{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.2427506213753108,
  "eval_steps": 500,
  "global_step": 1500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.7426025390625,
      "epoch": 0.004142502071251036,
      "grad_norm": 4.96875,
      "learning_rate": 2e-06,
      "loss": 10.7521,
      "mean_token_accuracy": 9.009009227156639e-05,
      "num_tokens": 9149.0,
      "step": 5
    },
    {
      "entropy": 10.742637062072754,
      "epoch": 0.008285004142502071,
      "grad_norm": 5.3125,
      "learning_rate": 4.5e-06,
      "loss": 10.7074,
      "mean_token_accuracy": 0.0,
      "num_tokens": 16800.0,
      "step": 10
    },
    {
      "entropy": 10.742610836029053,
      "epoch": 0.012427506213753107,
      "grad_norm": 4.90625,
      "learning_rate": 7e-06,
      "loss": 10.743,
      "mean_token_accuracy": 0.0,
      "num_tokens": 26617.0,
      "step": 15
    },
    {
      "entropy": 10.742677021026612,
      "epoch": 0.016570008285004142,
      "grad_norm": 5.15625,
      "learning_rate": 9.5e-06,
      "loss": 10.6772,
      "mean_token_accuracy": 0.0,
      "num_tokens": 35855.0,
      "step": 20
    },
    {
      "entropy": 10.742655372619629,
      "epoch": 0.020712510356255178,
      "grad_norm": 4.40625,
      "learning_rate": 1.2e-05,
      "loss": 10.5975,
      "mean_token_accuracy": 0.002011376223526895,
      "num_tokens": 45644.0,
      "step": 25
    },
    {
      "entropy": 10.742458724975586,
      "epoch": 0.024855012427506214,
      "grad_norm": 3.859375,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 10.4485,
      "mean_token_accuracy": 0.019727057497948407,
      "num_tokens": 54641.0,
      "step": 30
    },
    {
      "entropy": 10.741547870635987,
      "epoch": 0.02899751449875725,
      "grad_norm": 3.40625,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.3138,
      "mean_token_accuracy": 0.04090950228273869,
      "num_tokens": 63924.0,
      "step": 35
    },
    {
      "entropy": 10.739116668701172,
      "epoch": 0.033140016570008285,
      "grad_norm": 2.609375,
      "learning_rate": 1.95e-05,
      "loss": 10.2108,
      "mean_token_accuracy": 0.041662289202213286,
      "num_tokens": 73452.0,
      "step": 40
    },
    {
      "entropy": 10.735450172424317,
      "epoch": 0.037282518641259324,
      "grad_norm": 2.203125,
      "learning_rate": 2.2e-05,
      "loss": 10.0929,
      "mean_token_accuracy": 0.039270054548978806,
      "num_tokens": 82731.0,
      "step": 45
    },
    {
      "entropy": 10.733051109313966,
      "epoch": 0.041425020712510356,
      "grad_norm": 2.171875,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 10.0119,
      "mean_token_accuracy": 0.04254167750477791,
      "num_tokens": 92626.0,
      "step": 50
    },
    {
      "entropy": 10.732081508636474,
      "epoch": 0.045567522783761395,
      "grad_norm": 1.9921875,
      "learning_rate": 2.7e-05,
      "loss": 9.9713,
      "mean_token_accuracy": 0.04115805197507143,
      "num_tokens": 101380.0,
      "step": 55
    },
    {
      "entropy": 10.730613136291504,
      "epoch": 0.04971002485501243,
      "grad_norm": 1.828125,
      "learning_rate": 2.95e-05,
      "loss": 9.8815,
      "mean_token_accuracy": 0.041873040795326236,
      "num_tokens": 109895.0,
      "step": 60
    },
    {
      "entropy": 10.728298664093018,
      "epoch": 0.053852526926263466,
      "grad_norm": 1.9140625,
      "learning_rate": 3.2e-05,
      "loss": 9.8626,
      "mean_token_accuracy": 0.04000609517097473,
      "num_tokens": 119203.0,
      "step": 65
    },
    {
      "entropy": 10.7264967918396,
      "epoch": 0.0579950289975145,
      "grad_norm": 2.203125,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.7547,
      "mean_token_accuracy": 0.03999764695763588,
      "num_tokens": 127572.0,
      "step": 70
    },
    {
      "entropy": 10.723791694641113,
      "epoch": 0.06213753106876554,
      "grad_norm": 1.796875,
      "learning_rate": 3.7e-05,
      "loss": 9.7222,
      "mean_token_accuracy": 0.042604190111160276,
      "num_tokens": 137161.0,
      "step": 75
    },
    {
      "entropy": 10.719281482696534,
      "epoch": 0.06628003314001657,
      "grad_norm": 1.84375,
      "learning_rate": 3.95e-05,
      "loss": 9.6617,
      "mean_token_accuracy": 0.04084589965641498,
      "num_tokens": 145991.0,
      "step": 80
    },
    {
      "entropy": 10.712574768066407,
      "epoch": 0.07042253521126761,
      "grad_norm": 1.796875,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.598,
      "mean_token_accuracy": 0.041171186976134776,
      "num_tokens": 154927.0,
      "step": 85
    },
    {
      "entropy": 10.706455898284911,
      "epoch": 0.07456503728251865,
      "grad_norm": 1.953125,
      "learning_rate": 4.45e-05,
      "loss": 9.4593,
      "mean_token_accuracy": 0.042302171140909194,
      "num_tokens": 163234.0,
      "step": 90
    },
    {
      "entropy": 10.699676418304444,
      "epoch": 0.07870753935376967,
      "grad_norm": 1.8203125,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 9.4018,
      "mean_token_accuracy": 0.04330670088529587,
      "num_tokens": 171554.0,
      "step": 95
    },
    {
      "entropy": 10.688521003723144,
      "epoch": 0.08285004142502071,
      "grad_norm": 1.8046875,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 9.3311,
      "mean_token_accuracy": 0.047984727472066876,
      "num_tokens": 180171.0,
      "step": 100
    },
    {
      "entropy": 10.67230396270752,
      "epoch": 0.08699254349627175,
      "grad_norm": 1.8125,
      "learning_rate": 5.2e-05,
      "loss": 9.2155,
      "mean_token_accuracy": 0.057312173396348955,
      "num_tokens": 189216.0,
      "step": 105
    },
    {
      "entropy": 10.645287895202637,
      "epoch": 0.09113504556752279,
      "grad_norm": 1.71875,
      "learning_rate": 5.45e-05,
      "loss": 9.2219,
      "mean_token_accuracy": 0.052681009843945505,
      "num_tokens": 200022.0,
      "step": 110
    },
    {
      "entropy": 10.617320728302001,
      "epoch": 0.09527754763877382,
      "grad_norm": 1.828125,
      "learning_rate": 5.7e-05,
      "loss": 9.0422,
      "mean_token_accuracy": 0.05372999124228954,
      "num_tokens": 208686.0,
      "step": 115
    },
    {
      "entropy": 10.564020252227783,
      "epoch": 0.09942004971002485,
      "grad_norm": 1.8125,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.9322,
      "mean_token_accuracy": 0.054267140850424764,
      "num_tokens": 218266.0,
      "step": 120
    },
    {
      "entropy": 10.5111478805542,
      "epoch": 0.1035625517812759,
      "grad_norm": 1.625,
      "learning_rate": 6.2e-05,
      "loss": 8.8288,
      "mean_token_accuracy": 0.04845710247755051,
      "num_tokens": 227658.0,
      "step": 125
    },
    {
      "entropy": 10.454801654815673,
      "epoch": 0.10770505385252693,
      "grad_norm": 1.5234375,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.6344,
      "mean_token_accuracy": 0.06026250720024109,
      "num_tokens": 236301.0,
      "step": 130
    },
    {
      "entropy": 10.368635272979736,
      "epoch": 0.11184755592377796,
      "grad_norm": 1.609375,
      "learning_rate": 6.7e-05,
      "loss": 8.644,
      "mean_token_accuracy": 0.048985954746603964,
      "num_tokens": 246355.0,
      "step": 135
    },
    {
      "entropy": 10.31434211730957,
      "epoch": 0.115990057995029,
      "grad_norm": 1.5,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.4805,
      "mean_token_accuracy": 0.05331834591925144,
      "num_tokens": 256023.0,
      "step": 140
    },
    {
      "entropy": 10.216940021514892,
      "epoch": 0.12013256006628004,
      "grad_norm": 1.4140625,
      "learning_rate": 7.2e-05,
      "loss": 8.4309,
      "mean_token_accuracy": 0.04547450765967369,
      "num_tokens": 266893.0,
      "step": 145
    },
    {
      "entropy": 10.09643030166626,
      "epoch": 0.12427506213753108,
      "grad_norm": 1.328125,
      "learning_rate": 7.45e-05,
      "loss": 8.1871,
      "mean_token_accuracy": 0.0535514272749424,
      "num_tokens": 276075.0,
      "step": 150
    },
    {
      "entropy": 9.940307807922363,
      "epoch": 0.12841756420878211,
      "grad_norm": 1.234375,
      "learning_rate": 7.7e-05,
      "loss": 8.1018,
      "mean_token_accuracy": 0.055915596336126326,
      "num_tokens": 285529.0,
      "step": 155
    },
    {
      "entropy": 9.774145030975342,
      "epoch": 0.13256006628003314,
      "grad_norm": 1.1328125,
      "learning_rate": 7.950000000000001e-05,
      "loss": 8.0263,
      "mean_token_accuracy": 0.05546179860830307,
      "num_tokens": 294403.0,
      "step": 160
    },
    {
      "entropy": 9.571882343292236,
      "epoch": 0.13670256835128416,
      "grad_norm": 1.109375,
      "learning_rate": 8.2e-05,
      "loss": 7.9661,
      "mean_token_accuracy": 0.05883818194270134,
      "num_tokens": 303507.0,
      "step": 165
    },
    {
      "entropy": 9.380561351776123,
      "epoch": 0.14084507042253522,
      "grad_norm": 1.15625,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.7977,
      "mean_token_accuracy": 0.06465772800147533,
      "num_tokens": 312393.0,
      "step": 170
    },
    {
      "entropy": 9.121123886108398,
      "epoch": 0.14498757249378624,
      "grad_norm": 0.88671875,
      "learning_rate": 8.7e-05,
      "loss": 7.7071,
      "mean_token_accuracy": 0.06362416185438632,
      "num_tokens": 321309.0,
      "step": 175
    },
    {
      "entropy": 8.919694709777833,
      "epoch": 0.1491300745650373,
      "grad_norm": 0.94140625,
      "learning_rate": 8.95e-05,
      "loss": 7.7209,
      "mean_token_accuracy": 0.06159159280359745,
      "num_tokens": 330217.0,
      "step": 180
    },
    {
      "entropy": 8.755216789245605,
      "epoch": 0.15327257663628832,
      "grad_norm": 0.98046875,
      "learning_rate": 9.2e-05,
      "loss": 7.6795,
      "mean_token_accuracy": 0.061438082903623584,
      "num_tokens": 339587.0,
      "step": 185
    },
    {
      "entropy": 8.5862530708313,
      "epoch": 0.15741507870753935,
      "grad_norm": 1.140625,
      "learning_rate": 9.45e-05,
      "loss": 7.5864,
      "mean_token_accuracy": 0.06275138966739177,
      "num_tokens": 349367.0,
      "step": 190
    },
    {
      "entropy": 8.414569091796874,
      "epoch": 0.1615575807787904,
      "grad_norm": 0.9296875,
      "learning_rate": 9.7e-05,
      "loss": 7.4852,
      "mean_token_accuracy": 0.06507482491433621,
      "num_tokens": 358272.0,
      "step": 195
    },
    {
      "entropy": 8.34211015701294,
      "epoch": 0.16570008285004142,
      "grad_norm": 0.92578125,
      "learning_rate": 9.95e-05,
      "loss": 7.6662,
      "mean_token_accuracy": 0.06489905305206775,
      "num_tokens": 367709.0,
      "step": 200
    },
    {
      "entropy": 8.344364166259766,
      "epoch": 0.16984258492129245,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000102,
      "loss": 7.5038,
      "mean_token_accuracy": 0.06977917924523354,
      "num_tokens": 376180.0,
      "step": 205
    },
    {
      "entropy": 8.172031211853028,
      "epoch": 0.1739850869925435,
      "grad_norm": 0.84765625,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.6071,
      "mean_token_accuracy": 0.06480413824319839,
      "num_tokens": 385636.0,
      "step": 210
    },
    {
      "entropy": 8.20558671951294,
      "epoch": 0.17812758906379453,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000107,
      "loss": 7.4616,
      "mean_token_accuracy": 0.07232532948255539,
      "num_tokens": 394486.0,
      "step": 215
    },
    {
      "entropy": 8.116187858581544,
      "epoch": 0.18227009113504558,
      "grad_norm": 1.0,
      "learning_rate": 0.0001095,
      "loss": 7.4215,
      "mean_token_accuracy": 0.06673903428018094,
      "num_tokens": 403765.0,
      "step": 220
    },
    {
      "entropy": 8.066840410232544,
      "epoch": 0.1864125932062966,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000112,
      "loss": 7.4037,
      "mean_token_accuracy": 0.07279244847595692,
      "num_tokens": 412988.0,
      "step": 225
    },
    {
      "entropy": 8.047943782806396,
      "epoch": 0.19055509527754763,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0001145,
      "loss": 7.3598,
      "mean_token_accuracy": 0.07073828913271427,
      "num_tokens": 422096.0,
      "step": 230
    },
    {
      "entropy": 7.999078845977783,
      "epoch": 0.19469759734879868,
      "grad_norm": 1.078125,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.441,
      "mean_token_accuracy": 0.06653320901095867,
      "num_tokens": 431333.0,
      "step": 235
    },
    {
      "entropy": 8.124928903579711,
      "epoch": 0.1988400994200497,
      "grad_norm": 1.1875,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.4798,
      "mean_token_accuracy": 0.07552412785589695,
      "num_tokens": 440382.0,
      "step": 240
    },
    {
      "entropy": 7.967213916778564,
      "epoch": 0.20298260149130073,
      "grad_norm": 1.0,
      "learning_rate": 0.000122,
      "loss": 7.3739,
      "mean_token_accuracy": 0.07711177170276642,
      "num_tokens": 448469.0,
      "step": 245
    },
    {
      "entropy": 8.066874313354493,
      "epoch": 0.2071251035625518,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001245,
      "loss": 7.4382,
      "mean_token_accuracy": 0.07283047214150429,
      "num_tokens": 457356.0,
      "step": 250
    },
    {
      "entropy": 8.006765127182007,
      "epoch": 0.2112676056338028,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000127,
      "loss": 7.4372,
      "mean_token_accuracy": 0.07371915206313133,
      "num_tokens": 465928.0,
      "step": 255
    },
    {
      "entropy": 7.962953233718872,
      "epoch": 0.21541010770505387,
      "grad_norm": 1.03125,
      "learning_rate": 0.0001295,
      "loss": 7.3086,
      "mean_token_accuracy": 0.07748371884226798,
      "num_tokens": 475269.0,
      "step": 260
    },
    {
      "entropy": 7.846234321594238,
      "epoch": 0.2195526097763049,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000132,
      "loss": 7.2837,
      "mean_token_accuracy": 0.07470350190997124,
      "num_tokens": 485450.0,
      "step": 265
    },
    {
      "entropy": 7.911148262023926,
      "epoch": 0.22369511184755592,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.3174,
      "mean_token_accuracy": 0.07727364040911197,
      "num_tokens": 494029.0,
      "step": 270
    },
    {
      "entropy": 7.862251615524292,
      "epoch": 0.22783761391880697,
      "grad_norm": 1.046875,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.2856,
      "mean_token_accuracy": 0.07773918211460114,
      "num_tokens": 503253.0,
      "step": 275
    },
    {
      "entropy": 7.789915609359741,
      "epoch": 0.231980115990058,
      "grad_norm": 1.046875,
      "learning_rate": 0.0001395,
      "loss": 7.2873,
      "mean_token_accuracy": 0.07239806838333607,
      "num_tokens": 512857.0,
      "step": 280
    },
    {
      "entropy": 7.920849275588989,
      "epoch": 0.23612261806130902,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.3294,
      "mean_token_accuracy": 0.07429277002811432,
      "num_tokens": 522319.0,
      "step": 285
    },
    {
      "entropy": 7.767004537582397,
      "epoch": 0.24026512013256007,
      "grad_norm": 0.953125,
      "learning_rate": 0.0001445,
      "loss": 7.2731,
      "mean_token_accuracy": 0.07617050744593143,
      "num_tokens": 531469.0,
      "step": 290
    },
    {
      "entropy": 7.817885971069336,
      "epoch": 0.2444076222038111,
      "grad_norm": 1.171875,
      "learning_rate": 0.000147,
      "loss": 7.2883,
      "mean_token_accuracy": 0.07853077799081802,
      "num_tokens": 540470.0,
      "step": 295
    },
    {
      "entropy": 7.739614582061767,
      "epoch": 0.24855012427506215,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0001495,
      "loss": 7.2711,
      "mean_token_accuracy": 0.07688240185379983,
      "num_tokens": 550317.0,
      "step": 300
    },
    {
      "entropy": 7.804083776473999,
      "epoch": 0.2526926263463132,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000152,
      "loss": 7.2542,
      "mean_token_accuracy": 0.07694350108504296,
      "num_tokens": 559409.0,
      "step": 305
    },
    {
      "entropy": 7.721450233459473,
      "epoch": 0.25683512841756423,
      "grad_norm": 1.015625,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.2393,
      "mean_token_accuracy": 0.0897148072719574,
      "num_tokens": 568726.0,
      "step": 310
    },
    {
      "entropy": 7.8209740161895756,
      "epoch": 0.2609776304888152,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000157,
      "loss": 7.26,
      "mean_token_accuracy": 0.07505979798734189,
      "num_tokens": 577926.0,
      "step": 315
    },
    {
      "entropy": 7.7856145858764645,
      "epoch": 0.2651201325600663,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0001595,
      "loss": 7.2553,
      "mean_token_accuracy": 0.07944877073168755,
      "num_tokens": 586695.0,
      "step": 320
    },
    {
      "entropy": 7.730410480499268,
      "epoch": 0.26926263463131733,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000162,
      "loss": 7.264,
      "mean_token_accuracy": 0.07737773731350898,
      "num_tokens": 596250.0,
      "step": 325
    },
    {
      "entropy": 7.764742994308472,
      "epoch": 0.27340513670256833,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.2863,
      "mean_token_accuracy": 0.07472797855734825,
      "num_tokens": 605761.0,
      "step": 330
    },
    {
      "entropy": 7.720160007476807,
      "epoch": 0.2775476387738194,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.2106,
      "mean_token_accuracy": 0.07840372547507286,
      "num_tokens": 614313.0,
      "step": 335
    },
    {
      "entropy": 7.699429607391357,
      "epoch": 0.28169014084507044,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.2329,
      "mean_token_accuracy": 0.079537483304739,
      "num_tokens": 624470.0,
      "step": 340
    },
    {
      "entropy": 7.716359567642212,
      "epoch": 0.28583264291632143,
      "grad_norm": 1.015625,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.1781,
      "mean_token_accuracy": 0.07983956411480904,
      "num_tokens": 633458.0,
      "step": 345
    },
    {
      "entropy": 7.705728006362915,
      "epoch": 0.2899751449875725,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.1396,
      "mean_token_accuracy": 0.08111847266554832,
      "num_tokens": 643251.0,
      "step": 350
    },
    {
      "entropy": 7.584441757202148,
      "epoch": 0.29411764705882354,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000177,
      "loss": 7.2011,
      "mean_token_accuracy": 0.08007878810167313,
      "num_tokens": 652005.0,
      "step": 355
    },
    {
      "entropy": 7.649393129348755,
      "epoch": 0.2982601491300746,
      "grad_norm": 1.0625,
      "learning_rate": 0.0001795,
      "loss": 7.0934,
      "mean_token_accuracy": 0.08189089968800545,
      "num_tokens": 661003.0,
      "step": 360
    },
    {
      "entropy": 7.623857879638672,
      "epoch": 0.3024026512013256,
      "grad_norm": 1.0625,
      "learning_rate": 0.000182,
      "loss": 7.0933,
      "mean_token_accuracy": 0.08178795203566551,
      "num_tokens": 670221.0,
      "step": 365
    },
    {
      "entropy": 7.750538110733032,
      "epoch": 0.30654515327257664,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001845,
      "loss": 7.3477,
      "mean_token_accuracy": 0.07552824206650258,
      "num_tokens": 680756.0,
      "step": 370
    },
    {
      "entropy": 7.595584869384766,
      "epoch": 0.3106876553438277,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000187,
      "loss": 7.0678,
      "mean_token_accuracy": 0.08516154885292053,
      "num_tokens": 689641.0,
      "step": 375
    },
    {
      "entropy": 7.549710321426391,
      "epoch": 0.3148301574150787,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001895,
      "loss": 6.995,
      "mean_token_accuracy": 0.0825497955083847,
      "num_tokens": 698209.0,
      "step": 380
    },
    {
      "entropy": 7.580410575866699,
      "epoch": 0.31897265948632975,
      "grad_norm": 1.0,
      "learning_rate": 0.000192,
      "loss": 6.9957,
      "mean_token_accuracy": 0.08926377072930336,
      "num_tokens": 706712.0,
      "step": 385
    },
    {
      "entropy": 7.595743894577026,
      "epoch": 0.3231151615575808,
      "grad_norm": 1.078125,
      "learning_rate": 0.0001945,
      "loss": 7.1863,
      "mean_token_accuracy": 0.08130085095763206,
      "num_tokens": 716087.0,
      "step": 390
    },
    {
      "entropy": 7.5478174686431885,
      "epoch": 0.3272576636288318,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.0828,
      "mean_token_accuracy": 0.07734477743506432,
      "num_tokens": 725913.0,
      "step": 395
    },
    {
      "entropy": 7.606314182281494,
      "epoch": 0.33140016570008285,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.1209,
      "mean_token_accuracy": 0.0859277956187725,
      "num_tokens": 735020.0,
      "step": 400
    },
    {
      "entropy": 7.527130556106568,
      "epoch": 0.3355426677713339,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000202,
      "loss": 7.0165,
      "mean_token_accuracy": 0.07862741872668266,
      "num_tokens": 744299.0,
      "step": 405
    },
    {
      "entropy": 7.538954830169677,
      "epoch": 0.3396851698425849,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.9993,
      "mean_token_accuracy": 0.08304714113473892,
      "num_tokens": 753231.0,
      "step": 410
    },
    {
      "entropy": 7.5157403469085695,
      "epoch": 0.34382767191383595,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000207,
      "loss": 7.0619,
      "mean_token_accuracy": 0.08629989027976989,
      "num_tokens": 763043.0,
      "step": 415
    },
    {
      "entropy": 7.627204179763794,
      "epoch": 0.347970173985087,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002095,
      "loss": 7.0861,
      "mean_token_accuracy": 0.08552133813500404,
      "num_tokens": 771977.0,
      "step": 420
    },
    {
      "entropy": 7.432950973510742,
      "epoch": 0.352112676056338,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000212,
      "loss": 6.9878,
      "mean_token_accuracy": 0.08852855414152146,
      "num_tokens": 780285.0,
      "step": 425
    },
    {
      "entropy": 7.566639614105225,
      "epoch": 0.35625517812758906,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002145,
      "loss": 7.047,
      "mean_token_accuracy": 0.08173825293779373,
      "num_tokens": 789980.0,
      "step": 430
    },
    {
      "entropy": 7.483531761169433,
      "epoch": 0.3603976801988401,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021700000000000002,
      "loss": 7.0026,
      "mean_token_accuracy": 0.09233896881341934,
      "num_tokens": 798700.0,
      "step": 435
    },
    {
      "entropy": 7.38806381225586,
      "epoch": 0.36454018227009116,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0002195,
      "loss": 7.059,
      "mean_token_accuracy": 0.08308355435729027,
      "num_tokens": 808758.0,
      "step": 440
    },
    {
      "entropy": 7.56745285987854,
      "epoch": 0.36868268434134216,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000222,
      "loss": 7.0227,
      "mean_token_accuracy": 0.0889634720981121,
      "num_tokens": 817692.0,
      "step": 445
    },
    {
      "entropy": 7.4513147354125975,
      "epoch": 0.3728251864125932,
      "grad_norm": 0.96875,
      "learning_rate": 0.0002245,
      "loss": 7.074,
      "mean_token_accuracy": 0.09340473040938377,
      "num_tokens": 827052.0,
      "step": 450
    },
    {
      "entropy": 7.493870782852173,
      "epoch": 0.37696768848384427,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00022700000000000002,
      "loss": 7.0443,
      "mean_token_accuracy": 0.09431338384747505,
      "num_tokens": 836442.0,
      "step": 455
    },
    {
      "entropy": 7.542994403839112,
      "epoch": 0.38111019055509526,
      "grad_norm": 1.03125,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.9475,
      "mean_token_accuracy": 0.08114949613809586,
      "num_tokens": 846441.0,
      "step": 460
    },
    {
      "entropy": 7.422606658935547,
      "epoch": 0.3852526926263463,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.9848,
      "mean_token_accuracy": 0.09037281647324562,
      "num_tokens": 856108.0,
      "step": 465
    },
    {
      "entropy": 7.409952163696289,
      "epoch": 0.38939519469759737,
      "grad_norm": 1.015625,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.9198,
      "mean_token_accuracy": 0.08507395833730698,
      "num_tokens": 865788.0,
      "step": 470
    },
    {
      "entropy": 7.413805341720581,
      "epoch": 0.39353769676884837,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000237,
      "loss": 6.9738,
      "mean_token_accuracy": 0.0836036428809166,
      "num_tokens": 874833.0,
      "step": 475
    },
    {
      "entropy": 7.285798835754394,
      "epoch": 0.3976801988400994,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002395,
      "loss": 6.9254,
      "mean_token_accuracy": 0.08790519461035728,
      "num_tokens": 883685.0,
      "step": 480
    },
    {
      "entropy": 7.440755939483642,
      "epoch": 0.40182270091135047,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000242,
      "loss": 6.9147,
      "mean_token_accuracy": 0.08789389207959175,
      "num_tokens": 893908.0,
      "step": 485
    },
    {
      "entropy": 7.403847885131836,
      "epoch": 0.40596520298260147,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002445,
      "loss": 7.0129,
      "mean_token_accuracy": 0.08363201320171357,
      "num_tokens": 903081.0,
      "step": 490
    },
    {
      "entropy": 7.289476585388184,
      "epoch": 0.4101077050538525,
      "grad_norm": 1.03125,
      "learning_rate": 0.000247,
      "loss": 6.8789,
      "mean_token_accuracy": 0.08956534340977669,
      "num_tokens": 912515.0,
      "step": 495
    },
    {
      "entropy": 7.4823840141296385,
      "epoch": 0.4142502071251036,
      "grad_norm": 0.90625,
      "learning_rate": 0.0002495,
      "loss": 6.8952,
      "mean_token_accuracy": 0.09587915241718292,
      "num_tokens": 921056.0,
      "step": 500
    },
    {
      "epoch": 0.4142502071251036,
      "eval_entropy": 7.1376285746153485,
      "eval_loss": 7.034562110900879,
      "eval_mean_token_accuracy": 0.08812946460839273,
      "eval_num_tokens": 921056.0,
      "eval_runtime": 2.122,
      "eval_samples_per_second": 1765.767,
      "eval_steps_per_second": 221.015,
      "step": 500
    },
    {
      "entropy": 7.1874003410339355,
      "epoch": 0.4183927091963546,
      "grad_norm": 1.0625,
      "learning_rate": 0.000252,
      "loss": 6.8368,
      "mean_token_accuracy": 0.09126921370625496,
      "num_tokens": 931112.0,
      "step": 505
    },
    {
      "entropy": 7.335116291046143,
      "epoch": 0.4225352112676056,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0002545,
      "loss": 6.9556,
      "mean_token_accuracy": 0.08759467005729675,
      "num_tokens": 941120.0,
      "step": 510
    },
    {
      "entropy": 7.3472545623779295,
      "epoch": 0.4266777133388567,
      "grad_norm": 1.109375,
      "learning_rate": 0.000257,
      "loss": 6.9077,
      "mean_token_accuracy": 0.08984342738986015,
      "num_tokens": 950041.0,
      "step": 515
    },
    {
      "entropy": 7.321388053894043,
      "epoch": 0.43082021541010773,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002595,
      "loss": 6.8881,
      "mean_token_accuracy": 0.09532358199357986,
      "num_tokens": 958608.0,
      "step": 520
    },
    {
      "entropy": 7.433567953109741,
      "epoch": 0.43496271748135873,
      "grad_norm": 1.078125,
      "learning_rate": 0.000262,
      "loss": 6.994,
      "mean_token_accuracy": 0.08581585139036178,
      "num_tokens": 966901.0,
      "step": 525
    },
    {
      "entropy": 7.247237634658814,
      "epoch": 0.4391052195526098,
      "grad_norm": 1.03125,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.8413,
      "mean_token_accuracy": 0.09217345640063286,
      "num_tokens": 976198.0,
      "step": 530
    },
    {
      "entropy": 7.419485569000244,
      "epoch": 0.44324772162386084,
      "grad_norm": 1.078125,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.8981,
      "mean_token_accuracy": 0.0871151939034462,
      "num_tokens": 985248.0,
      "step": 535
    },
    {
      "entropy": 7.186427021026612,
      "epoch": 0.44739022369511183,
      "grad_norm": 1.0625,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.8256,
      "mean_token_accuracy": 0.09469342082738877,
      "num_tokens": 993289.0,
      "step": 540
    },
    {
      "entropy": 7.353890085220337,
      "epoch": 0.4515327257663629,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.8839,
      "mean_token_accuracy": 0.09135557413101196,
      "num_tokens": 1001784.0,
      "step": 545
    },
    {
      "entropy": 7.35116548538208,
      "epoch": 0.45567522783761394,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002745,
      "loss": 6.8471,
      "mean_token_accuracy": 0.09857566952705384,
      "num_tokens": 1010445.0,
      "step": 550
    },
    {
      "entropy": 7.249645519256592,
      "epoch": 0.45981772990886494,
      "grad_norm": 1.234375,
      "learning_rate": 0.000277,
      "loss": 6.8867,
      "mean_token_accuracy": 0.09312620386481285,
      "num_tokens": 1019084.0,
      "step": 555
    },
    {
      "entropy": 7.222869443893432,
      "epoch": 0.463960231980116,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002795,
      "loss": 6.8057,
      "mean_token_accuracy": 0.09473200142383575,
      "num_tokens": 1027263.0,
      "step": 560
    },
    {
      "entropy": 7.348370885848999,
      "epoch": 0.46810273405136704,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.91,
      "mean_token_accuracy": 0.08753768503665924,
      "num_tokens": 1036275.0,
      "step": 565
    },
    {
      "entropy": 7.228622531890869,
      "epoch": 0.47224523612261804,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002845,
      "loss": 6.952,
      "mean_token_accuracy": 0.08856018707156181,
      "num_tokens": 1045208.0,
      "step": 570
    },
    {
      "entropy": 7.336494016647339,
      "epoch": 0.4763877381938691,
      "grad_norm": 1.125,
      "learning_rate": 0.000287,
      "loss": 6.8686,
      "mean_token_accuracy": 0.09350670352578164,
      "num_tokens": 1054703.0,
      "step": 575
    },
    {
      "entropy": 7.101537704467773,
      "epoch": 0.48053024026512015,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002895,
      "loss": 6.7426,
      "mean_token_accuracy": 0.09196772649884224,
      "num_tokens": 1063652.0,
      "step": 580
    },
    {
      "entropy": 7.175399208068848,
      "epoch": 0.48467274233637114,
      "grad_norm": 1.015625,
      "learning_rate": 0.000292,
      "loss": 6.8364,
      "mean_token_accuracy": 0.08494803085923194,
      "num_tokens": 1072070.0,
      "step": 585
    },
    {
      "entropy": 7.374863576889038,
      "epoch": 0.4888152444076222,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002945,
      "loss": 6.8737,
      "mean_token_accuracy": 0.09276422187685966,
      "num_tokens": 1079992.0,
      "step": 590
    },
    {
      "entropy": 7.18324236869812,
      "epoch": 0.49295774647887325,
      "grad_norm": 1.09375,
      "learning_rate": 0.000297,
      "loss": 6.8582,
      "mean_token_accuracy": 0.09373369887471199,
      "num_tokens": 1088862.0,
      "step": 595
    },
    {
      "entropy": 7.20685133934021,
      "epoch": 0.4971002485501243,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002995,
      "loss": 6.6698,
      "mean_token_accuracy": 0.1027169682085514,
      "num_tokens": 1097614.0,
      "step": 600
    },
    {
      "entropy": 7.092188501358033,
      "epoch": 0.5012427506213754,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000302,
      "loss": 6.8528,
      "mean_token_accuracy": 0.0964401125907898,
      "num_tokens": 1107347.0,
      "step": 605
    },
    {
      "entropy": 7.277201890945435,
      "epoch": 0.5053852526926264,
      "grad_norm": 1.125,
      "learning_rate": 0.0003045,
      "loss": 6.8723,
      "mean_token_accuracy": 0.09177736937999725,
      "num_tokens": 1116707.0,
      "step": 610
    },
    {
      "entropy": 7.224560880661011,
      "epoch": 0.5095277547638773,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000307,
      "loss": 6.8637,
      "mean_token_accuracy": 0.09503144547343254,
      "num_tokens": 1126356.0,
      "step": 615
    },
    {
      "entropy": 7.175086212158203,
      "epoch": 0.5136702568351285,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003095,
      "loss": 6.8186,
      "mean_token_accuracy": 0.0897763006389141,
      "num_tokens": 1136281.0,
      "step": 620
    },
    {
      "entropy": 7.1897046089172365,
      "epoch": 0.5178127589063795,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000312,
      "loss": 6.7722,
      "mean_token_accuracy": 0.09368982836604119,
      "num_tokens": 1145466.0,
      "step": 625
    },
    {
      "entropy": 7.105846500396728,
      "epoch": 0.5219552609776305,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003145,
      "loss": 6.7725,
      "mean_token_accuracy": 0.09914444759488106,
      "num_tokens": 1154225.0,
      "step": 630
    },
    {
      "entropy": 7.1858007431030275,
      "epoch": 0.5260977630488816,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000317,
      "loss": 6.8155,
      "mean_token_accuracy": 0.0893120139837265,
      "num_tokens": 1163872.0,
      "step": 635
    },
    {
      "entropy": 7.202767562866211,
      "epoch": 0.5302402651201326,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003195,
      "loss": 6.7379,
      "mean_token_accuracy": 0.09808101058006287,
      "num_tokens": 1172695.0,
      "step": 640
    },
    {
      "entropy": 7.032454538345337,
      "epoch": 0.5343827671913836,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000322,
      "loss": 6.6203,
      "mean_token_accuracy": 0.10113452076911926,
      "num_tokens": 1181291.0,
      "step": 645
    },
    {
      "entropy": 7.1764801979064945,
      "epoch": 0.5385252692626347,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.788,
      "mean_token_accuracy": 0.09804049134254456,
      "num_tokens": 1189927.0,
      "step": 650
    },
    {
      "entropy": 7.1191915512084964,
      "epoch": 0.5426677713338857,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.7234,
      "mean_token_accuracy": 0.10310481786727906,
      "num_tokens": 1197913.0,
      "step": 655
    },
    {
      "entropy": 7.032878971099853,
      "epoch": 0.5468102734051367,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.6523,
      "mean_token_accuracy": 0.104758819937706,
      "num_tokens": 1207032.0,
      "step": 660
    },
    {
      "entropy": 7.1075170040130615,
      "epoch": 0.5509527754763878,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.7536,
      "mean_token_accuracy": 0.09772960171103477,
      "num_tokens": 1215710.0,
      "step": 665
    },
    {
      "entropy": 7.11876540184021,
      "epoch": 0.5550952775476388,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.7292,
      "mean_token_accuracy": 0.0940351977944374,
      "num_tokens": 1225619.0,
      "step": 670
    },
    {
      "entropy": 7.023571300506592,
      "epoch": 0.5592377796188898,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000337,
      "loss": 6.6637,
      "mean_token_accuracy": 0.09711888134479522,
      "num_tokens": 1235000.0,
      "step": 675
    },
    {
      "entropy": 6.982565116882324,
      "epoch": 0.5633802816901409,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003395,
      "loss": 6.664,
      "mean_token_accuracy": 0.098265540599823,
      "num_tokens": 1243594.0,
      "step": 680
    },
    {
      "entropy": 7.147447633743286,
      "epoch": 0.5675227837613919,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000342,
      "loss": 6.8558,
      "mean_token_accuracy": 0.09704434052109719,
      "num_tokens": 1252279.0,
      "step": 685
    },
    {
      "entropy": 7.213455104827881,
      "epoch": 0.5716652858326429,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.7541,
      "mean_token_accuracy": 0.09795698821544647,
      "num_tokens": 1261314.0,
      "step": 690
    },
    {
      "entropy": 6.968819999694825,
      "epoch": 0.575807787903894,
      "grad_norm": 0.90625,
      "learning_rate": 0.000347,
      "loss": 6.6257,
      "mean_token_accuracy": 0.10265169814229011,
      "num_tokens": 1270854.0,
      "step": 695
    },
    {
      "entropy": 7.012129592895508,
      "epoch": 0.579950289975145,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003495,
      "loss": 6.6127,
      "mean_token_accuracy": 0.10243444740772248,
      "num_tokens": 1279183.0,
      "step": 700
    },
    {
      "entropy": 7.013506698608398,
      "epoch": 0.584092792046396,
      "grad_norm": 1.0,
      "learning_rate": 0.000352,
      "loss": 6.6971,
      "mean_token_accuracy": 0.09987314119935035,
      "num_tokens": 1288347.0,
      "step": 705
    },
    {
      "entropy": 7.083859825134278,
      "epoch": 0.5882352941176471,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003545,
      "loss": 6.7262,
      "mean_token_accuracy": 0.09877969771623611,
      "num_tokens": 1297449.0,
      "step": 710
    },
    {
      "entropy": 7.100916719436645,
      "epoch": 0.5923777961888981,
      "grad_norm": 1.1875,
      "learning_rate": 0.000357,
      "loss": 6.6759,
      "mean_token_accuracy": 0.10137342810630798,
      "num_tokens": 1306232.0,
      "step": 715
    },
    {
      "entropy": 6.998146438598633,
      "epoch": 0.5965202982601492,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003595,
      "loss": 6.6866,
      "mean_token_accuracy": 0.10439095944166184,
      "num_tokens": 1315073.0,
      "step": 720
    },
    {
      "entropy": 7.042725706100464,
      "epoch": 0.6006628003314002,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000362,
      "loss": 6.7278,
      "mean_token_accuracy": 0.09690923467278481,
      "num_tokens": 1324629.0,
      "step": 725
    },
    {
      "entropy": 7.049885368347168,
      "epoch": 0.6048053024026512,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003645,
      "loss": 6.8315,
      "mean_token_accuracy": 0.09335754811763763,
      "num_tokens": 1334152.0,
      "step": 730
    },
    {
      "entropy": 7.204215431213379,
      "epoch": 0.6089478044739023,
      "grad_norm": 1.015625,
      "learning_rate": 0.000367,
      "loss": 6.8026,
      "mean_token_accuracy": 0.0956441655755043,
      "num_tokens": 1343063.0,
      "step": 735
    },
    {
      "entropy": 7.062389278411866,
      "epoch": 0.6130903065451533,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003695,
      "loss": 6.722,
      "mean_token_accuracy": 0.09436356574296952,
      "num_tokens": 1352632.0,
      "step": 740
    },
    {
      "entropy": 6.928182935714721,
      "epoch": 0.6172328086164043,
      "grad_norm": 1.15625,
      "learning_rate": 0.000372,
      "loss": 6.6537,
      "mean_token_accuracy": 0.09909828305244446,
      "num_tokens": 1361103.0,
      "step": 745
    },
    {
      "entropy": 7.017486238479615,
      "epoch": 0.6213753106876554,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003745,
      "loss": 6.6427,
      "mean_token_accuracy": 0.10182388573884964,
      "num_tokens": 1370658.0,
      "step": 750
    },
    {
      "entropy": 7.007332801818848,
      "epoch": 0.6255178127589064,
      "grad_norm": 0.953125,
      "learning_rate": 0.000377,
      "loss": 6.6679,
      "mean_token_accuracy": 0.09760173633694649,
      "num_tokens": 1379817.0,
      "step": 755
    },
    {
      "entropy": 7.043180799484253,
      "epoch": 0.6296603148301574,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003795,
      "loss": 6.6189,
      "mean_token_accuracy": 0.10276637375354766,
      "num_tokens": 1389329.0,
      "step": 760
    },
    {
      "entropy": 6.83769268989563,
      "epoch": 0.6338028169014085,
      "grad_norm": 1.0625,
      "learning_rate": 0.000382,
      "loss": 6.6539,
      "mean_token_accuracy": 0.10125284790992736,
      "num_tokens": 1399279.0,
      "step": 765
    },
    {
      "entropy": 6.993398857116699,
      "epoch": 0.6379453189726595,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003845,
      "loss": 6.6017,
      "mean_token_accuracy": 0.10439575612545013,
      "num_tokens": 1408765.0,
      "step": 770
    },
    {
      "entropy": 7.039279270172119,
      "epoch": 0.6420878210439105,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.7302,
      "mean_token_accuracy": 0.10342611595988274,
      "num_tokens": 1418615.0,
      "step": 775
    },
    {
      "entropy": 6.924956750869751,
      "epoch": 0.6462303231151616,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.6327,
      "mean_token_accuracy": 0.09766181856393814,
      "num_tokens": 1427359.0,
      "step": 780
    },
    {
      "entropy": 7.031212186813354,
      "epoch": 0.6503728251864126,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.7909,
      "mean_token_accuracy": 0.10112168565392494,
      "num_tokens": 1435516.0,
      "step": 785
    },
    {
      "entropy": 7.065572786331177,
      "epoch": 0.6545153272576636,
      "grad_norm": 1.125,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.7387,
      "mean_token_accuracy": 0.09975770190358162,
      "num_tokens": 1444438.0,
      "step": 790
    },
    {
      "entropy": 7.0253180980682375,
      "epoch": 0.6586578293289147,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.651,
      "mean_token_accuracy": 0.10093981400132179,
      "num_tokens": 1453740.0,
      "step": 795
    },
    {
      "entropy": 7.016492176055908,
      "epoch": 0.6628003314001657,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003995,
      "loss": 6.6457,
      "mean_token_accuracy": 0.10038311630487443,
      "num_tokens": 1462083.0,
      "step": 800
    },
    {
      "entropy": 6.8830955028533936,
      "epoch": 0.6669428334714167,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000402,
      "loss": 6.5764,
      "mean_token_accuracy": 0.10810273066163063,
      "num_tokens": 1471459.0,
      "step": 805
    },
    {
      "entropy": 6.919638729095459,
      "epoch": 0.6710853355426678,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004045,
      "loss": 6.5688,
      "mean_token_accuracy": 0.1042623020708561,
      "num_tokens": 1480516.0,
      "step": 810
    },
    {
      "entropy": 6.843572092056275,
      "epoch": 0.6752278376139188,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.4732,
      "mean_token_accuracy": 0.1075613982975483,
      "num_tokens": 1489268.0,
      "step": 815
    },
    {
      "entropy": 7.020244693756103,
      "epoch": 0.6793703396851698,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004095,
      "loss": 6.6765,
      "mean_token_accuracy": 0.10138711780309677,
      "num_tokens": 1498819.0,
      "step": 820
    },
    {
      "entropy": 6.910724115371704,
      "epoch": 0.6835128417564209,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000412,
      "loss": 6.6586,
      "mean_token_accuracy": 0.10495447292923928,
      "num_tokens": 1507852.0,
      "step": 825
    },
    {
      "entropy": 6.989867115020752,
      "epoch": 0.6876553438276719,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004145,
      "loss": 6.6807,
      "mean_token_accuracy": 0.10375865250825882,
      "num_tokens": 1517538.0,
      "step": 830
    },
    {
      "entropy": 6.873890399932861,
      "epoch": 0.6917978458989229,
      "grad_norm": 1.296875,
      "learning_rate": 0.000417,
      "loss": 6.563,
      "mean_token_accuracy": 0.1091550700366497,
      "num_tokens": 1525389.0,
      "step": 835
    },
    {
      "entropy": 6.926404094696045,
      "epoch": 0.695940347970174,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004195,
      "loss": 6.5379,
      "mean_token_accuracy": 0.10326163768768311,
      "num_tokens": 1534117.0,
      "step": 840
    },
    {
      "entropy": 6.861436605453491,
      "epoch": 0.700082850041425,
      "grad_norm": 1.140625,
      "learning_rate": 0.000422,
      "loss": 6.576,
      "mean_token_accuracy": 0.10448229536414147,
      "num_tokens": 1542610.0,
      "step": 845
    },
    {
      "entropy": 6.804575872421265,
      "epoch": 0.704225352112676,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004245,
      "loss": 6.5549,
      "mean_token_accuracy": 0.11098664253950119,
      "num_tokens": 1551771.0,
      "step": 850
    },
    {
      "entropy": 7.011364126205445,
      "epoch": 0.7083678541839271,
      "grad_norm": 1.203125,
      "learning_rate": 0.000427,
      "loss": 6.734,
      "mean_token_accuracy": 0.09743175655603409,
      "num_tokens": 1560231.0,
      "step": 855
    },
    {
      "entropy": 6.8614123344421385,
      "epoch": 0.7125103562551781,
      "grad_norm": 1.0,
      "learning_rate": 0.0004295,
      "loss": 6.6704,
      "mean_token_accuracy": 0.10075936913490295,
      "num_tokens": 1569578.0,
      "step": 860
    },
    {
      "entropy": 6.939766216278076,
      "epoch": 0.7166528583264291,
      "grad_norm": 1.03125,
      "learning_rate": 0.000432,
      "loss": 6.5619,
      "mean_token_accuracy": 0.10681614726781845,
      "num_tokens": 1579308.0,
      "step": 865
    },
    {
      "entropy": 6.910393333435058,
      "epoch": 0.7207953603976802,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004345,
      "loss": 6.7379,
      "mean_token_accuracy": 0.10018669590353965,
      "num_tokens": 1589506.0,
      "step": 870
    },
    {
      "entropy": 6.811976051330566,
      "epoch": 0.7249378624689312,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000437,
      "loss": 6.5377,
      "mean_token_accuracy": 0.10628587827086448,
      "num_tokens": 1598127.0,
      "step": 875
    },
    {
      "entropy": 6.834736251831055,
      "epoch": 0.7290803645401823,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004395,
      "loss": 6.549,
      "mean_token_accuracy": 0.10746697857975959,
      "num_tokens": 1608073.0,
      "step": 880
    },
    {
      "entropy": 6.892040348052978,
      "epoch": 0.7332228666114333,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000442,
      "loss": 6.5821,
      "mean_token_accuracy": 0.10727652609348297,
      "num_tokens": 1616272.0,
      "step": 885
    },
    {
      "entropy": 6.854152059555053,
      "epoch": 0.7373653686826843,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004445,
      "loss": 6.5853,
      "mean_token_accuracy": 0.10497673824429513,
      "num_tokens": 1625577.0,
      "step": 890
    },
    {
      "entropy": 6.974736309051513,
      "epoch": 0.7415078707539354,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000447,
      "loss": 6.6705,
      "mean_token_accuracy": 0.09881450086832047,
      "num_tokens": 1634902.0,
      "step": 895
    },
    {
      "entropy": 6.89474573135376,
      "epoch": 0.7456503728251864,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.5895,
      "mean_token_accuracy": 0.1074444942176342,
      "num_tokens": 1644131.0,
      "step": 900
    },
    {
      "entropy": 6.79931492805481,
      "epoch": 0.7497928748964374,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.5106,
      "mean_token_accuracy": 0.11076099425554276,
      "num_tokens": 1652272.0,
      "step": 905
    },
    {
      "entropy": 6.829667091369629,
      "epoch": 0.7539353769676885,
      "grad_norm": 1.125,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.5935,
      "mean_token_accuracy": 0.10850171595811844,
      "num_tokens": 1660962.0,
      "step": 910
    },
    {
      "entropy": 6.8230170726776125,
      "epoch": 0.7580778790389395,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.5328,
      "mean_token_accuracy": 0.10780560970306396,
      "num_tokens": 1670570.0,
      "step": 915
    },
    {
      "entropy": 6.78682131767273,
      "epoch": 0.7622203811101905,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.5977,
      "mean_token_accuracy": 0.10254386737942696,
      "num_tokens": 1679674.0,
      "step": 920
    },
    {
      "entropy": 6.933328247070312,
      "epoch": 0.7663628831814416,
      "grad_norm": 1.125,
      "learning_rate": 0.000462,
      "loss": 6.5662,
      "mean_token_accuracy": 0.10486595705151558,
      "num_tokens": 1688849.0,
      "step": 925
    },
    {
      "entropy": 6.694953775405883,
      "epoch": 0.7705053852526926,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004645,
      "loss": 6.4459,
      "mean_token_accuracy": 0.10817986652255059,
      "num_tokens": 1698565.0,
      "step": 930
    },
    {
      "entropy": 6.862428140640259,
      "epoch": 0.7746478873239436,
      "grad_norm": 1.09375,
      "learning_rate": 0.000467,
      "loss": 6.5348,
      "mean_token_accuracy": 0.11153487414121628,
      "num_tokens": 1708009.0,
      "step": 935
    },
    {
      "entropy": 6.8652912139892575,
      "epoch": 0.7787903893951947,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004695,
      "loss": 6.544,
      "mean_token_accuracy": 0.10657664686441422,
      "num_tokens": 1716497.0,
      "step": 940
    },
    {
      "entropy": 6.7234900951385494,
      "epoch": 0.7829328914664457,
      "grad_norm": 1.0625,
      "learning_rate": 0.000472,
      "loss": 6.5287,
      "mean_token_accuracy": 0.11238738149404526,
      "num_tokens": 1725609.0,
      "step": 945
    },
    {
      "entropy": 6.804434871673584,
      "epoch": 0.7870753935376967,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004745,
      "loss": 6.4561,
      "mean_token_accuracy": 0.11403901427984238,
      "num_tokens": 1734165.0,
      "step": 950
    },
    {
      "entropy": 6.788070774078369,
      "epoch": 0.7912178956089478,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000477,
      "loss": 6.58,
      "mean_token_accuracy": 0.10270701348781586,
      "num_tokens": 1744856.0,
      "step": 955
    },
    {
      "entropy": 6.728517675399781,
      "epoch": 0.7953603976801988,
      "grad_norm": 1.125,
      "learning_rate": 0.0004795,
      "loss": 6.5771,
      "mean_token_accuracy": 0.10236030369997025,
      "num_tokens": 1753386.0,
      "step": 960
    },
    {
      "entropy": 6.854466104507447,
      "epoch": 0.7995028997514498,
      "grad_norm": 1.078125,
      "learning_rate": 0.000482,
      "loss": 6.4332,
      "mean_token_accuracy": 0.11357746422290801,
      "num_tokens": 1762521.0,
      "step": 965
    },
    {
      "entropy": 6.700516653060913,
      "epoch": 0.8036454018227009,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004845,
      "loss": 6.4878,
      "mean_token_accuracy": 0.1087129294872284,
      "num_tokens": 1772403.0,
      "step": 970
    },
    {
      "entropy": 6.739825487136841,
      "epoch": 0.8077879038939519,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000487,
      "loss": 6.3968,
      "mean_token_accuracy": 0.11042674705386162,
      "num_tokens": 1780924.0,
      "step": 975
    },
    {
      "entropy": 6.788566207885742,
      "epoch": 0.8119304059652029,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004895,
      "loss": 6.5364,
      "mean_token_accuracy": 0.10567695051431655,
      "num_tokens": 1789992.0,
      "step": 980
    },
    {
      "entropy": 6.741111326217651,
      "epoch": 0.816072908036454,
      "grad_norm": 1.1875,
      "learning_rate": 0.000492,
      "loss": 6.5032,
      "mean_token_accuracy": 0.11023874580860138,
      "num_tokens": 1799513.0,
      "step": 985
    },
    {
      "entropy": 6.761337041854858,
      "epoch": 0.820215410107705,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004945,
      "loss": 6.3597,
      "mean_token_accuracy": 0.11022111549973487,
      "num_tokens": 1807393.0,
      "step": 990
    },
    {
      "entropy": 6.774386739730835,
      "epoch": 0.824357912178956,
      "grad_norm": 1.15625,
      "learning_rate": 0.000497,
      "loss": 6.6022,
      "mean_token_accuracy": 0.10382014736533166,
      "num_tokens": 1817284.0,
      "step": 995
    },
    {
      "entropy": 6.767058658599853,
      "epoch": 0.8285004142502072,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004995,
      "loss": 6.375,
      "mean_token_accuracy": 0.1134818322956562,
      "num_tokens": 1825591.0,
      "step": 1000
    },
    {
      "epoch": 0.8285004142502072,
      "eval_entropy": 6.4873610498554415,
      "eval_loss": 6.588028907775879,
      "eval_mean_token_accuracy": 0.11022937250162747,
      "eval_num_tokens": 1825591.0,
      "eval_runtime": 2.1242,
      "eval_samples_per_second": 1763.952,
      "eval_steps_per_second": 220.788,
      "step": 1000
    },
    {
      "entropy": 6.784157371520996,
      "epoch": 0.8326429163214581,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004999998547681232,
      "loss": 6.5023,
      "mean_token_accuracy": 0.11439347863197327,
      "num_tokens": 1834655.0,
      "step": 1005
    },
    {
      "entropy": 6.565783596038818,
      "epoch": 0.8367854183927091,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999992647639447,
      "loss": 6.3503,
      "mean_token_accuracy": 0.1175505943596363,
      "num_tokens": 1844187.0,
      "step": 1010
    },
    {
      "entropy": 6.671072387695313,
      "epoch": 0.8409279204639603,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999982209116618,
      "loss": 6.4409,
      "mean_token_accuracy": 0.10719897747039794,
      "num_tokens": 1852768.0,
      "step": 1015
    },
    {
      "entropy": 6.695590972900391,
      "epoch": 0.8450704225352113,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999967232133796,
      "loss": 6.4767,
      "mean_token_accuracy": 0.1183364823460579,
      "num_tokens": 1862095.0,
      "step": 1020
    },
    {
      "entropy": 6.7737802028656,
      "epoch": 0.8492129246064622,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999947716721196,
      "loss": 6.5222,
      "mean_token_accuracy": 0.10554626062512398,
      "num_tokens": 1871239.0,
      "step": 1025
    },
    {
      "entropy": 6.712663888931274,
      "epoch": 0.8533554266777134,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999923662918179,
      "loss": 6.515,
      "mean_token_accuracy": 0.11144773066043853,
      "num_tokens": 1880065.0,
      "step": 1030
    },
    {
      "entropy": 6.769164752960205,
      "epoch": 0.8574979287489644,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999895070773266,
      "loss": 6.5543,
      "mean_token_accuracy": 0.10713088288903236,
      "num_tokens": 1889065.0,
      "step": 1035
    },
    {
      "entropy": 6.731873178482056,
      "epoch": 0.8616404308202155,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999861940344129,
      "loss": 6.5513,
      "mean_token_accuracy": 0.1086876131594181,
      "num_tokens": 1898238.0,
      "step": 1040
    },
    {
      "entropy": 6.776325654983521,
      "epoch": 0.8657829328914665,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999824271697597,
      "loss": 6.5038,
      "mean_token_accuracy": 0.1181743711233139,
      "num_tokens": 1906780.0,
      "step": 1045
    },
    {
      "entropy": 6.801963806152344,
      "epoch": 0.8699254349627175,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499978206490965,
      "loss": 6.5841,
      "mean_token_accuracy": 0.10914266481995583,
      "num_tokens": 1916183.0,
      "step": 1050
    },
    {
      "entropy": 6.635178756713867,
      "epoch": 0.8740679370339686,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999735320065427,
      "loss": 6.4376,
      "mean_token_accuracy": 0.11982565298676491,
      "num_tokens": 1925382.0,
      "step": 1055
    },
    {
      "entropy": 6.6497776985168455,
      "epoch": 0.8782104391052196,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999684037259216,
      "loss": 6.2866,
      "mean_token_accuracy": 0.1231508955359459,
      "num_tokens": 1933639.0,
      "step": 1060
    },
    {
      "entropy": 6.584863090515137,
      "epoch": 0.8823529411764706,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999628216594456,
      "loss": 6.5596,
      "mean_token_accuracy": 0.10988020077347756,
      "num_tokens": 1943044.0,
      "step": 1065
    },
    {
      "entropy": 6.758437156677246,
      "epoch": 0.8864954432477217,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499956785818375,
      "loss": 6.4002,
      "mean_token_accuracy": 0.11250741928815841,
      "num_tokens": 1951777.0,
      "step": 1070
    },
    {
      "entropy": 6.580097198486328,
      "epoch": 0.8906379453189727,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999502962148843,
      "loss": 6.4733,
      "mean_token_accuracy": 0.1164925940334797,
      "num_tokens": 1961356.0,
      "step": 1075
    },
    {
      "entropy": 6.8584808826446535,
      "epoch": 0.8947804473902237,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499943352862064,
      "loss": 6.5297,
      "mean_token_accuracy": 0.11108277216553689,
      "num_tokens": 1970341.0,
      "step": 1080
    },
    {
      "entropy": 6.662361335754395,
      "epoch": 0.8989229494614748,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999359557739194,
      "loss": 6.3848,
      "mean_token_accuracy": 0.1149318166077137,
      "num_tokens": 1978576.0,
      "step": 1085
    },
    {
      "entropy": 6.662960624694824,
      "epoch": 0.9030654515327258,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999281049653713,
      "loss": 6.4318,
      "mean_token_accuracy": 0.11755185425281525,
      "num_tokens": 1987229.0,
      "step": 1090
    },
    {
      "entropy": 6.685955619812011,
      "epoch": 0.9072079536039768,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999198004522557,
      "loss": 6.3431,
      "mean_token_accuracy": 0.124117149412632,
      "num_tokens": 1996473.0,
      "step": 1095
    },
    {
      "entropy": 6.601357793807983,
      "epoch": 0.9113504556752279,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999110422513237,
      "loss": 6.3803,
      "mean_token_accuracy": 0.11419567912817001,
      "num_tokens": 2005551.0,
      "step": 1100
    },
    {
      "entropy": 6.621028995513916,
      "epoch": 0.9154929577464789,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999018303802414,
      "loss": 6.4823,
      "mean_token_accuracy": 0.11499169245362281,
      "num_tokens": 2015012.0,
      "step": 1105
    },
    {
      "entropy": 6.696759366989136,
      "epoch": 0.9196354598177299,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998921648575904,
      "loss": 6.4869,
      "mean_token_accuracy": 0.11509674936532974,
      "num_tokens": 2024435.0,
      "step": 1110
    },
    {
      "entropy": 6.658921909332276,
      "epoch": 0.923777961888981,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998820457028669,
      "loss": 6.39,
      "mean_token_accuracy": 0.11696809232234955,
      "num_tokens": 2033991.0,
      "step": 1115
    },
    {
      "entropy": 6.638371849060059,
      "epoch": 0.927920463960232,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998714729364824,
      "loss": 6.3985,
      "mean_token_accuracy": 0.12039039358496666,
      "num_tokens": 2043117.0,
      "step": 1120
    },
    {
      "entropy": 6.695077419281006,
      "epoch": 0.932062966031483,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998604465797634,
      "loss": 6.4111,
      "mean_token_accuracy": 0.11857732310891152,
      "num_tokens": 2052304.0,
      "step": 1125
    },
    {
      "entropy": 6.52111029624939,
      "epoch": 0.9362054681027341,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998489666549512,
      "loss": 6.3676,
      "mean_token_accuracy": 0.12015296593308449,
      "num_tokens": 2061217.0,
      "step": 1130
    },
    {
      "entropy": 6.554986667633057,
      "epoch": 0.9403479701739851,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998370331852021,
      "loss": 6.3543,
      "mean_token_accuracy": 0.11446144282817841,
      "num_tokens": 2070796.0,
      "step": 1135
    },
    {
      "entropy": 6.708700561523438,
      "epoch": 0.9444904722452361,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998246461945872,
      "loss": 6.4765,
      "mean_token_accuracy": 0.11509892418980598,
      "num_tokens": 2080202.0,
      "step": 1140
    },
    {
      "entropy": 6.634687614440918,
      "epoch": 0.9486329743164872,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998118057080923,
      "loss": 6.393,
      "mean_token_accuracy": 0.11585810706019402,
      "num_tokens": 2089788.0,
      "step": 1145
    },
    {
      "entropy": 6.633106660842896,
      "epoch": 0.9527754763877382,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997985117516183,
      "loss": 6.4111,
      "mean_token_accuracy": 0.11044302135705948,
      "num_tokens": 2098727.0,
      "step": 1150
    },
    {
      "entropy": 6.56185474395752,
      "epoch": 0.9569179784589892,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004997847643519803,
      "loss": 6.3168,
      "mean_token_accuracy": 0.11371051520109177,
      "num_tokens": 2107645.0,
      "step": 1155
    },
    {
      "entropy": 6.496697521209716,
      "epoch": 0.9610604805302403,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997705635369085,
      "loss": 6.299,
      "mean_token_accuracy": 0.11548860892653465,
      "num_tokens": 2116613.0,
      "step": 1160
    },
    {
      "entropy": 6.596785545349121,
      "epoch": 0.9652029826014913,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997559093350474,
      "loss": 6.2959,
      "mean_token_accuracy": 0.11894101276993752,
      "num_tokens": 2125585.0,
      "step": 1165
    },
    {
      "entropy": 6.679542684555054,
      "epoch": 0.9693454846727423,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997408017759562,
      "loss": 6.4112,
      "mean_token_accuracy": 0.11748032122850419,
      "num_tokens": 2134795.0,
      "step": 1170
    },
    {
      "entropy": 6.610104560852051,
      "epoch": 0.9734879867439934,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997252408901084,
      "loss": 6.4127,
      "mean_token_accuracy": 0.11549609526991844,
      "num_tokens": 2143699.0,
      "step": 1175
    },
    {
      "entropy": 6.6923322677612305,
      "epoch": 0.9776304888152444,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997092267088922,
      "loss": 6.359,
      "mean_token_accuracy": 0.11639811396598816,
      "num_tokens": 2152187.0,
      "step": 1180
    },
    {
      "entropy": 6.513570928573609,
      "epoch": 0.9817729908864954,
      "grad_norm": 1.125,
      "learning_rate": 0.0004996927592646097,
      "loss": 6.4218,
      "mean_token_accuracy": 0.11397558972239494,
      "num_tokens": 2160941.0,
      "step": 1185
    },
    {
      "entropy": 6.680602073669434,
      "epoch": 0.9859154929577465,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996758385904776,
      "loss": 6.4021,
      "mean_token_accuracy": 0.11822220534086228,
      "num_tokens": 2168946.0,
      "step": 1190
    },
    {
      "entropy": 6.4418701171875,
      "epoch": 0.9900579950289975,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996584647206269,
      "loss": 6.307,
      "mean_token_accuracy": 0.11671008840203285,
      "num_tokens": 2177662.0,
      "step": 1195
    },
    {
      "entropy": 6.607698106765747,
      "epoch": 0.9942004971002486,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996406376901026,
      "loss": 6.4069,
      "mean_token_accuracy": 0.12819217070937156,
      "num_tokens": 2187461.0,
      "step": 1200
    },
    {
      "entropy": 6.504300785064697,
      "epoch": 0.9983429991714996,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996223575348635,
      "loss": 6.2895,
      "mean_token_accuracy": 0.12742482796311377,
      "num_tokens": 2196878.0,
      "step": 1205
    },
    {
      "entropy": 6.635723876953125,
      "epoch": 1.0024855012427507,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996036242917828,
      "loss": 6.3007,
      "mean_token_accuracy": 0.11356158629059791,
      "num_tokens": 2206178.0,
      "step": 1210
    },
    {
      "entropy": 6.521615219116211,
      "epoch": 1.0066280033140016,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995844379986476,
      "loss": 6.1736,
      "mean_token_accuracy": 0.1223662681877613,
      "num_tokens": 2215330.0,
      "step": 1215
    },
    {
      "entropy": 6.593561172485352,
      "epoch": 1.0107705053852527,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995647986941587,
      "loss": 6.1712,
      "mean_token_accuracy": 0.12720630392432214,
      "num_tokens": 2224146.0,
      "step": 1220
    },
    {
      "entropy": 6.5283712387084964,
      "epoch": 1.0149130074565038,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995447064179308,
      "loss": 6.1551,
      "mean_token_accuracy": 0.13192698806524278,
      "num_tokens": 2233646.0,
      "step": 1225
    },
    {
      "entropy": 6.364728593826294,
      "epoch": 1.0190555095277547,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499524161210492,
      "loss": 6.0263,
      "mean_token_accuracy": 0.1275729440152645,
      "num_tokens": 2243663.0,
      "step": 1230
    },
    {
      "entropy": 6.589374303817749,
      "epoch": 1.0231980115990058,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995031631132844,
      "loss": 6.2014,
      "mean_token_accuracy": 0.13359731659293175,
      "num_tokens": 2252564.0,
      "step": 1235
    },
    {
      "entropy": 6.493811464309692,
      "epoch": 1.027340513670257,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004994817121686633,
      "loss": 6.2099,
      "mean_token_accuracy": 0.12453857064247131,
      "num_tokens": 2262504.0,
      "step": 1240
    },
    {
      "entropy": 6.53023624420166,
      "epoch": 1.0314830157415078,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004994598084198979,
      "loss": 6.0827,
      "mean_token_accuracy": 0.13783227279782295,
      "num_tokens": 2270553.0,
      "step": 1245
    },
    {
      "entropy": 6.445224046707153,
      "epoch": 1.035625517812759,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004994374519111701,
      "loss": 6.1134,
      "mean_token_accuracy": 0.12587740272283554,
      "num_tokens": 2279319.0,
      "step": 1250
    },
    {
      "entropy": 6.521056652069092,
      "epoch": 1.03976801988401,
      "grad_norm": 1.125,
      "learning_rate": 0.0004994146426875757,
      "loss": 6.2428,
      "mean_token_accuracy": 0.1287379577755928,
      "num_tokens": 2288749.0,
      "step": 1255
    },
    {
      "entropy": 6.496748733520508,
      "epoch": 1.043910521955261,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004993913807951232,
      "loss": 6.1241,
      "mean_token_accuracy": 0.12613506391644477,
      "num_tokens": 2298272.0,
      "step": 1260
    },
    {
      "entropy": 6.513681507110595,
      "epoch": 1.048053024026512,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993676662807345,
      "loss": 6.1752,
      "mean_token_accuracy": 0.13274361863732337,
      "num_tokens": 2307230.0,
      "step": 1265
    },
    {
      "entropy": 6.4921036720275875,
      "epoch": 1.0521955260977631,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004993434991922443,
      "loss": 6.251,
      "mean_token_accuracy": 0.12720409855246545,
      "num_tokens": 2316306.0,
      "step": 1270
    },
    {
      "entropy": 6.523912334442139,
      "epoch": 1.056338028169014,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993188795784002,
      "loss": 6.1805,
      "mean_token_accuracy": 0.1223751187324524,
      "num_tokens": 2325672.0,
      "step": 1275
    },
    {
      "entropy": 6.428217792510987,
      "epoch": 1.0604805302402651,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004992938074888628,
      "loss": 6.1681,
      "mean_token_accuracy": 0.1255595475435257,
      "num_tokens": 2335251.0,
      "step": 1280
    },
    {
      "entropy": 6.473890256881714,
      "epoch": 1.0646230323115162,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004992682829742052,
      "loss": 6.125,
      "mean_token_accuracy": 0.12794472500681878,
      "num_tokens": 2344054.0,
      "step": 1285
    },
    {
      "entropy": 6.3825085163116455,
      "epoch": 1.0687655343827671,
      "grad_norm": 1.125,
      "learning_rate": 0.0004992423060859132,
      "loss": 6.1274,
      "mean_token_accuracy": 0.12978773862123488,
      "num_tokens": 2353493.0,
      "step": 1290
    },
    {
      "entropy": 6.579702091217041,
      "epoch": 1.0729080364540182,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004992158768763849,
      "loss": 6.2352,
      "mean_token_accuracy": 0.12934877946972848,
      "num_tokens": 2362687.0,
      "step": 1295
    },
    {
      "entropy": 6.51726803779602,
      "epoch": 1.0770505385252693,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991889953989308,
      "loss": 6.2101,
      "mean_token_accuracy": 0.13052549585700035,
      "num_tokens": 2372325.0,
      "step": 1300
    },
    {
      "entropy": 6.475685262680054,
      "epoch": 1.0811930405965202,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004991616617077741,
      "loss": 6.2296,
      "mean_token_accuracy": 0.12784886062145234,
      "num_tokens": 2381987.0,
      "step": 1305
    },
    {
      "entropy": 6.639652824401855,
      "epoch": 1.0853355426677713,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004991338758580496,
      "loss": 6.1472,
      "mean_token_accuracy": 0.1295457474887371,
      "num_tokens": 2390901.0,
      "step": 1310
    },
    {
      "entropy": 6.358977031707764,
      "epoch": 1.0894780447390224,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991056379058045,
      "loss": 6.1883,
      "mean_token_accuracy": 0.12295814007520675,
      "num_tokens": 2400756.0,
      "step": 1315
    },
    {
      "entropy": 6.40067458152771,
      "epoch": 1.0936205468102733,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004990769479079976,
      "loss": 5.9995,
      "mean_token_accuracy": 0.13081807941198348,
      "num_tokens": 2409526.0,
      "step": 1320
    },
    {
      "entropy": 6.402806425094605,
      "epoch": 1.0977630488815244,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004990478059225003,
      "loss": 6.1774,
      "mean_token_accuracy": 0.12853156328201293,
      "num_tokens": 2419583.0,
      "step": 1325
    },
    {
      "entropy": 6.509050941467285,
      "epoch": 1.1019055509527755,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004990182120080945,
      "loss": 6.142,
      "mean_token_accuracy": 0.13302397802472116,
      "num_tokens": 2429439.0,
      "step": 1330
    },
    {
      "entropy": 6.362121295928955,
      "epoch": 1.1060480530240264,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004989881662244748,
      "loss": 6.128,
      "mean_token_accuracy": 0.12412297874689102,
      "num_tokens": 2438467.0,
      "step": 1335
    },
    {
      "entropy": 6.464797878265381,
      "epoch": 1.1101905550952775,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004989576686322468,
      "loss": 6.0819,
      "mean_token_accuracy": 0.13460188284516333,
      "num_tokens": 2446619.0,
      "step": 1340
    },
    {
      "entropy": 6.365456485748291,
      "epoch": 1.1143330571665286,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004989267192929272,
      "loss": 6.1521,
      "mean_token_accuracy": 0.12620313614606857,
      "num_tokens": 2454998.0,
      "step": 1345
    },
    {
      "entropy": 6.402712297439575,
      "epoch": 1.1184755592377795,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004988953182689445,
      "loss": 6.1088,
      "mean_token_accuracy": 0.13071864768862723,
      "num_tokens": 2463497.0,
      "step": 1350
    },
    {
      "entropy": 6.442966413497925,
      "epoch": 1.1226180613090306,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004988634656236378,
      "loss": 6.0379,
      "mean_token_accuracy": 0.1323768101632595,
      "num_tokens": 2471763.0,
      "step": 1355
    },
    {
      "entropy": 6.382192516326905,
      "epoch": 1.1267605633802817,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004988311614212576,
      "loss": 6.0982,
      "mean_token_accuracy": 0.13209657073020936,
      "num_tokens": 2479949.0,
      "step": 1360
    },
    {
      "entropy": 6.453052568435669,
      "epoch": 1.1309030654515326,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004987984057269648,
      "loss": 6.1482,
      "mean_token_accuracy": 0.1316144049167633,
      "num_tokens": 2489097.0,
      "step": 1365
    },
    {
      "entropy": 6.231172561645508,
      "epoch": 1.1350455675227837,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004987651986068311,
      "loss": 5.9739,
      "mean_token_accuracy": 0.13615623116493225,
      "num_tokens": 2497625.0,
      "step": 1370
    },
    {
      "entropy": 6.410304975509644,
      "epoch": 1.1391880695940348,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004987315401278394,
      "loss": 6.2206,
      "mean_token_accuracy": 0.12649596706032754,
      "num_tokens": 2507107.0,
      "step": 1375
    },
    {
      "entropy": 6.553795623779297,
      "epoch": 1.143330571665286,
      "grad_norm": 1.25,
      "learning_rate": 0.0004986974303578819,
      "loss": 6.2198,
      "mean_token_accuracy": 0.12320356070995331,
      "num_tokens": 2516629.0,
      "step": 1380
    },
    {
      "entropy": 6.4057948112487795,
      "epoch": 1.1474730737365368,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004986628693657621,
      "loss": 6.1043,
      "mean_token_accuracy": 0.13110435828566552,
      "num_tokens": 2525924.0,
      "step": 1385
    },
    {
      "entropy": 6.529660415649414,
      "epoch": 1.151615575807788,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004986278572211933,
      "loss": 6.2581,
      "mean_token_accuracy": 0.12878933995962144,
      "num_tokens": 2535586.0,
      "step": 1390
    },
    {
      "entropy": 6.49504132270813,
      "epoch": 1.1557580778790388,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004985923939947987,
      "loss": 6.3228,
      "mean_token_accuracy": 0.12478614002466201,
      "num_tokens": 2546685.0,
      "step": 1395
    },
    {
      "entropy": 6.359566736221313,
      "epoch": 1.15990057995029,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004985564797581115,
      "loss": 6.0478,
      "mean_token_accuracy": 0.13317504152655602,
      "num_tokens": 2556873.0,
      "step": 1400
    },
    {
      "entropy": 6.434709787368774,
      "epoch": 1.164043082021541,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004985201145835747,
      "loss": 6.0913,
      "mean_token_accuracy": 0.13183084428310393,
      "num_tokens": 2565222.0,
      "step": 1405
    },
    {
      "entropy": 6.3213330745697025,
      "epoch": 1.1681855840927922,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000498483298544541,
      "loss": 6.0768,
      "mean_token_accuracy": 0.12748995423316956,
      "num_tokens": 2575467.0,
      "step": 1410
    },
    {
      "entropy": 6.398600482940674,
      "epoch": 1.172328086164043,
      "grad_norm": 1.15625,
      "learning_rate": 0.000498446031715272,
      "loss": 6.0972,
      "mean_token_accuracy": 0.13319127038121223,
      "num_tokens": 2583963.0,
      "step": 1415
    },
    {
      "entropy": 6.394404697418213,
      "epoch": 1.1764705882352942,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004984083141709391,
      "loss": 6.1374,
      "mean_token_accuracy": 0.133284143358469,
      "num_tokens": 2593350.0,
      "step": 1420
    },
    {
      "entropy": 6.300214338302612,
      "epoch": 1.180613090306545,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000498370145987623,
      "loss": 6.0387,
      "mean_token_accuracy": 0.13577536046504973,
      "num_tokens": 2602191.0,
      "step": 1425
    },
    {
      "entropy": 6.413609552383423,
      "epoch": 1.1847555923777962,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004983315272423127,
      "loss": 6.175,
      "mean_token_accuracy": 0.1315205328166485,
      "num_tokens": 2613590.0,
      "step": 1430
    },
    {
      "entropy": 6.330083179473877,
      "epoch": 1.1888980944490473,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004982924580129066,
      "loss": 5.9919,
      "mean_token_accuracy": 0.13426781222224235,
      "num_tokens": 2622072.0,
      "step": 1435
    },
    {
      "entropy": 6.408995532989502,
      "epoch": 1.1930405965202984,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004982529383782116,
      "loss": 6.1871,
      "mean_token_accuracy": 0.1286151461303234,
      "num_tokens": 2630904.0,
      "step": 1440
    },
    {
      "entropy": 6.3900384426116945,
      "epoch": 1.1971830985915493,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004982129684179432,
      "loss": 6.1579,
      "mean_token_accuracy": 0.12926432937383653,
      "num_tokens": 2639925.0,
      "step": 1445
    },
    {
      "entropy": 6.350859832763672,
      "epoch": 1.2013256006628004,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004981725482127251,
      "loss": 6.0588,
      "mean_token_accuracy": 0.1335553601384163,
      "num_tokens": 2649397.0,
      "step": 1450
    },
    {
      "entropy": 6.4356773853302,
      "epoch": 1.2054681027340513,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004981316778440892,
      "loss": 6.1985,
      "mean_token_accuracy": 0.13050690963864325,
      "num_tokens": 2658877.0,
      "step": 1455
    },
    {
      "entropy": 6.354866790771484,
      "epoch": 1.2096106048053024,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004980903573944756,
      "loss": 6.0988,
      "mean_token_accuracy": 0.1269911654293537,
      "num_tokens": 2668518.0,
      "step": 1460
    },
    {
      "entropy": 6.373867654800415,
      "epoch": 1.2137531068765535,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004980485869472322,
      "loss": 6.1852,
      "mean_token_accuracy": 0.12791354209184647,
      "num_tokens": 2677459.0,
      "step": 1465
    },
    {
      "entropy": 6.468676900863647,
      "epoch": 1.2178956089478046,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004980063665866143,
      "loss": 6.1557,
      "mean_token_accuracy": 0.12465783432126046,
      "num_tokens": 2686668.0,
      "step": 1470
    },
    {
      "entropy": 6.215289306640625,
      "epoch": 1.2220381110190555,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004979636963977853,
      "loss": 5.9811,
      "mean_token_accuracy": 0.1315767101943493,
      "num_tokens": 2695545.0,
      "step": 1475
    },
    {
      "entropy": 6.380659484863282,
      "epoch": 1.2261806130903066,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004979205764668156,
      "loss": 6.1323,
      "mean_token_accuracy": 0.1333031803369522,
      "num_tokens": 2705514.0,
      "step": 1480
    },
    {
      "entropy": 6.364181184768677,
      "epoch": 1.2303231151615575,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004978770068806825,
      "loss": 6.0352,
      "mean_token_accuracy": 0.13151699304580688,
      "num_tokens": 2714373.0,
      "step": 1485
    },
    {
      "entropy": 6.372594594955444,
      "epoch": 1.2344656172328086,
      "grad_norm": 1.078125,
      "learning_rate": 0.000497832987727271,
      "loss": 6.1622,
      "mean_token_accuracy": 0.1286887414753437,
      "num_tokens": 2723887.0,
      "step": 1490
    },
    {
      "entropy": 6.338165140151977,
      "epoch": 1.2386081193040597,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004977885190953724,
      "loss": 6.0249,
      "mean_token_accuracy": 0.13291778564453124,
      "num_tokens": 2733591.0,
      "step": 1495
    },
    {
      "entropy": 6.403629159927368,
      "epoch": 1.2427506213753108,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004977436010746846,
      "loss": 6.1543,
      "mean_token_accuracy": 0.1351762518286705,
      "num_tokens": 2741638.0,
      "step": 1500
    },
    {
      "epoch": 1.2427506213753108,
      "eval_entropy": 6.184635734761447,
      "eval_loss": 6.306137561798096,
      "eval_mean_token_accuracy": 0.12917301741871498,
      "eval_num_tokens": 2741638.0,
      "eval_runtime": 2.1134,
      "eval_samples_per_second": 1772.967,
      "eval_steps_per_second": 221.917,
      "step": 1500
    }
  ],
  "logging_steps": 5,
  "max_steps": 12060,
  "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": 606177840936960.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}