{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0811249323958896,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 4.802441883087158,
      "epoch": 0.005408328826392645,
      "grad_norm": 13.375,
      "learning_rate": 2e-06,
      "loss": 14.4628,
      "mean_token_accuracy": 0.0,
      "num_tokens": 10942.0,
      "step": 5
    },
    {
      "entropy": 4.81857442855835,
      "epoch": 0.01081665765278529,
      "grad_norm": 16.25,
      "learning_rate": 4.5e-06,
      "loss": 14.3862,
      "mean_token_accuracy": 0.0,
      "num_tokens": 21887.0,
      "step": 10
    },
    {
      "entropy": 4.852184391021728,
      "epoch": 0.016224986479177934,
      "grad_norm": 18.375,
      "learning_rate": 7e-06,
      "loss": 14.2078,
      "mean_token_accuracy": 0.0,
      "num_tokens": 34345.0,
      "step": 15
    },
    {
      "entropy": 5.0442163944244385,
      "epoch": 0.02163331530557058,
      "grad_norm": 27.0,
      "learning_rate": 9.5e-06,
      "loss": 13.6012,
      "mean_token_accuracy": 0.0,
      "num_tokens": 45445.0,
      "step": 20
    },
    {
      "entropy": 6.944517469406128,
      "epoch": 0.02704164413196322,
      "grad_norm": 19.5,
      "learning_rate": 1.2e-05,
      "loss": 12.0589,
      "mean_token_accuracy": 0.0,
      "num_tokens": 58237.0,
      "step": 25
    },
    {
      "entropy": 10.193405723571777,
      "epoch": 0.03244997295835587,
      "grad_norm": 3.921875,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 10.8364,
      "mean_token_accuracy": 0.002917378023266792,
      "num_tokens": 68098.0,
      "step": 30
    },
    {
      "entropy": 10.690443992614746,
      "epoch": 0.03785830178474851,
      "grad_norm": 3.234375,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.5773,
      "mean_token_accuracy": 0.014713712595403194,
      "num_tokens": 79867.0,
      "step": 35
    },
    {
      "entropy": 10.687717247009278,
      "epoch": 0.04326663061114116,
      "grad_norm": 2.90625,
      "learning_rate": 1.95e-05,
      "loss": 10.2534,
      "mean_token_accuracy": 0.016428307350724935,
      "num_tokens": 91257.0,
      "step": 40
    },
    {
      "entropy": 10.658015537261964,
      "epoch": 0.048674959437533805,
      "grad_norm": 2.515625,
      "learning_rate": 2.2e-05,
      "loss": 9.9866,
      "mean_token_accuracy": 0.024076672736555338,
      "num_tokens": 103474.0,
      "step": 45
    },
    {
      "entropy": 10.61853494644165,
      "epoch": 0.05408328826392644,
      "grad_norm": 2.53125,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.7234,
      "mean_token_accuracy": 0.03784517608582973,
      "num_tokens": 114830.0,
      "step": 50
    },
    {
      "entropy": 10.597916889190675,
      "epoch": 0.05949161709031909,
      "grad_norm": 2.171875,
      "learning_rate": 2.7e-05,
      "loss": 9.5693,
      "mean_token_accuracy": 0.035866692662239075,
      "num_tokens": 126017.0,
      "step": 55
    },
    {
      "entropy": 10.58803596496582,
      "epoch": 0.06489994591671173,
      "grad_norm": 2.171875,
      "learning_rate": 2.95e-05,
      "loss": 9.4863,
      "mean_token_accuracy": 0.040941419266164306,
      "num_tokens": 138509.0,
      "step": 60
    },
    {
      "entropy": 10.57759370803833,
      "epoch": 0.07030827474310439,
      "grad_norm": 2.046875,
      "learning_rate": 3.2e-05,
      "loss": 9.4155,
      "mean_token_accuracy": 0.040815659798681736,
      "num_tokens": 148353.0,
      "step": 65
    },
    {
      "entropy": 10.564495277404784,
      "epoch": 0.07571660356949703,
      "grad_norm": 2.0,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.3856,
      "mean_token_accuracy": 0.03659801315516233,
      "num_tokens": 160269.0,
      "step": 70
    },
    {
      "entropy": 10.558395195007325,
      "epoch": 0.08112493239588967,
      "grad_norm": 2.140625,
      "learning_rate": 3.7e-05,
      "loss": 9.2438,
      "mean_token_accuracy": 0.036995376273989675,
      "num_tokens": 171568.0,
      "step": 75
    },
    {
      "entropy": 10.547275161743164,
      "epoch": 0.08653326122228232,
      "grad_norm": 2.140625,
      "learning_rate": 3.95e-05,
      "loss": 9.1614,
      "mean_token_accuracy": 0.0399360004812479,
      "num_tokens": 184243.0,
      "step": 80
    },
    {
      "entropy": 10.52041311264038,
      "epoch": 0.09194159004867496,
      "grad_norm": 2.09375,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.0786,
      "mean_token_accuracy": 0.039298290386796,
      "num_tokens": 194887.0,
      "step": 85
    },
    {
      "entropy": 10.486145210266113,
      "epoch": 0.09734991887506761,
      "grad_norm": 2.0,
      "learning_rate": 4.45e-05,
      "loss": 8.9895,
      "mean_token_accuracy": 0.04011793658137321,
      "num_tokens": 206476.0,
      "step": 90
    },
    {
      "entropy": 10.448410129547119,
      "epoch": 0.10275824770146025,
      "grad_norm": 1.9609375,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.8381,
      "mean_token_accuracy": 0.04185208044946194,
      "num_tokens": 216578.0,
      "step": 95
    },
    {
      "entropy": 10.391289806365966,
      "epoch": 0.10816657652785289,
      "grad_norm": 1.8515625,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.7399,
      "mean_token_accuracy": 0.0479723758995533,
      "num_tokens": 226941.0,
      "step": 100
    },
    {
      "entropy": 10.326226139068604,
      "epoch": 0.11357490535424554,
      "grad_norm": 1.7890625,
      "learning_rate": 5.2e-05,
      "loss": 8.6146,
      "mean_token_accuracy": 0.05267183780670166,
      "num_tokens": 238060.0,
      "step": 105
    },
    {
      "entropy": 10.213607120513917,
      "epoch": 0.11898323418063818,
      "grad_norm": 1.9140625,
      "learning_rate": 5.45e-05,
      "loss": 8.4421,
      "mean_token_accuracy": 0.055974167212843894,
      "num_tokens": 249137.0,
      "step": 110
    },
    {
      "entropy": 10.048615741729737,
      "epoch": 0.12439156300703083,
      "grad_norm": 1.6484375,
      "learning_rate": 5.7e-05,
      "loss": 8.3336,
      "mean_token_accuracy": 0.05477573201060295,
      "num_tokens": 260712.0,
      "step": 115
    },
    {
      "entropy": 9.906066799163819,
      "epoch": 0.12979989183342347,
      "grad_norm": 1.59375,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.1638,
      "mean_token_accuracy": 0.05388197749853134,
      "num_tokens": 271039.0,
      "step": 120
    },
    {
      "entropy": 9.787034893035889,
      "epoch": 0.1352082206598161,
      "grad_norm": 1.6015625,
      "learning_rate": 6.2e-05,
      "loss": 8.0076,
      "mean_token_accuracy": 0.05202497914433479,
      "num_tokens": 281353.0,
      "step": 125
    },
    {
      "entropy": 9.50912857055664,
      "epoch": 0.14061654948620878,
      "grad_norm": 1.140625,
      "learning_rate": 6.450000000000001e-05,
      "loss": 7.9456,
      "mean_token_accuracy": 0.054855041578412055,
      "num_tokens": 292221.0,
      "step": 130
    },
    {
      "entropy": 9.263109493255616,
      "epoch": 0.1460248783126014,
      "grad_norm": 1.109375,
      "learning_rate": 6.7e-05,
      "loss": 7.8241,
      "mean_token_accuracy": 0.0536174725741148,
      "num_tokens": 303353.0,
      "step": 135
    },
    {
      "entropy": 8.991836738586425,
      "epoch": 0.15143320713899405,
      "grad_norm": 1.03125,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.7511,
      "mean_token_accuracy": 0.056266501918435095,
      "num_tokens": 314165.0,
      "step": 140
    },
    {
      "entropy": 8.685503292083741,
      "epoch": 0.1568415359653867,
      "grad_norm": 0.93359375,
      "learning_rate": 7.2e-05,
      "loss": 7.5805,
      "mean_token_accuracy": 0.05333330743014812,
      "num_tokens": 326387.0,
      "step": 145
    },
    {
      "entropy": 8.49657220840454,
      "epoch": 0.16224986479177933,
      "grad_norm": 0.91015625,
      "learning_rate": 7.45e-05,
      "loss": 7.5908,
      "mean_token_accuracy": 0.05400457009673119,
      "num_tokens": 339155.0,
      "step": 150
    },
    {
      "entropy": 8.256601428985595,
      "epoch": 0.167658193618172,
      "grad_norm": 1.1796875,
      "learning_rate": 7.7e-05,
      "loss": 7.5015,
      "mean_token_accuracy": 0.05675696507096291,
      "num_tokens": 350643.0,
      "step": 155
    },
    {
      "entropy": 8.110760021209718,
      "epoch": 0.17306652244456464,
      "grad_norm": 1.2421875,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.4385,
      "mean_token_accuracy": 0.05687093436717987,
      "num_tokens": 361977.0,
      "step": 160
    },
    {
      "entropy": 7.999344825744629,
      "epoch": 0.17847485127095727,
      "grad_norm": 0.96484375,
      "learning_rate": 8.2e-05,
      "loss": 7.4928,
      "mean_token_accuracy": 0.06705345250666142,
      "num_tokens": 373604.0,
      "step": 165
    },
    {
      "entropy": 7.95798511505127,
      "epoch": 0.1838831800973499,
      "grad_norm": 0.90234375,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.4608,
      "mean_token_accuracy": 0.07175817862153053,
      "num_tokens": 384306.0,
      "step": 170
    },
    {
      "entropy": 7.838786649703979,
      "epoch": 0.18929150892374255,
      "grad_norm": 1.234375,
      "learning_rate": 8.7e-05,
      "loss": 7.3682,
      "mean_token_accuracy": 0.07707577534019946,
      "num_tokens": 396479.0,
      "step": 175
    },
    {
      "entropy": 7.802298307418823,
      "epoch": 0.19469983775013522,
      "grad_norm": 0.9453125,
      "learning_rate": 8.95e-05,
      "loss": 7.3187,
      "mean_token_accuracy": 0.07556967809796333,
      "num_tokens": 408489.0,
      "step": 180
    },
    {
      "entropy": 7.777406930923462,
      "epoch": 0.20010816657652786,
      "grad_norm": 1.109375,
      "learning_rate": 9.2e-05,
      "loss": 7.3304,
      "mean_token_accuracy": 0.08097317069768906,
      "num_tokens": 419339.0,
      "step": 185
    },
    {
      "entropy": 7.777624797821045,
      "epoch": 0.2055164954029205,
      "grad_norm": 1.0234375,
      "learning_rate": 9.45e-05,
      "loss": 7.2935,
      "mean_token_accuracy": 0.08264814168214799,
      "num_tokens": 430486.0,
      "step": 190
    },
    {
      "entropy": 7.753584861755371,
      "epoch": 0.21092482422931313,
      "grad_norm": 1.0859375,
      "learning_rate": 9.7e-05,
      "loss": 7.2943,
      "mean_token_accuracy": 0.08632798045873642,
      "num_tokens": 440414.0,
      "step": 195
    },
    {
      "entropy": 7.748952054977417,
      "epoch": 0.21633315305570577,
      "grad_norm": 1.046875,
      "learning_rate": 9.95e-05,
      "loss": 7.2391,
      "mean_token_accuracy": 0.08323732763528824,
      "num_tokens": 451475.0,
      "step": 200
    },
    {
      "entropy": 7.659665584564209,
      "epoch": 0.22174148188209844,
      "grad_norm": 0.90625,
      "learning_rate": 0.000102,
      "loss": 7.3028,
      "mean_token_accuracy": 0.08789955824613571,
      "num_tokens": 462053.0,
      "step": 205
    },
    {
      "entropy": 7.667898845672608,
      "epoch": 0.22714981070849108,
      "grad_norm": 1.015625,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.2495,
      "mean_token_accuracy": 0.08465297371149064,
      "num_tokens": 473207.0,
      "step": 210
    },
    {
      "entropy": 7.687205266952515,
      "epoch": 0.23255813953488372,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000107,
      "loss": 7.2412,
      "mean_token_accuracy": 0.08715058863162994,
      "num_tokens": 484999.0,
      "step": 215
    },
    {
      "entropy": 7.709909105300904,
      "epoch": 0.23796646836127636,
      "grad_norm": 1.015625,
      "learning_rate": 0.0001095,
      "loss": 7.2455,
      "mean_token_accuracy": 0.08413905426859855,
      "num_tokens": 496772.0,
      "step": 220
    },
    {
      "entropy": 7.729339981079102,
      "epoch": 0.24337479718766902,
      "grad_norm": 1.140625,
      "learning_rate": 0.000112,
      "loss": 7.2107,
      "mean_token_accuracy": 0.0819280207157135,
      "num_tokens": 507782.0,
      "step": 225
    },
    {
      "entropy": 7.6187958240509035,
      "epoch": 0.24878312601406166,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001145,
      "loss": 7.113,
      "mean_token_accuracy": 0.08953967168927193,
      "num_tokens": 518950.0,
      "step": 230
    },
    {
      "entropy": 7.6271685600280765,
      "epoch": 0.2541914548404543,
      "grad_norm": 1.046875,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.1849,
      "mean_token_accuracy": 0.09068010970950127,
      "num_tokens": 530795.0,
      "step": 235
    },
    {
      "entropy": 7.659736394882202,
      "epoch": 0.25959978366684694,
      "grad_norm": 1.15625,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.1958,
      "mean_token_accuracy": 0.08313242197036744,
      "num_tokens": 542878.0,
      "step": 240
    },
    {
      "entropy": 7.627575349807739,
      "epoch": 0.2650081124932396,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000122,
      "loss": 7.0933,
      "mean_token_accuracy": 0.08872637823224068,
      "num_tokens": 553682.0,
      "step": 245
    },
    {
      "entropy": 7.5292261123657225,
      "epoch": 0.2704164413196322,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001245,
      "loss": 7.0941,
      "mean_token_accuracy": 0.0952358789741993,
      "num_tokens": 564433.0,
      "step": 250
    },
    {
      "entropy": 7.531745386123657,
      "epoch": 0.27582477014602486,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000127,
      "loss": 7.1263,
      "mean_token_accuracy": 0.09562606438994407,
      "num_tokens": 576001.0,
      "step": 255
    },
    {
      "entropy": 7.514750719070435,
      "epoch": 0.28123309897241755,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0001295,
      "loss": 7.0509,
      "mean_token_accuracy": 0.09575334340333938,
      "num_tokens": 587138.0,
      "step": 260
    },
    {
      "entropy": 7.472469186782837,
      "epoch": 0.2866414277988102,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000132,
      "loss": 7.0111,
      "mean_token_accuracy": 0.097767373919487,
      "num_tokens": 597572.0,
      "step": 265
    },
    {
      "entropy": 7.394614315032959,
      "epoch": 0.2920497566252028,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013450000000000002,
      "loss": 6.9821,
      "mean_token_accuracy": 0.10268613770604133,
      "num_tokens": 608879.0,
      "step": 270
    },
    {
      "entropy": 7.507109117507935,
      "epoch": 0.29745808545159547,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.0138,
      "mean_token_accuracy": 0.0923161268234253,
      "num_tokens": 620439.0,
      "step": 275
    },
    {
      "entropy": 7.481431484222412,
      "epoch": 0.3028664142779881,
      "grad_norm": 1.015625,
      "learning_rate": 0.0001395,
      "loss": 7.0666,
      "mean_token_accuracy": 0.0980131283402443,
      "num_tokens": 632543.0,
      "step": 280
    },
    {
      "entropy": 7.478473043441772,
      "epoch": 0.30827474310438074,
      "grad_norm": 1.140625,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.1097,
      "mean_token_accuracy": 0.10091490522027016,
      "num_tokens": 643369.0,
      "step": 285
    },
    {
      "entropy": 7.441432142257691,
      "epoch": 0.3136830719307734,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001445,
      "loss": 6.9728,
      "mean_token_accuracy": 0.1002442181110382,
      "num_tokens": 655328.0,
      "step": 290
    },
    {
      "entropy": 7.446781730651855,
      "epoch": 0.319091400757166,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000147,
      "loss": 7.0188,
      "mean_token_accuracy": 0.09976664856076241,
      "num_tokens": 665780.0,
      "step": 295
    },
    {
      "entropy": 7.409105348587036,
      "epoch": 0.32449972958355866,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001495,
      "loss": 6.9372,
      "mean_token_accuracy": 0.10388917848467827,
      "num_tokens": 676566.0,
      "step": 300
    },
    {
      "entropy": 7.41029577255249,
      "epoch": 0.3299080584099513,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000152,
      "loss": 6.9241,
      "mean_token_accuracy": 0.09972899705171585,
      "num_tokens": 687407.0,
      "step": 305
    },
    {
      "entropy": 7.285754060745239,
      "epoch": 0.335316387236344,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015450000000000001,
      "loss": 6.8611,
      "mean_token_accuracy": 0.10674165338277816,
      "num_tokens": 698162.0,
      "step": 310
    },
    {
      "entropy": 7.2760748863220215,
      "epoch": 0.34072471606273663,
      "grad_norm": 1.265625,
      "learning_rate": 0.000157,
      "loss": 6.8426,
      "mean_token_accuracy": 0.10251836776733399,
      "num_tokens": 708743.0,
      "step": 315
    },
    {
      "entropy": 7.402875375747681,
      "epoch": 0.34613304488912927,
      "grad_norm": 1.03125,
      "learning_rate": 0.0001595,
      "loss": 7.0535,
      "mean_token_accuracy": 0.10290157496929168,
      "num_tokens": 719828.0,
      "step": 320
    },
    {
      "entropy": 7.430617713928223,
      "epoch": 0.3515413737155219,
      "grad_norm": 1.25,
      "learning_rate": 0.000162,
      "loss": 7.0055,
      "mean_token_accuracy": 0.10132194757461548,
      "num_tokens": 732514.0,
      "step": 325
    },
    {
      "entropy": 7.293907737731933,
      "epoch": 0.35694970254191455,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.8334,
      "mean_token_accuracy": 0.10402422174811363,
      "num_tokens": 742391.0,
      "step": 330
    },
    {
      "entropy": 7.197569942474365,
      "epoch": 0.3623580313683072,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.834,
      "mean_token_accuracy": 0.10959370359778405,
      "num_tokens": 753638.0,
      "step": 335
    },
    {
      "entropy": 7.37444748878479,
      "epoch": 0.3677663601946998,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.0511,
      "mean_token_accuracy": 0.09830658435821533,
      "num_tokens": 764850.0,
      "step": 340
    },
    {
      "entropy": 7.327763652801513,
      "epoch": 0.37317468902109246,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.8284,
      "mean_token_accuracy": 0.10949721187353134,
      "num_tokens": 776192.0,
      "step": 345
    },
    {
      "entropy": 7.142797756195068,
      "epoch": 0.3785830178474851,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.7698,
      "mean_token_accuracy": 0.1114330142736435,
      "num_tokens": 787051.0,
      "step": 350
    },
    {
      "entropy": 7.214595127105713,
      "epoch": 0.3839913466738778,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000177,
      "loss": 6.81,
      "mean_token_accuracy": 0.10584235936403275,
      "num_tokens": 799281.0,
      "step": 355
    },
    {
      "entropy": 7.2480724334716795,
      "epoch": 0.38939967550027044,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001795,
      "loss": 6.8244,
      "mean_token_accuracy": 0.10464156717061997,
      "num_tokens": 810003.0,
      "step": 360
    },
    {
      "entropy": 7.165778160095215,
      "epoch": 0.3948080043266631,
      "grad_norm": 1.6015625,
      "learning_rate": 0.000182,
      "loss": 6.7971,
      "mean_token_accuracy": 0.10737397745251656,
      "num_tokens": 819749.0,
      "step": 365
    },
    {
      "entropy": 7.157060241699218,
      "epoch": 0.4002163331530557,
      "grad_norm": 1.109375,
      "learning_rate": 0.0001845,
      "loss": 6.798,
      "mean_token_accuracy": 0.11348775699734688,
      "num_tokens": 830552.0,
      "step": 370
    },
    {
      "entropy": 7.2775880813598635,
      "epoch": 0.40562466197944835,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000187,
      "loss": 6.8943,
      "mean_token_accuracy": 0.11229263693094253,
      "num_tokens": 840495.0,
      "step": 375
    },
    {
      "entropy": 7.101148653030395,
      "epoch": 0.411032990805841,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001895,
      "loss": 6.8311,
      "mean_token_accuracy": 0.11005991026759147,
      "num_tokens": 852915.0,
      "step": 380
    },
    {
      "entropy": 7.19731650352478,
      "epoch": 0.41644131963223363,
      "grad_norm": 1.21875,
      "learning_rate": 0.000192,
      "loss": 6.7841,
      "mean_token_accuracy": 0.11650079712271691,
      "num_tokens": 862762.0,
      "step": 385
    },
    {
      "entropy": 7.14437313079834,
      "epoch": 0.42184964845862627,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001945,
      "loss": 6.7342,
      "mean_token_accuracy": 0.11399752795696258,
      "num_tokens": 873023.0,
      "step": 390
    },
    {
      "entropy": 7.184113311767578,
      "epoch": 0.4272579772850189,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.8456,
      "mean_token_accuracy": 0.10619384720921517,
      "num_tokens": 885281.0,
      "step": 395
    },
    {
      "entropy": 7.0833296298980715,
      "epoch": 0.43266630611141155,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.8406,
      "mean_token_accuracy": 0.10883399248123168,
      "num_tokens": 896210.0,
      "step": 400
    },
    {
      "entropy": 7.069676637649536,
      "epoch": 0.43807463493780424,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000202,
      "loss": 6.7519,
      "mean_token_accuracy": 0.11440604701638221,
      "num_tokens": 909028.0,
      "step": 405
    },
    {
      "entropy": 7.126840543746948,
      "epoch": 0.4434829637641969,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.6773,
      "mean_token_accuracy": 0.118293134868145,
      "num_tokens": 919710.0,
      "step": 410
    },
    {
      "entropy": 7.0679234027862545,
      "epoch": 0.4488912925905895,
      "grad_norm": 1.03125,
      "learning_rate": 0.000207,
      "loss": 6.7387,
      "mean_token_accuracy": 0.1108441911637783,
      "num_tokens": 930019.0,
      "step": 415
    },
    {
      "entropy": 7.100546026229859,
      "epoch": 0.45429962141698216,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002095,
      "loss": 6.6661,
      "mean_token_accuracy": 0.11544709652662277,
      "num_tokens": 940535.0,
      "step": 420
    },
    {
      "entropy": 6.917388868331909,
      "epoch": 0.4597079502433748,
      "grad_norm": 1.15625,
      "learning_rate": 0.000212,
      "loss": 6.6207,
      "mean_token_accuracy": 0.11555901542305946,
      "num_tokens": 951390.0,
      "step": 425
    },
    {
      "entropy": 7.044145488739014,
      "epoch": 0.46511627906976744,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002145,
      "loss": 6.6635,
      "mean_token_accuracy": 0.11623054891824722,
      "num_tokens": 963010.0,
      "step": 430
    },
    {
      "entropy": 7.005122280120849,
      "epoch": 0.4705246078961601,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.629,
      "mean_token_accuracy": 0.12106695622205735,
      "num_tokens": 974032.0,
      "step": 435
    },
    {
      "entropy": 7.042861032485962,
      "epoch": 0.4759329367225527,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002195,
      "loss": 6.6675,
      "mean_token_accuracy": 0.1184677854180336,
      "num_tokens": 983954.0,
      "step": 440
    },
    {
      "entropy": 7.009529399871826,
      "epoch": 0.48134126554894535,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000222,
      "loss": 6.7104,
      "mean_token_accuracy": 0.11903695166110992,
      "num_tokens": 994135.0,
      "step": 445
    },
    {
      "entropy": 6.992412567138672,
      "epoch": 0.48674959437533805,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002245,
      "loss": 6.6665,
      "mean_token_accuracy": 0.11242141798138619,
      "num_tokens": 1006275.0,
      "step": 450
    },
    {
      "entropy": 6.83262996673584,
      "epoch": 0.4921579232017307,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.482,
      "mean_token_accuracy": 0.12549787536263465,
      "num_tokens": 1017767.0,
      "step": 455
    },
    {
      "entropy": 6.956169319152832,
      "epoch": 0.4975662520281233,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.6167,
      "mean_token_accuracy": 0.12303209006786346,
      "num_tokens": 1028286.0,
      "step": 460
    },
    {
      "entropy": 6.925253438949585,
      "epoch": 0.502974580854516,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.6247,
      "mean_token_accuracy": 0.11958765834569932,
      "num_tokens": 1040555.0,
      "step": 465
    },
    {
      "entropy": 7.025730276107788,
      "epoch": 0.5083829096809086,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.7262,
      "mean_token_accuracy": 0.12084319666028023,
      "num_tokens": 1053668.0,
      "step": 470
    },
    {
      "entropy": 6.828843450546264,
      "epoch": 0.5137912385073012,
      "grad_norm": 1.734375,
      "learning_rate": 0.000237,
      "loss": 6.6761,
      "mean_token_accuracy": 0.11609490439295769,
      "num_tokens": 1066042.0,
      "step": 475
    },
    {
      "entropy": 6.954677152633667,
      "epoch": 0.5191995673336939,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002395,
      "loss": 6.6768,
      "mean_token_accuracy": 0.1181111603975296,
      "num_tokens": 1077843.0,
      "step": 480
    },
    {
      "entropy": 6.920350646972656,
      "epoch": 0.5246078961600865,
      "grad_norm": 1.390625,
      "learning_rate": 0.000242,
      "loss": 6.5678,
      "mean_token_accuracy": 0.12242683097720146,
      "num_tokens": 1089448.0,
      "step": 485
    },
    {
      "entropy": 6.856211614608765,
      "epoch": 0.5300162249864792,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002445,
      "loss": 6.6594,
      "mean_token_accuracy": 0.12442992702126503,
      "num_tokens": 1100989.0,
      "step": 490
    },
    {
      "entropy": 7.0379091739654545,
      "epoch": 0.5354245538128718,
      "grad_norm": 1.359375,
      "learning_rate": 0.000247,
      "loss": 6.7094,
      "mean_token_accuracy": 0.12095102891325951,
      "num_tokens": 1113012.0,
      "step": 495
    },
    {
      "entropy": 6.903893089294433,
      "epoch": 0.5408328826392644,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002495,
      "loss": 6.6584,
      "mean_token_accuracy": 0.11823340728878975,
      "num_tokens": 1123551.0,
      "step": 500
    },
    {
      "epoch": 0.5408328826392644,
      "eval_entropy": 7.013918313113126,
      "eval_loss": 6.79232931137085,
      "eval_mean_token_accuracy": 0.11552235864272171,
      "eval_num_tokens": 1123551.0,
      "eval_runtime": 2.625,
      "eval_samples_per_second": 1339.063,
      "eval_steps_per_second": 167.621,
      "step": 500
    },
    {
      "entropy": 6.937011671066284,
      "epoch": 0.5462412114656571,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000252,
      "loss": 6.5714,
      "mean_token_accuracy": 0.1215690866112709,
      "num_tokens": 1133838.0,
      "step": 505
    },
    {
      "entropy": 6.832457637786865,
      "epoch": 0.5516495402920497,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002545,
      "loss": 6.6089,
      "mean_token_accuracy": 0.11541855782270431,
      "num_tokens": 1146005.0,
      "step": 510
    },
    {
      "entropy": 6.898182916641235,
      "epoch": 0.5570578691184424,
      "grad_norm": 1.375,
      "learning_rate": 0.000257,
      "loss": 6.6986,
      "mean_token_accuracy": 0.11636286452412606,
      "num_tokens": 1157328.0,
      "step": 515
    },
    {
      "entropy": 6.86890025138855,
      "epoch": 0.5624661979448351,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002595,
      "loss": 6.5041,
      "mean_token_accuracy": 0.1261405475437641,
      "num_tokens": 1168592.0,
      "step": 520
    },
    {
      "entropy": 6.840576457977295,
      "epoch": 0.5678745267712277,
      "grad_norm": 1.265625,
      "learning_rate": 0.000262,
      "loss": 6.6185,
      "mean_token_accuracy": 0.12170294821262359,
      "num_tokens": 1178680.0,
      "step": 525
    },
    {
      "entropy": 6.846650314331055,
      "epoch": 0.5732828555976204,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.5657,
      "mean_token_accuracy": 0.12621748968958854,
      "num_tokens": 1188825.0,
      "step": 530
    },
    {
      "entropy": 6.873343515396118,
      "epoch": 0.578691184424013,
      "grad_norm": 1.0625,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.6197,
      "mean_token_accuracy": 0.12520118281245232,
      "num_tokens": 1201375.0,
      "step": 535
    },
    {
      "entropy": 6.885719203948975,
      "epoch": 0.5840995132504057,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.7389,
      "mean_token_accuracy": 0.12146739810705184,
      "num_tokens": 1211915.0,
      "step": 540
    },
    {
      "entropy": 6.8768409252166744,
      "epoch": 0.5895078420767983,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.6373,
      "mean_token_accuracy": 0.12586278915405275,
      "num_tokens": 1223151.0,
      "step": 545
    },
    {
      "entropy": 6.8410814762115475,
      "epoch": 0.5949161709031909,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002745,
      "loss": 6.5612,
      "mean_token_accuracy": 0.1256933443248272,
      "num_tokens": 1234229.0,
      "step": 550
    },
    {
      "entropy": 6.868610095977783,
      "epoch": 0.6003244997295836,
      "grad_norm": 1.28125,
      "learning_rate": 0.000277,
      "loss": 6.5732,
      "mean_token_accuracy": 0.1251695491373539,
      "num_tokens": 1245298.0,
      "step": 555
    },
    {
      "entropy": 6.753601741790772,
      "epoch": 0.6057328285559762,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002795,
      "loss": 6.5604,
      "mean_token_accuracy": 0.12367500364780426,
      "num_tokens": 1256154.0,
      "step": 560
    },
    {
      "entropy": 6.8078868865966795,
      "epoch": 0.6111411573823688,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.4867,
      "mean_token_accuracy": 0.12721323147416114,
      "num_tokens": 1267404.0,
      "step": 565
    },
    {
      "entropy": 6.796451997756958,
      "epoch": 0.6165494862087615,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002845,
      "loss": 6.5311,
      "mean_token_accuracy": 0.12494980618357658,
      "num_tokens": 1278894.0,
      "step": 570
    },
    {
      "entropy": 6.720667362213135,
      "epoch": 0.6219578150351541,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000287,
      "loss": 6.5174,
      "mean_token_accuracy": 0.1313262701034546,
      "num_tokens": 1290408.0,
      "step": 575
    },
    {
      "entropy": 6.79968957901001,
      "epoch": 0.6273661438615468,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002895,
      "loss": 6.5932,
      "mean_token_accuracy": 0.12316951155662537,
      "num_tokens": 1301508.0,
      "step": 580
    },
    {
      "entropy": 6.8444901466369625,
      "epoch": 0.6327744726879394,
      "grad_norm": 1.234375,
      "learning_rate": 0.000292,
      "loss": 6.5698,
      "mean_token_accuracy": 0.1288432337343693,
      "num_tokens": 1312551.0,
      "step": 585
    },
    {
      "entropy": 6.841915416717529,
      "epoch": 0.638182801514332,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002945,
      "loss": 6.6691,
      "mean_token_accuracy": 0.12365695238113403,
      "num_tokens": 1324876.0,
      "step": 590
    },
    {
      "entropy": 6.743932723999023,
      "epoch": 0.6435911303407247,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000297,
      "loss": 6.5384,
      "mean_token_accuracy": 0.12478130236268044,
      "num_tokens": 1335401.0,
      "step": 595
    },
    {
      "entropy": 6.72650032043457,
      "epoch": 0.6489994591671173,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002995,
      "loss": 6.4655,
      "mean_token_accuracy": 0.12780711874365808,
      "num_tokens": 1348265.0,
      "step": 600
    },
    {
      "entropy": 6.636385107040406,
      "epoch": 0.65440778799351,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000302,
      "loss": 6.4219,
      "mean_token_accuracy": 0.13813963606953622,
      "num_tokens": 1359078.0,
      "step": 605
    },
    {
      "entropy": 6.532011079788208,
      "epoch": 0.6598161168199026,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003045,
      "loss": 6.3346,
      "mean_token_accuracy": 0.12941448390483856,
      "num_tokens": 1369499.0,
      "step": 610
    },
    {
      "entropy": 6.7241157531738285,
      "epoch": 0.6652244456462953,
      "grad_norm": 1.203125,
      "learning_rate": 0.000307,
      "loss": 6.4365,
      "mean_token_accuracy": 0.12599155828356742,
      "num_tokens": 1380329.0,
      "step": 615
    },
    {
      "entropy": 6.65777325630188,
      "epoch": 0.670632774472688,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003095,
      "loss": 6.5301,
      "mean_token_accuracy": 0.12588285207748412,
      "num_tokens": 1393047.0,
      "step": 620
    },
    {
      "entropy": 6.6903739929199215,
      "epoch": 0.6760411032990806,
      "grad_norm": 1.109375,
      "learning_rate": 0.000312,
      "loss": 6.5108,
      "mean_token_accuracy": 0.1278846800327301,
      "num_tokens": 1405600.0,
      "step": 625
    },
    {
      "entropy": 6.655543231964112,
      "epoch": 0.6814494321254733,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003145,
      "loss": 6.4692,
      "mean_token_accuracy": 0.12778620719909667,
      "num_tokens": 1416670.0,
      "step": 630
    },
    {
      "entropy": 6.649730110168457,
      "epoch": 0.6868577609518659,
      "grad_norm": 1.296875,
      "learning_rate": 0.000317,
      "loss": 6.4681,
      "mean_token_accuracy": 0.1273884281516075,
      "num_tokens": 1428929.0,
      "step": 635
    },
    {
      "entropy": 6.69134488105774,
      "epoch": 0.6922660897782585,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003195,
      "loss": 6.3925,
      "mean_token_accuracy": 0.12978972792625426,
      "num_tokens": 1439298.0,
      "step": 640
    },
    {
      "entropy": 6.6626770973205565,
      "epoch": 0.6976744186046512,
      "grad_norm": 1.171875,
      "learning_rate": 0.000322,
      "loss": 6.4936,
      "mean_token_accuracy": 0.13316625058650972,
      "num_tokens": 1449853.0,
      "step": 645
    },
    {
      "entropy": 6.727707958221435,
      "epoch": 0.7030827474310438,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.5531,
      "mean_token_accuracy": 0.1245946742594242,
      "num_tokens": 1460816.0,
      "step": 650
    },
    {
      "entropy": 6.585225248336792,
      "epoch": 0.7084910762574365,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.4105,
      "mean_token_accuracy": 0.12925421670079232,
      "num_tokens": 1473003.0,
      "step": 655
    },
    {
      "entropy": 6.654827499389649,
      "epoch": 0.7138994050838291,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.5262,
      "mean_token_accuracy": 0.12104681655764579,
      "num_tokens": 1484745.0,
      "step": 660
    },
    {
      "entropy": 6.636510896682739,
      "epoch": 0.7193077339102217,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.4073,
      "mean_token_accuracy": 0.1321321666240692,
      "num_tokens": 1495803.0,
      "step": 665
    },
    {
      "entropy": 6.633009243011474,
      "epoch": 0.7247160627366144,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.4088,
      "mean_token_accuracy": 0.1358610361814499,
      "num_tokens": 1507504.0,
      "step": 670
    },
    {
      "entropy": 6.664580249786377,
      "epoch": 0.730124391563007,
      "grad_norm": 1.078125,
      "learning_rate": 0.000337,
      "loss": 6.5178,
      "mean_token_accuracy": 0.12738374322652818,
      "num_tokens": 1518765.0,
      "step": 675
    },
    {
      "entropy": 6.511888456344605,
      "epoch": 0.7355327203893997,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003395,
      "loss": 6.3856,
      "mean_token_accuracy": 0.13164084926247596,
      "num_tokens": 1529481.0,
      "step": 680
    },
    {
      "entropy": 6.642051362991333,
      "epoch": 0.7409410492157923,
      "grad_norm": 1.28125,
      "learning_rate": 0.000342,
      "loss": 6.4132,
      "mean_token_accuracy": 0.13513489440083504,
      "num_tokens": 1539612.0,
      "step": 685
    },
    {
      "entropy": 6.646273422241211,
      "epoch": 0.7463493780421849,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.5721,
      "mean_token_accuracy": 0.12754858955740928,
      "num_tokens": 1551370.0,
      "step": 690
    },
    {
      "entropy": 6.6850049018859865,
      "epoch": 0.7517577068685776,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000347,
      "loss": 6.4495,
      "mean_token_accuracy": 0.13616673350334169,
      "num_tokens": 1561662.0,
      "step": 695
    },
    {
      "entropy": 6.5182334899902346,
      "epoch": 0.7571660356949702,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003495,
      "loss": 6.4259,
      "mean_token_accuracy": 0.1308984115719795,
      "num_tokens": 1573479.0,
      "step": 700
    },
    {
      "entropy": 6.6419507503509525,
      "epoch": 0.7625743645213628,
      "grad_norm": 1.375,
      "learning_rate": 0.000352,
      "loss": 6.3767,
      "mean_token_accuracy": 0.1341143622994423,
      "num_tokens": 1583684.0,
      "step": 705
    },
    {
      "entropy": 6.539843273162842,
      "epoch": 0.7679826933477556,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003545,
      "loss": 6.4271,
      "mean_token_accuracy": 0.13350974842906,
      "num_tokens": 1595694.0,
      "step": 710
    },
    {
      "entropy": 6.517860841751099,
      "epoch": 0.7733910221741482,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000357,
      "loss": 6.4161,
      "mean_token_accuracy": 0.12861086875200273,
      "num_tokens": 1606104.0,
      "step": 715
    },
    {
      "entropy": 6.653214359283448,
      "epoch": 0.7787993510005409,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003595,
      "loss": 6.4917,
      "mean_token_accuracy": 0.13703173398971558,
      "num_tokens": 1617368.0,
      "step": 720
    },
    {
      "entropy": 6.583222913742065,
      "epoch": 0.7842076798269335,
      "grad_norm": 1.0625,
      "learning_rate": 0.000362,
      "loss": 6.5023,
      "mean_token_accuracy": 0.12887703627347946,
      "num_tokens": 1630485.0,
      "step": 725
    },
    {
      "entropy": 6.571360731124878,
      "epoch": 0.7896160086533262,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003645,
      "loss": 6.3358,
      "mean_token_accuracy": 0.13361840322613716,
      "num_tokens": 1641411.0,
      "step": 730
    },
    {
      "entropy": 6.496680021286011,
      "epoch": 0.7950243374797188,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000367,
      "loss": 6.3417,
      "mean_token_accuracy": 0.1377808481454849,
      "num_tokens": 1653737.0,
      "step": 735
    },
    {
      "entropy": 6.627929592132569,
      "epoch": 0.8004326663061114,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003695,
      "loss": 6.5447,
      "mean_token_accuracy": 0.12955381646752356,
      "num_tokens": 1665910.0,
      "step": 740
    },
    {
      "entropy": 6.509966039657593,
      "epoch": 0.8058409951325041,
      "grad_norm": 1.203125,
      "learning_rate": 0.000372,
      "loss": 6.3154,
      "mean_token_accuracy": 0.1310278929769993,
      "num_tokens": 1676927.0,
      "step": 745
    },
    {
      "entropy": 6.544528341293335,
      "epoch": 0.8112493239588967,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003745,
      "loss": 6.3675,
      "mean_token_accuracy": 0.13159878849983214,
      "num_tokens": 1688167.0,
      "step": 750
    },
    {
      "entropy": 6.473386955261231,
      "epoch": 0.8166576527852893,
      "grad_norm": 1.25,
      "learning_rate": 0.000377,
      "loss": 6.4376,
      "mean_token_accuracy": 0.136484544724226,
      "num_tokens": 1699899.0,
      "step": 755
    },
    {
      "entropy": 6.567677068710327,
      "epoch": 0.822065981611682,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003795,
      "loss": 6.3601,
      "mean_token_accuracy": 0.1403407372534275,
      "num_tokens": 1709862.0,
      "step": 760
    },
    {
      "entropy": 6.378448438644409,
      "epoch": 0.8274743104380746,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000382,
      "loss": 6.3464,
      "mean_token_accuracy": 0.13634308502078057,
      "num_tokens": 1719902.0,
      "step": 765
    },
    {
      "entropy": 6.528745365142822,
      "epoch": 0.8328826392644673,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003845,
      "loss": 6.3591,
      "mean_token_accuracy": 0.13441544622182847,
      "num_tokens": 1730351.0,
      "step": 770
    },
    {
      "entropy": 6.510399198532104,
      "epoch": 0.8382909680908599,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.3384,
      "mean_token_accuracy": 0.13328076303005218,
      "num_tokens": 1741783.0,
      "step": 775
    },
    {
      "entropy": 6.322606420516967,
      "epoch": 0.8436992969172525,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.3697,
      "mean_token_accuracy": 0.13814696222543715,
      "num_tokens": 1752952.0,
      "step": 780
    },
    {
      "entropy": 6.606351613998413,
      "epoch": 0.8491076257436452,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.3959,
      "mean_token_accuracy": 0.13247718662023544,
      "num_tokens": 1764014.0,
      "step": 785
    },
    {
      "entropy": 6.447211217880249,
      "epoch": 0.8545159545700378,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.3455,
      "mean_token_accuracy": 0.13544300198554993,
      "num_tokens": 1776154.0,
      "step": 790
    },
    {
      "entropy": 6.516087865829467,
      "epoch": 0.8599242833964305,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.4172,
      "mean_token_accuracy": 0.13449665680527687,
      "num_tokens": 1787292.0,
      "step": 795
    },
    {
      "entropy": 6.481107139587403,
      "epoch": 0.8653326122228231,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003995,
      "loss": 6.3821,
      "mean_token_accuracy": 0.13621541485190392,
      "num_tokens": 1799469.0,
      "step": 800
    },
    {
      "entropy": 6.4164855003356935,
      "epoch": 0.8707409410492158,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000402,
      "loss": 6.2457,
      "mean_token_accuracy": 0.14523041546344756,
      "num_tokens": 1810091.0,
      "step": 805
    },
    {
      "entropy": 6.484155941009521,
      "epoch": 0.8761492698756085,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004045,
      "loss": 6.405,
      "mean_token_accuracy": 0.13718786239624023,
      "num_tokens": 1820835.0,
      "step": 810
    },
    {
      "entropy": 6.451174211502075,
      "epoch": 0.8815575987020011,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.364,
      "mean_token_accuracy": 0.13937411159276963,
      "num_tokens": 1831868.0,
      "step": 815
    },
    {
      "entropy": 6.407666540145874,
      "epoch": 0.8869659275283938,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004095,
      "loss": 6.3273,
      "mean_token_accuracy": 0.13826591596007348,
      "num_tokens": 1842040.0,
      "step": 820
    },
    {
      "entropy": 6.53410325050354,
      "epoch": 0.8923742563547864,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000412,
      "loss": 6.4122,
      "mean_token_accuracy": 0.1363173633813858,
      "num_tokens": 1853765.0,
      "step": 825
    },
    {
      "entropy": 6.515021228790284,
      "epoch": 0.897782585181179,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004145,
      "loss": 6.4173,
      "mean_token_accuracy": 0.13506757691502572,
      "num_tokens": 1866081.0,
      "step": 830
    },
    {
      "entropy": 6.419595050811767,
      "epoch": 0.9031909140075717,
      "grad_norm": 1.234375,
      "learning_rate": 0.000417,
      "loss": 6.3213,
      "mean_token_accuracy": 0.1396598130464554,
      "num_tokens": 1876754.0,
      "step": 835
    },
    {
      "entropy": 6.540018558502197,
      "epoch": 0.9085992428339643,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004195,
      "loss": 6.4374,
      "mean_token_accuracy": 0.13417667225003244,
      "num_tokens": 1888299.0,
      "step": 840
    },
    {
      "entropy": 6.325070524215699,
      "epoch": 0.914007571660357,
      "grad_norm": 1.09375,
      "learning_rate": 0.000422,
      "loss": 6.2112,
      "mean_token_accuracy": 0.14675599336624146,
      "num_tokens": 1899027.0,
      "step": 845
    },
    {
      "entropy": 6.455481433868409,
      "epoch": 0.9194159004867496,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004245,
      "loss": 6.3336,
      "mean_token_accuracy": 0.13757723793387414,
      "num_tokens": 1911043.0,
      "step": 850
    },
    {
      "entropy": 6.381244230270386,
      "epoch": 0.9248242293131422,
      "grad_norm": 1.125,
      "learning_rate": 0.000427,
      "loss": 6.2676,
      "mean_token_accuracy": 0.13861764669418336,
      "num_tokens": 1921295.0,
      "step": 855
    },
    {
      "entropy": 6.472774600982666,
      "epoch": 0.9302325581395349,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004295,
      "loss": 6.3636,
      "mean_token_accuracy": 0.14171690344810486,
      "num_tokens": 1932944.0,
      "step": 860
    },
    {
      "entropy": 6.311262941360473,
      "epoch": 0.9356408869659275,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000432,
      "loss": 6.2982,
      "mean_token_accuracy": 0.13167063891887665,
      "num_tokens": 1944072.0,
      "step": 865
    },
    {
      "entropy": 6.397356128692627,
      "epoch": 0.9410492157923201,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004345,
      "loss": 6.3051,
      "mean_token_accuracy": 0.14503950625658035,
      "num_tokens": 1954798.0,
      "step": 870
    },
    {
      "entropy": 6.3590000629425045,
      "epoch": 0.9464575446187128,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000437,
      "loss": 6.2638,
      "mean_token_accuracy": 0.14990901201963425,
      "num_tokens": 1965304.0,
      "step": 875
    },
    {
      "entropy": 6.482557010650635,
      "epoch": 0.9518658734451054,
      "grad_norm": 1.125,
      "learning_rate": 0.0004395,
      "loss": 6.3088,
      "mean_token_accuracy": 0.13219384998083114,
      "num_tokens": 1976797.0,
      "step": 880
    },
    {
      "entropy": 6.331461238861084,
      "epoch": 0.9572742022714981,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000442,
      "loss": 6.335,
      "mean_token_accuracy": 0.13832038938999175,
      "num_tokens": 1987448.0,
      "step": 885
    },
    {
      "entropy": 6.1668407917022705,
      "epoch": 0.9626825310978907,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004445,
      "loss": 6.1689,
      "mean_token_accuracy": 0.13529722392559052,
      "num_tokens": 2000948.0,
      "step": 890
    },
    {
      "entropy": 6.391203689575195,
      "epoch": 0.9680908599242833,
      "grad_norm": 1.140625,
      "learning_rate": 0.000447,
      "loss": 6.2438,
      "mean_token_accuracy": 0.13901959732174873,
      "num_tokens": 2011807.0,
      "step": 895
    },
    {
      "entropy": 6.355871772766113,
      "epoch": 0.9734991887506761,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.3258,
      "mean_token_accuracy": 0.13166970759630203,
      "num_tokens": 2022835.0,
      "step": 900
    },
    {
      "entropy": 6.3004621982574465,
      "epoch": 0.9789075175770687,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.233,
      "mean_token_accuracy": 0.13900147005915642,
      "num_tokens": 2034331.0,
      "step": 905
    },
    {
      "entropy": 6.364811372756958,
      "epoch": 0.9843158464034614,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.242,
      "mean_token_accuracy": 0.13642507493495942,
      "num_tokens": 2047002.0,
      "step": 910
    },
    {
      "entropy": 6.2332769393920895,
      "epoch": 0.989724175229854,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.1739,
      "mean_token_accuracy": 0.14840540438890457,
      "num_tokens": 2056299.0,
      "step": 915
    },
    {
      "entropy": 6.289453458786011,
      "epoch": 0.9951325040562466,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.2991,
      "mean_token_accuracy": 0.14142733812332153,
      "num_tokens": 2068031.0,
      "step": 920
    },
    {
      "entropy": 6.354425801171197,
      "epoch": 1.0,
      "grad_norm": 1.9296875,
      "learning_rate": 0.000462,
      "loss": 6.2578,
      "mean_token_accuracy": 0.1445001173350546,
      "num_tokens": 2076889.0,
      "step": 925
    },
    {
      "entropy": 6.325593137741089,
      "epoch": 1.0054083288263926,
      "grad_norm": 1.25,
      "learning_rate": 0.0004645,
      "loss": 6.0941,
      "mean_token_accuracy": 0.1489081636071205,
      "num_tokens": 2087316.0,
      "step": 930
    },
    {
      "entropy": 6.279916524887085,
      "epoch": 1.0108166576527853,
      "grad_norm": 1.078125,
      "learning_rate": 0.000467,
      "loss": 6.0993,
      "mean_token_accuracy": 0.1413825236260891,
      "num_tokens": 2098522.0,
      "step": 935
    },
    {
      "entropy": 6.153361511230469,
      "epoch": 1.016224986479178,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004695,
      "loss": 6.0827,
      "mean_token_accuracy": 0.14982683211565018,
      "num_tokens": 2110015.0,
      "step": 940
    },
    {
      "entropy": 6.294631814956665,
      "epoch": 1.0216333153055706,
      "grad_norm": 1.140625,
      "learning_rate": 0.000472,
      "loss": 6.1968,
      "mean_token_accuracy": 0.13932642787694932,
      "num_tokens": 2121333.0,
      "step": 945
    },
    {
      "entropy": 6.309599113464356,
      "epoch": 1.0270416441319632,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004745,
      "loss": 6.0672,
      "mean_token_accuracy": 0.15198333263397218,
      "num_tokens": 2131385.0,
      "step": 950
    },
    {
      "entropy": 6.302496814727784,
      "epoch": 1.0324499729583558,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000477,
      "loss": 6.1164,
      "mean_token_accuracy": 0.15040989816188813,
      "num_tokens": 2143274.0,
      "step": 955
    },
    {
      "entropy": 6.08814058303833,
      "epoch": 1.0378583017847485,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004795,
      "loss": 6.0022,
      "mean_token_accuracy": 0.150777405500412,
      "num_tokens": 2154529.0,
      "step": 960
    },
    {
      "entropy": 6.278039503097534,
      "epoch": 1.043266630611141,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000482,
      "loss": 6.1564,
      "mean_token_accuracy": 0.1439770504832268,
      "num_tokens": 2165949.0,
      "step": 965
    },
    {
      "entropy": 6.256474876403809,
      "epoch": 1.0486749594375337,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004845,
      "loss": 6.1395,
      "mean_token_accuracy": 0.14725938364863395,
      "num_tokens": 2176710.0,
      "step": 970
    },
    {
      "entropy": 6.2467734813690186,
      "epoch": 1.0540832882639264,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000487,
      "loss": 6.1028,
      "mean_token_accuracy": 0.15126092433929444,
      "num_tokens": 2189470.0,
      "step": 975
    },
    {
      "entropy": 6.138709020614624,
      "epoch": 1.059491617090319,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004895,
      "loss": 6.0725,
      "mean_token_accuracy": 0.15428618490695953,
      "num_tokens": 2200266.0,
      "step": 980
    },
    {
      "entropy": 6.209649991989136,
      "epoch": 1.0648999459167117,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000492,
      "loss": 6.086,
      "mean_token_accuracy": 0.15003882795572282,
      "num_tokens": 2210736.0,
      "step": 985
    },
    {
      "entropy": 6.291779804229736,
      "epoch": 1.0703082747431043,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004945,
      "loss": 6.197,
      "mean_token_accuracy": 0.14641276448965074,
      "num_tokens": 2221057.0,
      "step": 990
    },
    {
      "entropy": 6.208144187927246,
      "epoch": 1.075716603569497,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000497,
      "loss": 6.1826,
      "mean_token_accuracy": 0.14815677255392073,
      "num_tokens": 2232352.0,
      "step": 995
    },
    {
      "entropy": 6.239853429794311,
      "epoch": 1.0811249323958896,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004995,
      "loss": 6.0704,
      "mean_token_accuracy": 0.1508354812860489,
      "num_tokens": 2244655.0,
      "step": 1000
    },
    {
      "epoch": 1.0811249323958896,
      "eval_entropy": 6.042540579492395,
      "eval_loss": 6.387476444244385,
      "eval_mean_token_accuracy": 0.14328089801763946,
      "eval_num_tokens": 2244655.0,
      "eval_runtime": 2.5132,
      "eval_samples_per_second": 1398.64,
      "eval_steps_per_second": 175.079,
      "step": 1000
    }
  ],
  "logging_steps": 5,
  "max_steps": 9240,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3492617103360000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}