean_token_accuracy": 0.179922017455101,
      "num_tokens": 51409925.0,
      "step": 20280
    },
    {
      "entropy": 6.118048858642578,
      "epoch": 2.341027120600115,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044602298596838264,
      "loss": 5.8405,
      "mean_token_accuracy": 0.1788712352514267,
      "num_tokens": 51422840.0,
      "step": 20285
    },
    {
      "entropy": 6.166372537612915,
      "epoch": 2.341604154645124,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044599616936839853,
      "loss": 5.7468,
      "mean_token_accuracy": 0.18387030363082885,
      "num_tokens": 51434626.0,
      "step": 20290
    },
    {
      "entropy": 6.143677663803101,
      "epoch": 2.342181188690133,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004459693470170432,
      "loss": 5.7893,
      "mean_token_accuracy": 0.18312890082597733,
      "num_tokens": 51446358.0,
      "step": 20295
    },
    {
      "entropy": 6.095878314971924,
      "epoch": 2.3427582227351413,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004459425189152187,
      "loss": 5.8486,
      "mean_token_accuracy": 0.17457541972398757,
      "num_tokens": 51459029.0,
      "step": 20300
    },
    {
      "entropy": 6.137482166290283,
      "epoch": 2.34333525678015,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00044591568506382757,
      "loss": 5.7421,
      "mean_token_accuracy": 0.1879407659173012,
      "num_tokens": 51471250.0,
      "step": 20305
    },
    {
      "entropy": 6.124959659576416,
      "epoch": 2.3439122908251586,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00044588884546377226,
      "loss": 5.813,
      "mean_token_accuracy": 0.18058087378740312,
      "num_tokens": 51482456.0,
      "step": 20310
    },
    {
      "entropy": 6.076003170013427,
      "epoch": 2.3444893248701675,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004458620001159555,
      "loss": 5.7507,
      "mean_token_accuracy": 0.18680650293827056,
      "num_tokens": 51495413.0,
      "step": 20315
    },
    {
      "entropy": 6.200300979614258,
      "epoch": 2.345066358915176,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004458351490212803,
      "loss": 5.808,
      "mean_token_accuracy": 0.18212637603282927,
      "num_tokens": 51507068.0,
      "step": 20320
    },
    {
      "entropy": 6.009555149078369,
      "epoch": 2.3456433929601848,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044580829218064964,
      "loss": 5.6724,
      "mean_token_accuracy": 0.1920401856303215,
      "num_tokens": 51519553.0,
      "step": 20325
    },
    {
      "entropy": 6.0924159526824955,
      "epoch": 2.346220427005193,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000445781429594967,
      "loss": 5.7843,
      "mean_token_accuracy": 0.18538253754377365,
      "num_tokens": 51531027.0,
      "step": 20330
    },
    {
      "entropy": 6.11998028755188,
      "epoch": 2.346797461050202,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004457545612651357,
      "loss": 5.779,
      "mean_token_accuracy": 0.1800786241889,
      "num_tokens": 51543883.0,
      "step": 20335
    },
    {
      "entropy": 6.125153541564941,
      "epoch": 2.3473744950952105,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044572768719205964,
      "loss": 5.7937,
      "mean_token_accuracy": 0.18416469544172287,
      "num_tokens": 51557514.0,
      "step": 20340
    },
    {
      "entropy": 6.131639289855957,
      "epoch": 2.3479515291402193,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044570080737664264,
      "loss": 5.8154,
      "mean_token_accuracy": 0.18182428926229477,
      "num_tokens": 51569990.0,
      "step": 20345
    },
    {
      "entropy": 6.171053218841553,
      "epoch": 2.3485285631852277,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044567392181978874,
      "loss": 5.8113,
      "mean_token_accuracy": 0.18461856842041016,
      "num_tokens": 51581794.0,
      "step": 20350
    },
    {
      "entropy": 6.087177896499634,
      "epoch": 2.3491055972302366,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00044564703052240223,
      "loss": 5.719,
      "mean_token_accuracy": 0.1859034538269043,
      "num_tokens": 51594121.0,
      "step": 20355
    },
    {
      "entropy": 6.1076192378997805,
      "epoch": 2.349682631275245,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004456201334853876,
      "loss": 5.7459,
      "mean_token_accuracy": 0.1899031490087509,
      "num_tokens": 51606418.0,
      "step": 20360
    },
    {
      "entropy": 6.127715396881103,
      "epoch": 2.350259665320254,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044559323070964956,
      "loss": 5.8468,
      "mean_token_accuracy": 0.17938563972711563,
      "num_tokens": 51619529.0,
      "step": 20365
    },
    {
      "entropy": 6.14036283493042,
      "epoch": 2.3508366993652627,
      "grad_norm": 0.984375,
      "learning_rate": 0.000445566322196093,
      "loss": 5.8944,
      "mean_token_accuracy": 0.16992413401603698,
      "num_tokens": 51631724.0,
      "step": 20370
    },
    {
      "entropy": 6.197034549713135,
      "epoch": 2.351413733410271,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004455394079456228,
      "loss": 5.8119,
      "mean_token_accuracy": 0.17992000579833983,
      "num_tokens": 51643908.0,
      "step": 20375
    },
    {
      "entropy": 6.194287633895874,
      "epoch": 2.35199076745528,
      "grad_norm": 0.96875,
      "learning_rate": 0.00044551248795914446,
      "loss": 5.8751,
      "mean_token_accuracy": 0.17746468335390092,
      "num_tokens": 51654873.0,
      "step": 20380
    },
    {
      "entropy": 6.098720455169678,
      "epoch": 2.3525678015002884,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004454855622375632,
      "loss": 5.7758,
      "mean_token_accuracy": 0.18222216069698333,
      "num_tokens": 51667765.0,
      "step": 20385
    },
    {
      "entropy": 6.0380621433258055,
      "epoch": 2.3531448355452973,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004454586307817848,
      "loss": 5.7956,
      "mean_token_accuracy": 0.1861286550760269,
      "num_tokens": 51680950.0,
      "step": 20390
    },
    {
      "entropy": 6.117996311187744,
      "epoch": 2.3537218695903057,
      "grad_norm": 1.03125,
      "learning_rate": 0.000445431693592715,
      "loss": 5.7574,
      "mean_token_accuracy": 0.19613803774118424,
      "num_tokens": 51696437.0,
      "step": 20395
    },
    {
      "entropy": 6.108854055404663,
      "epoch": 2.3542989036353146,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004454047506712598,
      "loss": 5.8279,
      "mean_token_accuracy": 0.1779678449034691,
      "num_tokens": 51709142.0,
      "step": 20400
    },
    {
      "entropy": 6.01763596534729,
      "epoch": 2.354875937680323,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044537780201832545,
      "loss": 5.7428,
      "mean_token_accuracy": 0.18623868376016617,
      "num_tokens": 51721197.0,
      "step": 20405
    },
    {
      "entropy": 6.16437783241272,
      "epoch": 2.355452971725332,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004453508476348184,
      "loss": 5.8272,
      "mean_token_accuracy": 0.18277889341115952,
      "num_tokens": 51734921.0,
      "step": 20410
    },
    {
      "entropy": 6.1244378089904785,
      "epoch": 2.3560300057703403,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004453238875216452,
      "loss": 5.8062,
      "mean_token_accuracy": 0.18378682285547257,
      "num_tokens": 51747143.0,
      "step": 20415
    },
    {
      "entropy": 6.13089919090271,
      "epoch": 2.356607039815349,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004452969216797127,
      "loss": 5.8143,
      "mean_token_accuracy": 0.1732479974627495,
      "num_tokens": 51760203.0,
      "step": 20420
    },
    {
      "entropy": 6.174628162384034,
      "epoch": 2.357184073860358,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004452699501099277,
      "loss": 5.8164,
      "mean_token_accuracy": 0.18061094731092453,
      "num_tokens": 51772287.0,
      "step": 20425
    },
    {
      "entropy": 6.142187976837159,
      "epoch": 2.3577611079053664,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044524297281319766,
      "loss": 5.846,
      "mean_token_accuracy": 0.1800048053264618,
      "num_tokens": 51783893.0,
      "step": 20430
    },
    {
      "entropy": 6.176952695846557,
      "epoch": 2.358338141950375,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00044521598979042963,
      "loss": 5.8818,
      "mean_token_accuracy": 0.1868714153766632,
      "num_tokens": 51795482.0,
      "step": 20435
    },
    {
      "entropy": 6.132177829742432,
      "epoch": 2.3589151759953837,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044518900104253154,
      "loss": 5.7973,
      "mean_token_accuracy": 0.18500807136297226,
      "num_tokens": 51810306.0,
      "step": 20440
    },
    {
      "entropy": 6.1797784805297855,
      "epoch": 2.3594922100403926,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004451620065704108,
      "loss": 5.8661,
      "mean_token_accuracy": 0.17727299332618712,
      "num_tokens": 51824297.0,
      "step": 20445
    },
    {
      "entropy": 6.1277471542358395,
      "epoch": 2.360069244085401,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00044513500637497546,
      "loss": 5.7492,
      "mean_token_accuracy": 0.183962182700634,
      "num_tokens": 51835870.0,
      "step": 20450
    },
    {
      "entropy": 6.150326299667358,
      "epoch": 2.36064627813041,
      "grad_norm": 1.328125,
      "learning_rate": 0.00044510800045713373,
      "loss": 5.8432,
      "mean_token_accuracy": 0.18107542991638184,
      "num_tokens": 51847235.0,
      "step": 20455
    },
    {
      "entropy": 6.184855604171753,
      "epoch": 2.3612233121754183,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044508098881779396,
      "loss": 5.8137,
      "mean_token_accuracy": 0.18033018410205842,
      "num_tokens": 51859192.0,
      "step": 20460
    },
    {
      "entropy": 6.0830518245697025,
      "epoch": 2.361800346220427,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004450539714578645,
      "loss": 5.6947,
      "mean_token_accuracy": 0.1948489725589752,
      "num_tokens": 51872463.0,
      "step": 20465
    },
    {
      "entropy": 6.112808513641357,
      "epoch": 2.3623773802654355,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004450269483782543,
      "loss": 5.8659,
      "mean_token_accuracy": 0.17560944259166716,
      "num_tokens": 51886320.0,
      "step": 20470
    },
    {
      "entropy": 6.193296813964844,
      "epoch": 2.3629544143104444,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004449999195798721,
      "loss": 5.8334,
      "mean_token_accuracy": 0.1775003954768181,
      "num_tokens": 51898670.0,
      "step": 20475
    },
    {
      "entropy": 6.181326961517334,
      "epoch": 2.363531448355453,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044497288506362706,
      "loss": 5.8907,
      "mean_token_accuracy": 0.17409141808748246,
      "num_tokens": 51911510.0,
      "step": 20480
    },
    {
      "entropy": 6.1368450164794925,
      "epoch": 2.3641084824004617,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004449458448304284,
      "loss": 5.7865,
      "mean_token_accuracy": 0.19429122358560563,
      "num_tokens": 51923322.0,
      "step": 20485
    },
    {
      "entropy": 6.087029457092285,
      "epoch": 2.36468551644547,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044491879888118574,
      "loss": 5.7039,
      "mean_token_accuracy": 0.1871379718184471,
      "num_tokens": 51935010.0,
      "step": 20490
    },
    {
      "entropy": 6.093118000030517,
      "epoch": 2.365262550490479,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004448917472168087,
      "loss": 5.7232,
      "mean_token_accuracy": 0.1882448047399521,
      "num_tokens": 51949083.0,
      "step": 20495
    },
    {
      "entropy": 6.066104459762573,
      "epoch": 2.365839584535488,
      "grad_norm": 0.9375,
      "learning_rate": 0.00044486468983820707,
      "loss": 5.787,
      "mean_token_accuracy": 0.17642031162977218,
      "num_tokens": 51962150.0,
      "step": 20500
    },
    {
      "entropy": 6.188040781021118,
      "epoch": 2.3664166185804962,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000444837626746291,
      "loss": 5.8463,
      "mean_token_accuracy": 0.17642795592546462,
      "num_tokens": 51974867.0,
      "step": 20505
    },
    {
      "entropy": 6.14934720993042,
      "epoch": 2.366993652625505,
      "grad_norm": 1.0,
      "learning_rate": 0.0004448105579419707,
      "loss": 5.7477,
      "mean_token_accuracy": 0.1852025493979454,
      "num_tokens": 51986654.0,
      "step": 20510
    },
    {
      "entropy": 6.053664112091065,
      "epoch": 2.3675706866705135,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004447834834261567,
      "loss": 5.7819,
      "mean_token_accuracy": 0.17870779186487198,
      "num_tokens": 51999402.0,
      "step": 20515
    },
    {
      "entropy": 6.183165597915649,
      "epoch": 2.3681477207155224,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004447564031997595,
      "loss": 5.8383,
      "mean_token_accuracy": 0.18007162511348723,
      "num_tokens": 52013583.0,
      "step": 20520
    },
    {
      "entropy": 6.171697282791138,
      "epoch": 2.368724754760531,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00044472931726369,
      "loss": 5.8625,
      "mean_token_accuracy": 0.18298262655735015,
      "num_tokens": 52026892.0,
      "step": 20525
    },
    {
      "entropy": 6.057624244689942,
      "epoch": 2.3693017888055397,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044470222561885926,
      "loss": 5.7437,
      "mean_token_accuracy": 0.19220590144395827,
      "num_tokens": 52039837.0,
      "step": 20530
    },
    {
      "entropy": 6.191092443466187,
      "epoch": 2.369878822850548,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044467512826617845,
      "loss": 5.7966,
      "mean_token_accuracy": 0.18481701016426086,
      "num_tokens": 52054079.0,
      "step": 20535
    },
    {
      "entropy": 6.185373163223266,
      "epoch": 2.370455856895557,
      "grad_norm": 0.984375,
      "learning_rate": 0.00044464802520655897,
      "loss": 5.8022,
      "mean_token_accuracy": 0.17992214113473892,
      "num_tokens": 52066895.0,
      "step": 20540
    },
    {
      "entropy": 6.056500625610352,
      "epoch": 2.3710328909405654,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004446209164409124,
      "loss": 5.7931,
      "mean_token_accuracy": 0.1860213205218315,
      "num_tokens": 52078983.0,
      "step": 20545
    },
    {
      "entropy": 6.088406991958618,
      "epoch": 2.371609924985574,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004445938019701506,
      "loss": 5.8546,
      "mean_token_accuracy": 0.17890825271606445,
      "num_tokens": 52091692.0,
      "step": 20550
    },
    {
      "entropy": 6.2114182472229,
      "epoch": 2.3721869590305826,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004445666817951855,
      "loss": 5.7735,
      "mean_token_accuracy": 0.18435920178890228,
      "num_tokens": 52103010.0,
      "step": 20555
    },
    {
      "entropy": 6.084341430664063,
      "epoch": 2.3727639930755915,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004445395559169293,
      "loss": 5.7723,
      "mean_token_accuracy": 0.1898090735077858,
      "num_tokens": 52115185.0,
      "step": 20560
    },
    {
      "entropy": 6.11562762260437,
      "epoch": 2.3733410271206,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004445124243362943,
      "loss": 5.7834,
      "mean_token_accuracy": 0.1789586752653122,
      "num_tokens": 52127713.0,
      "step": 20565
    },
    {
      "entropy": 6.1114589214324955,
      "epoch": 2.3739180611656088,
      "grad_norm": 1.125,
      "learning_rate": 0.0004444852870541932,
      "loss": 5.7453,
      "mean_token_accuracy": 0.18846319615840912,
      "num_tokens": 52141127.0,
      "step": 20570
    },
    {
      "entropy": 6.067537879943847,
      "epoch": 2.3744950952106176,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004444581440715386,
      "loss": 5.7653,
      "mean_token_accuracy": 0.19737626761198043,
      "num_tokens": 52154063.0,
      "step": 20575
    },
    {
      "entropy": 6.148746585845947,
      "epoch": 2.375072129255626,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00044443099538924347,
      "loss": 5.7966,
      "mean_token_accuracy": 0.17939671874046326,
      "num_tokens": 52165605.0,
      "step": 20580
    },
    {
      "entropy": 6.181947708129883,
      "epoch": 2.375649163300635,
      "grad_norm": 1.125,
      "learning_rate": 0.0004444038410082211,
      "loss": 5.7823,
      "mean_token_accuracy": 0.18514414876699448,
      "num_tokens": 52178739.0,
      "step": 20585
    },
    {
      "entropy": 6.182062816619873,
      "epoch": 2.3762261973456433,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004443766809293846,
      "loss": 5.8961,
      "mean_token_accuracy": 0.1761386513710022,
      "num_tokens": 52192056.0,
      "step": 20590
    },
    {
      "entropy": 6.111511993408203,
      "epoch": 2.376803231390652,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004443495151536475,
      "loss": 5.7423,
      "mean_token_accuracy": 0.18427986055612564,
      "num_tokens": 52204791.0,
      "step": 20595
    },
    {
      "entropy": 6.1407757759094235,
      "epoch": 2.3773802654356606,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004443223436819237,
      "loss": 5.8031,
      "mean_token_accuracy": 0.18613592237234117,
      "num_tokens": 52216929.0,
      "step": 20600
    },
    {
      "entropy": 6.188626480102539,
      "epoch": 2.3779572994806695,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00044429516651512687,
      "loss": 5.9074,
      "mean_token_accuracy": 0.17905638366937637,
      "num_tokens": 52228946.0,
      "step": 20605
    },
    {
      "entropy": 6.1470903873443605,
      "epoch": 2.378534333525678,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044426798365417133,
      "loss": 5.7831,
      "mean_token_accuracy": 0.1831638351082802,
      "num_tokens": 52242114.0,
      "step": 20610
    },
    {
      "entropy": 6.121111917495727,
      "epoch": 2.3791113675706868,
      "grad_norm": 0.953125,
      "learning_rate": 0.00044424079509997104,
      "loss": 5.8322,
      "mean_token_accuracy": 0.17893511205911636,
      "num_tokens": 52255434.0,
      "step": 20615
    },
    {
      "entropy": 6.167761135101318,
      "epoch": 2.379688401615695,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004442136008534409,
      "loss": 5.8527,
      "mean_token_accuracy": 0.18079368621110917,
      "num_tokens": 52268774.0,
      "step": 20620
    },
    {
      "entropy": 6.141374254226685,
      "epoch": 2.380265435660704,
      "grad_norm": 0.953125,
      "learning_rate": 0.00044418640091549525,
      "loss": 5.8104,
      "mean_token_accuracy": 0.18754349201917647,
      "num_tokens": 52282046.0,
      "step": 20625
    },
    {
      "entropy": 6.104725790023804,
      "epoch": 2.3808424697057124,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000444159195287049,
      "loss": 5.7956,
      "mean_token_accuracy": 0.18250571340322494,
      "num_tokens": 52295266.0,
      "step": 20630
    },
    {
      "entropy": 6.105867338180542,
      "epoch": 2.3814195037507213,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004441319839690173,
      "loss": 5.7413,
      "mean_token_accuracy": 0.18568430840969086,
      "num_tokens": 52307530.0,
      "step": 20635
    },
    {
      "entropy": 6.08230881690979,
      "epoch": 2.3819965377957297,
      "grad_norm": 1.0,
      "learning_rate": 0.0004441047669623153,
      "loss": 5.7273,
      "mean_token_accuracy": 0.19005152136087416,
      "num_tokens": 52319461.0,
      "step": 20640
    },
    {
      "entropy": 6.080320930480957,
      "epoch": 2.3825735718407386,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044407754426785845,
      "loss": 5.755,
      "mean_token_accuracy": 0.18198447227478026,
      "num_tokens": 52332163.0,
      "step": 20645
    },
    {
      "entropy": 6.232722616195678,
      "epoch": 2.3831506058857475,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004440503158865625,
      "loss": 5.9384,
      "mean_token_accuracy": 0.17457229644060135,
      "num_tokens": 52344970.0,
      "step": 20650
    },
    {
      "entropy": 6.172081375122071,
      "epoch": 2.383727639930756,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044402308181934295,
      "loss": 5.8516,
      "mean_token_accuracy": 0.1812719702720642,
      "num_tokens": 52358378.0,
      "step": 20655
    },
    {
      "entropy": 6.110611867904663,
      "epoch": 2.3843046739757647,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004439958420671162,
      "loss": 5.784,
      "mean_token_accuracy": 0.17903131395578384,
      "num_tokens": 52369335.0,
      "step": 20660
    },
    {
      "entropy": 6.128928899765015,
      "epoch": 2.384881708020773,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044396859663079814,
      "loss": 5.7647,
      "mean_token_accuracy": 0.1880863532423973,
      "num_tokens": 52381573.0,
      "step": 20665
    },
    {
      "entropy": 6.118908166885376,
      "epoch": 2.385458742065782,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044394134551130533,
      "loss": 5.7642,
      "mean_token_accuracy": 0.1875980630517006,
      "num_tokens": 52395316.0,
      "step": 20670
    },
    {
      "entropy": 6.1461201190948485,
      "epoch": 2.3860357761107904,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004439140887095542,
      "loss": 5.7545,
      "mean_token_accuracy": 0.18303051143884658,
      "num_tokens": 52407751.0,
      "step": 20675
    },
    {
      "entropy": 6.041077280044556,
      "epoch": 2.3866128101557993,
      "grad_norm": 0.984375,
      "learning_rate": 0.00044388682622646165,
      "loss": 5.7977,
      "mean_token_accuracy": 0.18750965893268584,
      "num_tokens": 52420880.0,
      "step": 20680
    },
    {
      "entropy": 6.029811429977417,
      "epoch": 2.3871898442008077,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004438595580629446,
      "loss": 5.7916,
      "mean_token_accuracy": 0.1861064016819,
      "num_tokens": 52433600.0,
      "step": 20685
    },
    {
      "entropy": 6.102794981002807,
      "epoch": 2.3877668782458166,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004438322842199202,
      "loss": 5.7174,
      "mean_token_accuracy": 0.19216446429491044,
      "num_tokens": 52446983.0,
      "step": 20690
    },
    {
      "entropy": 6.172792100906372,
      "epoch": 2.388343912290825,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004438050046983057,
      "loss": 5.8329,
      "mean_token_accuracy": 0.17529330402612686,
      "num_tokens": 52460668.0,
      "step": 20695
    },
    {
      "entropy": 6.098419952392578,
      "epoch": 2.388920946335834,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044377771949901876,
      "loss": 5.773,
      "mean_token_accuracy": 0.18325713127851487,
      "num_tokens": 52472838.0,
      "step": 20700
    },
    {
      "entropy": 6.110059356689453,
      "epoch": 2.3894979803808427,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00044375042862297703,
      "loss": 5.8545,
      "mean_token_accuracy": 0.17736202478408813,
      "num_tokens": 52485412.0,
      "step": 20705
    },
    {
      "entropy": 6.131012916564941,
      "epoch": 2.390075014425851,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044372313207109856,
      "loss": 5.8227,
      "mean_token_accuracy": 0.18876413404941558,
      "num_tokens": 52497440.0,
      "step": 20710
    },
    {
      "entropy": 6.171064805984497,
      "epoch": 2.3906520484708595,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00044369582984430127,
      "loss": 5.8199,
      "mean_token_accuracy": 0.18323802053928376,
      "num_tokens": 52510549.0,
      "step": 20715
    },
    {
      "entropy": 6.074308156967163,
      "epoch": 2.3912290825158684,
      "grad_norm": 1.125,
      "learning_rate": 0.00044366852194350354,
      "loss": 5.768,
      "mean_token_accuracy": 0.1865235060453415,
      "num_tokens": 52522542.0,
      "step": 20720
    },
    {
      "entropy": 6.057948589324951,
      "epoch": 2.3918061165608773,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004436412083696239,
      "loss": 5.7135,
      "mean_token_accuracy": 0.1839848667383194,
      "num_tokens": 52535607.0,
      "step": 20725
    },
    {
      "entropy": 6.150481510162353,
      "epoch": 2.3923831506058857,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00044361388912358095,
      "loss": 5.7465,
      "mean_token_accuracy": 0.18799040615558624,
      "num_tokens": 52548515.0,
      "step": 20730
    },
    {
      "entropy": 6.160697746276855,
      "epoch": 2.3929601846508946,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004435865642062936,
      "loss": 5.8136,
      "mean_token_accuracy": 0.1764257550239563,
      "num_tokens": 52561142.0,
      "step": 20735
    },
    {
      "entropy": 6.11927056312561,
      "epoch": 2.393537218695903,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004435592336186809,
      "loss": 5.7798,
      "mean_token_accuracy": 0.18851767778396605,
      "num_tokens": 52573355.0,
      "step": 20740
    },
    {
      "entropy": 6.162110710144043,
      "epoch": 2.394114252740912,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004435318973616622,
      "loss": 5.8765,
      "mean_token_accuracy": 0.17309651374816895,
      "num_tokens": 52586353.0,
      "step": 20745
    },
    {
      "entropy": 6.146757125854492,
      "epoch": 2.3946912867859202,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044350455543615665,
      "loss": 5.8012,
      "mean_token_accuracy": 0.1842290937900543,
      "num_tokens": 52597673.0,
      "step": 20750
    },
    {
      "entropy": 6.051271677017212,
      "epoch": 2.395268320830929,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004434772078430843,
      "loss": 5.8232,
      "mean_token_accuracy": 0.18001710176467894,
      "num_tokens": 52609165.0,
      "step": 20755
    },
    {
      "entropy": 6.1048181533813475,
      "epoch": 2.3958453548759375,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004434498545833646,
      "loss": 5.8604,
      "mean_token_accuracy": 0.18053965866565705,
      "num_tokens": 52623335.0,
      "step": 20760
    },
    {
      "entropy": 6.138064908981323,
      "epoch": 2.3964223889209464,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004434224956579177,
      "loss": 5.7255,
      "mean_token_accuracy": 0.18789040595293044,
      "num_tokens": 52635754.0,
      "step": 20765
    },
    {
      "entropy": 6.087193965911865,
      "epoch": 2.396999422965955,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004433951310676639,
      "loss": 5.8261,
      "mean_token_accuracy": 0.18568842858076096,
      "num_tokens": 52649095.0,
      "step": 20770
    },
    {
      "entropy": 6.109959363937378,
      "epoch": 2.3975764570109637,
      "grad_norm": 1.125,
      "learning_rate": 0.00044336776081352347,
      "loss": 5.7292,
      "mean_token_accuracy": 0.19300491660833358,
      "num_tokens": 52661421.0,
      "step": 20775
    },
    {
      "entropy": 6.018739891052246,
      "epoch": 2.3981534910559725,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044334038489641706,
      "loss": 5.66,
      "mean_token_accuracy": 0.19040346294641494,
      "num_tokens": 52674865.0,
      "step": 20780
    },
    {
      "entropy": 6.078220701217651,
      "epoch": 2.398730525100981,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044331300331726544,
      "loss": 5.7986,
      "mean_token_accuracy": 0.18137383908033372,
      "num_tokens": 52687241.0,
      "step": 20785
    },
    {
      "entropy": 6.131845951080322,
      "epoch": 2.3993075591459894,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044328561607698947,
      "loss": 5.8114,
      "mean_token_accuracy": 0.18047611266374589,
      "num_tokens": 52699567.0,
      "step": 20790
    },
    {
      "entropy": 6.079652309417725,
      "epoch": 2.3998845931909982,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004432582231765105,
      "loss": 5.7512,
      "mean_token_accuracy": 0.18098798990249634,
      "num_tokens": 52711607.0,
      "step": 20795
    },
    {
      "entropy": 6.076911735534668,
      "epoch": 2.400461627236007,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044323082461674974,
      "loss": 5.692,
      "mean_token_accuracy": 0.19030100256204605,
      "num_tokens": 52724256.0,
      "step": 20800
    },
    {
      "entropy": 6.0936439037323,
      "epoch": 2.4010386612810155,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004432034203986287,
      "loss": 5.7538,
      "mean_token_accuracy": 0.1939753621816635,
      "num_tokens": 52736356.0,
      "step": 20805
    },
    {
      "entropy": 6.0900688648223875,
      "epoch": 2.4016156953260244,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004431760105230693,
      "loss": 5.7501,
      "mean_token_accuracy": 0.18549337834119797,
      "num_tokens": 52748623.0,
      "step": 20810
    },
    {
      "entropy": 6.161285924911499,
      "epoch": 2.402192729371033,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00044314859499099325,
      "loss": 5.8062,
      "mean_token_accuracy": 0.17826490849256516,
      "num_tokens": 52761479.0,
      "step": 20815
    },
    {
      "entropy": 6.116491031646729,
      "epoch": 2.4027697634160416,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044312117380332274,
      "loss": 5.7716,
      "mean_token_accuracy": 0.1940062701702118,
      "num_tokens": 52774681.0,
      "step": 20820
    },
    {
      "entropy": 6.152537298202515,
      "epoch": 2.40334679746105,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044309374696098,
      "loss": 5.8191,
      "mean_token_accuracy": 0.18741075098514556,
      "num_tokens": 52787495.0,
      "step": 20825
    },
    {
      "entropy": 6.099158906936646,
      "epoch": 2.403923831506059,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004430663144648876,
      "loss": 5.8407,
      "mean_token_accuracy": 0.18023990243673324,
      "num_tokens": 52799958.0,
      "step": 20830
    },
    {
      "entropy": 6.157201623916626,
      "epoch": 2.4045008655510673,
      "grad_norm": 1.0,
      "learning_rate": 0.00044303887631596823,
      "loss": 5.7917,
      "mean_token_accuracy": 0.18298872858285903,
      "num_tokens": 52812033.0,
      "step": 20835
    },
    {
      "entropy": 6.152058219909668,
      "epoch": 2.405077899596076,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004430114325151447,
      "loss": 5.8124,
      "mean_token_accuracy": 0.17951597273349762,
      "num_tokens": 52824269.0,
      "step": 20840
    },
    {
      "entropy": 6.1759552478790285,
      "epoch": 2.4056549336410846,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004429839830633401,
      "loss": 5.8404,
      "mean_token_accuracy": 0.18397446274757384,
      "num_tokens": 52837180.0,
      "step": 20845
    },
    {
      "entropy": 6.146060562133789,
      "epoch": 2.4062319676860935,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004429565279614777,
      "loss": 5.8285,
      "mean_token_accuracy": 0.17569271475076675,
      "num_tokens": 52849751.0,
      "step": 20850
    },
    {
      "entropy": 6.100464105606079,
      "epoch": 2.4068090017311023,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044292906721048094,
      "loss": 5.7365,
      "mean_token_accuracy": 0.18632608652114868,
      "num_tokens": 52862902.0,
      "step": 20855
    },
    {
      "entropy": 6.073440599441528,
      "epoch": 2.4073860357761108,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004429016008112733,
      "loss": 5.7584,
      "mean_token_accuracy": 0.18374385237693786,
      "num_tokens": 52876104.0,
      "step": 20860
    },
    {
      "entropy": 6.14205756187439,
      "epoch": 2.4079630698211196,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004428741287647788,
      "loss": 5.8561,
      "mean_token_accuracy": 0.1776397243142128,
      "num_tokens": 52889173.0,
      "step": 20865
    },
    {
      "entropy": 6.079821062088013,
      "epoch": 2.408540103866128,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00044284665107192136,
      "loss": 5.6975,
      "mean_token_accuracy": 0.19375640898942947,
      "num_tokens": 52902064.0,
      "step": 20870
    },
    {
      "entropy": 6.1201738834381105,
      "epoch": 2.409117137911137,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004428191677336251,
      "loss": 5.7941,
      "mean_token_accuracy": 0.17660316228866577,
      "num_tokens": 52913895.0,
      "step": 20875
    },
    {
      "entropy": 6.150294399261474,
      "epoch": 2.4096941719561453,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004427916787508146,
      "loss": 5.7439,
      "mean_token_accuracy": 0.18648834973573686,
      "num_tokens": 52925589.0,
      "step": 20880
    },
    {
      "entropy": 6.118099927902222,
      "epoch": 2.410271206001154,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004427641841244144,
      "loss": 5.7772,
      "mean_token_accuracy": 0.19391288310289384,
      "num_tokens": 52940519.0,
      "step": 20885
    },
    {
      "entropy": 6.026507997512818,
      "epoch": 2.4108482400461626,
      "grad_norm": 1.03125,
      "learning_rate": 0.000442736683855349,
      "loss": 5.7764,
      "mean_token_accuracy": 0.18228887170553207,
      "num_tokens": 52954205.0,
      "step": 20890
    },
    {
      "entropy": 6.095542001724243,
      "epoch": 2.4114252740911715,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004427091779445437,
      "loss": 5.8029,
      "mean_token_accuracy": 0.18271701633930207,
      "num_tokens": 52967949.0,
      "step": 20895
    },
    {
      "entropy": 6.067535305023194,
      "epoch": 2.41200230813618,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004426816663929235,
      "loss": 5.6918,
      "mean_token_accuracy": 0.19003389477729798,
      "num_tokens": 52980125.0,
      "step": 20900
    },
    {
      "entropy": 5.998403787612915,
      "epoch": 2.4125793421811887,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044265414920141366,
      "loss": 5.6835,
      "mean_token_accuracy": 0.1972445949912071,
      "num_tokens": 52992721.0,
      "step": 20905
    },
    {
      "entropy": 6.070257663726807,
      "epoch": 2.413156376226197,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044262662637093987,
      "loss": 5.8095,
      "mean_token_accuracy": 0.18321685045957564,
      "num_tokens": 53004278.0,
      "step": 20910
    },
    {
      "entropy": 6.1697649478912355,
      "epoch": 2.413733410271206,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044259909790242776,
      "loss": 5.7736,
      "mean_token_accuracy": 0.18285899758338928,
      "num_tokens": 53018290.0,
      "step": 20915
    },
    {
      "entropy": 6.037852764129639,
      "epoch": 2.4143104443162144,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004425715637968032,
      "loss": 5.6901,
      "mean_token_accuracy": 0.1975919172167778,
      "num_tokens": 53031458.0,
      "step": 20920
    },
    {
      "entropy": 6.0538571834564205,
      "epoch": 2.4148874783612233,
      "grad_norm": 1.0,
      "learning_rate": 0.0004425440240549923,
      "loss": 5.7461,
      "mean_token_accuracy": 0.18474527150392533,
      "num_tokens": 53042894.0,
      "step": 20925
    },
    {
      "entropy": 6.194273948669434,
      "epoch": 2.415464512406232,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044251647867792147,
      "loss": 5.898,
      "mean_token_accuracy": 0.18324737697839738,
      "num_tokens": 53056206.0,
      "step": 20930
    },
    {
      "entropy": 6.0965221405029295,
      "epoch": 2.4160415464512406,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044248892766651706,
      "loss": 5.7618,
      "mean_token_accuracy": 0.1864438995718956,
      "num_tokens": 53069249.0,
      "step": 20935
    },
    {
      "entropy": 6.086435222625733,
      "epoch": 2.4166185804962494,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004424613710217057,
      "loss": 5.8277,
      "mean_token_accuracy": 0.17954965084791183,
      "num_tokens": 53080716.0,
      "step": 20940
    },
    {
      "entropy": 6.116638708114624,
      "epoch": 2.417195614541258,
      "grad_norm": 0.96875,
      "learning_rate": 0.00044243380874441437,
      "loss": 5.8097,
      "mean_token_accuracy": 0.18204215466976165,
      "num_tokens": 53093671.0,
      "step": 20945
    },
    {
      "entropy": 6.161101818084717,
      "epoch": 2.4177726485862667,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044240624083557,
      "loss": 5.8142,
      "mean_token_accuracy": 0.1833065554499626,
      "num_tokens": 53106057.0,
      "step": 20950
    },
    {
      "entropy": 6.09749174118042,
      "epoch": 2.418349682631275,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00044237866729609994,
      "loss": 5.7608,
      "mean_token_accuracy": 0.1861227720975876,
      "num_tokens": 53118902.0,
      "step": 20955
    },
    {
      "entropy": 6.077785444259644,
      "epoch": 2.418926716676284,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004423510881269315,
      "loss": 5.8686,
      "mean_token_accuracy": 0.17929997742176057,
      "num_tokens": 53133084.0,
      "step": 20960
    },
    {
      "entropy": 6.142805719375611,
      "epoch": 2.4195037507212924,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004423235033289924,
      "loss": 5.6952,
      "mean_token_accuracy": 0.1858563795685768,
      "num_tokens": 53144877.0,
      "step": 20965
    },
    {
      "entropy": 6.096131420135498,
      "epoch": 2.4200807847663013,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004422959129032103,
      "loss": 5.7391,
      "mean_token_accuracy": 0.18631935119628906,
      "num_tokens": 53158328.0,
      "step": 20970
    },
    {
      "entropy": 6.069164276123047,
      "epoch": 2.4206578188113097,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044226831685051333,
      "loss": 5.7927,
      "mean_token_accuracy": 0.17545884549617768,
      "num_tokens": 53169769.0,
      "step": 20975
    },
    {
      "entropy": 6.176840019226074,
      "epoch": 2.4212348528563186,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004422407151718296,
      "loss": 5.7886,
      "mean_token_accuracy": 0.18747422248125076,
      "num_tokens": 53182686.0,
      "step": 20980
    },
    {
      "entropy": 6.1123366355896,
      "epoch": 2.4218118869013274,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044221310786808746,
      "loss": 5.8012,
      "mean_token_accuracy": 0.18763403296470643,
      "num_tokens": 53195720.0,
      "step": 20985
    },
    {
      "entropy": 6.010849523544311,
      "epoch": 2.422388920946336,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004421854949402155,
      "loss": 5.6811,
      "mean_token_accuracy": 0.19400209784507752,
      "num_tokens": 53209068.0,
      "step": 20990
    },
    {
      "entropy": 6.059459400177002,
      "epoch": 2.4229659549913443,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044215787638914245,
      "loss": 5.7444,
      "mean_token_accuracy": 0.18343445807695388,
      "num_tokens": 53220739.0,
      "step": 20995
    },
    {
      "entropy": 6.068580913543701,
      "epoch": 2.423542989036353,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004421302522157973,
      "loss": 5.8025,
      "mean_token_accuracy": 0.17986188530921937,
      "num_tokens": 53232633.0,
      "step": 21000
    },
    {
      "epoch": 2.423542989036353,
      "eval_entropy": 5.989283062769738,
      "eval_loss": 5.924990653991699,
      "eval_mean_token_accuracy": 0.1836731506022737,
      "eval_num_tokens": 53232633.0,
      "eval_runtime": 23.264,
      "eval_samples_per_second": 1209.42,
      "eval_steps_per_second": 151.178,
      "step": 21000
    }
  ],
  "logging_steps": 5,
  "max_steps": 86650,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 3000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 7.8581480122368e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}