{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.06609166914510425,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.691955471038819,
      "epoch": 0.0003304583457255213,
      "grad_norm": 12.8125,
      "learning_rate": 2e-06,
      "loss": 10.8273,
      "mean_token_accuracy": 0.0,
      "num_tokens": 10288.0,
      "step": 5
    },
    {
      "entropy": 10.691919994354247,
      "epoch": 0.0006609166914510426,
      "grad_norm": 10.875,
      "learning_rate": 4.5e-06,
      "loss": 10.7719,
      "mean_token_accuracy": 0.0,
      "num_tokens": 20647.0,
      "step": 10
    },
    {
      "entropy": 10.6910964012146,
      "epoch": 0.000991375037176564,
      "grad_norm": 8.6875,
      "learning_rate": 7e-06,
      "loss": 10.5372,
      "mean_token_accuracy": 0.016732628224417567,
      "num_tokens": 30934.0,
      "step": 15
    },
    {
      "entropy": 10.682924270629883,
      "epoch": 0.0013218333829020852,
      "grad_norm": 5.9375,
      "learning_rate": 9.5e-06,
      "loss": 10.2268,
      "mean_token_accuracy": 0.039138264022767544,
      "num_tokens": 41985.0,
      "step": 20
    },
    {
      "entropy": 10.630903339385986,
      "epoch": 0.0016522917286276066,
      "grad_norm": 3.859375,
      "learning_rate": 1.2e-05,
      "loss": 9.9501,
      "mean_token_accuracy": 0.04244728535413742,
      "num_tokens": 52249.0,
      "step": 25
    },
    {
      "entropy": 10.59464464187622,
      "epoch": 0.001982750074353128,
      "grad_norm": 3.171875,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 9.8224,
      "mean_token_accuracy": 0.039565733075141905,
      "num_tokens": 63022.0,
      "step": 30
    },
    {
      "entropy": 10.583414649963379,
      "epoch": 0.002313208420078649,
      "grad_norm": 2.8125,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 9.7332,
      "mean_token_accuracy": 0.042595500499010085,
      "num_tokens": 74862.0,
      "step": 35
    },
    {
      "entropy": 10.582724380493165,
      "epoch": 0.0026436667658041703,
      "grad_norm": 2.484375,
      "learning_rate": 1.95e-05,
      "loss": 9.7531,
      "mean_token_accuracy": 0.03622140642255545,
      "num_tokens": 86588.0,
      "step": 40
    },
    {
      "entropy": 10.594015216827392,
      "epoch": 0.0029741251115296915,
      "grad_norm": 2.5625,
      "learning_rate": 2.2e-05,
      "loss": 9.6536,
      "mean_token_accuracy": 0.037497786805033684,
      "num_tokens": 96248.0,
      "step": 45
    },
    {
      "entropy": 10.578313446044922,
      "epoch": 0.003304583457255213,
      "grad_norm": 2.21875,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.6594,
      "mean_token_accuracy": 0.03918336480855942,
      "num_tokens": 107167.0,
      "step": 50
    },
    {
      "entropy": 10.555775451660157,
      "epoch": 0.0036350418029807343,
      "grad_norm": 2.265625,
      "learning_rate": 2.7e-05,
      "loss": 9.5383,
      "mean_token_accuracy": 0.043136011064052585,
      "num_tokens": 117207.0,
      "step": 55
    },
    {
      "entropy": 10.549382877349853,
      "epoch": 0.003965500148706256,
      "grad_norm": 2.109375,
      "learning_rate": 2.95e-05,
      "loss": 9.5206,
      "mean_token_accuracy": 0.04122119285166263,
      "num_tokens": 127335.0,
      "step": 60
    },
    {
      "entropy": 10.5144211769104,
      "epoch": 0.004295958494431777,
      "grad_norm": 2.28125,
      "learning_rate": 3.2e-05,
      "loss": 9.4157,
      "mean_token_accuracy": 0.051365657895803454,
      "num_tokens": 138329.0,
      "step": 65
    },
    {
      "entropy": 10.402010536193847,
      "epoch": 0.004626416840157298,
      "grad_norm": 2.234375,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.3531,
      "mean_token_accuracy": 0.04862799309194088,
      "num_tokens": 149993.0,
      "step": 70
    },
    {
      "entropy": 10.443185424804687,
      "epoch": 0.00495687518588282,
      "grad_norm": 2.0625,
      "learning_rate": 3.7e-05,
      "loss": 9.2381,
      "mean_token_accuracy": 0.05048903152346611,
      "num_tokens": 160872.0,
      "step": 75
    },
    {
      "entropy": 10.446312522888183,
      "epoch": 0.005287333531608341,
      "grad_norm": 1.9765625,
      "learning_rate": 3.95e-05,
      "loss": 9.2065,
      "mean_token_accuracy": 0.04706168882548809,
      "num_tokens": 172141.0,
      "step": 80
    },
    {
      "entropy": 10.442552852630616,
      "epoch": 0.005617791877333862,
      "grad_norm": 2.046875,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.036,
      "mean_token_accuracy": 0.05399964451789856,
      "num_tokens": 183762.0,
      "step": 85
    },
    {
      "entropy": 10.36781177520752,
      "epoch": 0.005948250223059383,
      "grad_norm": 1.8984375,
      "learning_rate": 4.45e-05,
      "loss": 8.9932,
      "mean_token_accuracy": 0.05456707440316677,
      "num_tokens": 196109.0,
      "step": 90
    },
    {
      "entropy": 10.29437780380249,
      "epoch": 0.006278708568784905,
      "grad_norm": 2.453125,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.7929,
      "mean_token_accuracy": 0.061743784323334695,
      "num_tokens": 205671.0,
      "step": 95
    },
    {
      "entropy": 10.216643905639648,
      "epoch": 0.006609166914510426,
      "grad_norm": 1.96875,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.7749,
      "mean_token_accuracy": 0.056532309576869014,
      "num_tokens": 217256.0,
      "step": 100
    },
    {
      "entropy": 10.16990032196045,
      "epoch": 0.006939625260235947,
      "grad_norm": 2.0625,
      "learning_rate": 5.2e-05,
      "loss": 8.6032,
      "mean_token_accuracy": 0.06618640683591366,
      "num_tokens": 228066.0,
      "step": 105
    },
    {
      "entropy": 10.083656120300294,
      "epoch": 0.007270083605961469,
      "grad_norm": 1.9609375,
      "learning_rate": 5.45e-05,
      "loss": 8.5398,
      "mean_token_accuracy": 0.06366933546960354,
      "num_tokens": 238643.0,
      "step": 110
    },
    {
      "entropy": 10.01426429748535,
      "epoch": 0.00760054195168699,
      "grad_norm": 1.9453125,
      "learning_rate": 5.7e-05,
      "loss": 8.4763,
      "mean_token_accuracy": 0.0619691651314497,
      "num_tokens": 249812.0,
      "step": 115
    },
    {
      "entropy": 9.912741661071777,
      "epoch": 0.007931000297412512,
      "grad_norm": 1.4921875,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.3405,
      "mean_token_accuracy": 0.06653297729790211,
      "num_tokens": 261942.0,
      "step": 120
    },
    {
      "entropy": 9.791018199920654,
      "epoch": 0.008261458643138032,
      "grad_norm": 2.109375,
      "learning_rate": 6.2e-05,
      "loss": 8.179,
      "mean_token_accuracy": 0.07168376296758652,
      "num_tokens": 272537.0,
      "step": 125
    },
    {
      "entropy": 9.592787837982177,
      "epoch": 0.008591916988863553,
      "grad_norm": 1.5078125,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.0936,
      "mean_token_accuracy": 0.06862132512032985,
      "num_tokens": 283610.0,
      "step": 130
    },
    {
      "entropy": 9.465274620056153,
      "epoch": 0.008922375334589075,
      "grad_norm": 1.2734375,
      "learning_rate": 6.7e-05,
      "loss": 8.0333,
      "mean_token_accuracy": 0.06386486329138279,
      "num_tokens": 294257.0,
      "step": 135
    },
    {
      "entropy": 9.326618099212647,
      "epoch": 0.009252833680314597,
      "grad_norm": 1.4453125,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.9842,
      "mean_token_accuracy": 0.060935232788324356,
      "num_tokens": 304405.0,
      "step": 140
    },
    {
      "entropy": 9.061574840545655,
      "epoch": 0.009583292026040118,
      "grad_norm": 1.1875,
      "learning_rate": 7.2e-05,
      "loss": 7.7675,
      "mean_token_accuracy": 0.07102037817239762,
      "num_tokens": 314413.0,
      "step": 145
    },
    {
      "entropy": 8.9353346824646,
      "epoch": 0.00991375037176564,
      "grad_norm": 1.3125,
      "learning_rate": 7.45e-05,
      "loss": 7.7801,
      "mean_token_accuracy": 0.06595919616520404,
      "num_tokens": 325010.0,
      "step": 150
    },
    {
      "entropy": 8.705840969085694,
      "epoch": 0.01024420871749116,
      "grad_norm": 1.3359375,
      "learning_rate": 7.7e-05,
      "loss": 7.7226,
      "mean_token_accuracy": 0.06655254885554314,
      "num_tokens": 336421.0,
      "step": 155
    },
    {
      "entropy": 8.54863624572754,
      "epoch": 0.010574667063216681,
      "grad_norm": 1.2890625,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.5898,
      "mean_token_accuracy": 0.07376853227615357,
      "num_tokens": 346930.0,
      "step": 160
    },
    {
      "entropy": 8.424325942993164,
      "epoch": 0.010905125408942203,
      "grad_norm": 1.3359375,
      "learning_rate": 8.2e-05,
      "loss": 7.6097,
      "mean_token_accuracy": 0.07119092084467411,
      "num_tokens": 357874.0,
      "step": 165
    },
    {
      "entropy": 8.46063985824585,
      "epoch": 0.011235583754667725,
      "grad_norm": 1.203125,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.6735,
      "mean_token_accuracy": 0.06998510733246803,
      "num_tokens": 370138.0,
      "step": 170
    },
    {
      "entropy": 8.31341323852539,
      "epoch": 0.011566042100393246,
      "grad_norm": 1.2109375,
      "learning_rate": 8.7e-05,
      "loss": 7.6173,
      "mean_token_accuracy": 0.06727928221225739,
      "num_tokens": 379999.0,
      "step": 175
    },
    {
      "entropy": 8.19371862411499,
      "epoch": 0.011896500446118766,
      "grad_norm": 0.96484375,
      "learning_rate": 8.95e-05,
      "loss": 7.4712,
      "mean_token_accuracy": 0.07500850819051266,
      "num_tokens": 392754.0,
      "step": 180
    },
    {
      "entropy": 8.132864665985107,
      "epoch": 0.012226958791844288,
      "grad_norm": 1.0625,
      "learning_rate": 9.2e-05,
      "loss": 7.5206,
      "mean_token_accuracy": 0.06753306314349175,
      "num_tokens": 403771.0,
      "step": 185
    },
    {
      "entropy": 8.108299064636231,
      "epoch": 0.01255741713756981,
      "grad_norm": 2.125,
      "learning_rate": 9.45e-05,
      "loss": 7.5477,
      "mean_token_accuracy": 0.07148384340107442,
      "num_tokens": 415381.0,
      "step": 190
    },
    {
      "entropy": 8.155412006378175,
      "epoch": 0.01288787548329533,
      "grad_norm": 1.3125,
      "learning_rate": 9.7e-05,
      "loss": 7.4227,
      "mean_token_accuracy": 0.07254996970295906,
      "num_tokens": 425919.0,
      "step": 195
    },
    {
      "entropy": 8.072521829605103,
      "epoch": 0.013218333829020852,
      "grad_norm": 1.296875,
      "learning_rate": 9.95e-05,
      "loss": 7.3718,
      "mean_token_accuracy": 0.06771104149520397,
      "num_tokens": 437753.0,
      "step": 200
    },
    {
      "entropy": 7.959475183486939,
      "epoch": 0.013548792174746372,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000102,
      "loss": 7.3961,
      "mean_token_accuracy": 0.07540397495031356,
      "num_tokens": 449079.0,
      "step": 205
    },
    {
      "entropy": 7.956682586669922,
      "epoch": 0.013879250520471894,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.3773,
      "mean_token_accuracy": 0.07537127286195755,
      "num_tokens": 459250.0,
      "step": 210
    },
    {
      "entropy": 7.86835880279541,
      "epoch": 0.014209708866197416,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000107,
      "loss": 7.3512,
      "mean_token_accuracy": 0.07468458339571953,
      "num_tokens": 470898.0,
      "step": 215
    },
    {
      "entropy": 7.843323755264282,
      "epoch": 0.014540167211922937,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001095,
      "loss": 7.3402,
      "mean_token_accuracy": 0.07922964245080948,
      "num_tokens": 481116.0,
      "step": 220
    },
    {
      "entropy": 7.847117757797241,
      "epoch": 0.014870625557648459,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000112,
      "loss": 7.349,
      "mean_token_accuracy": 0.07711787447333336,
      "num_tokens": 492639.0,
      "step": 225
    },
    {
      "entropy": 7.853758859634399,
      "epoch": 0.01520108390337398,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001145,
      "loss": 7.3957,
      "mean_token_accuracy": 0.07615091502666474,
      "num_tokens": 504838.0,
      "step": 230
    },
    {
      "entropy": 7.945042896270752,
      "epoch": 0.0155315422490995,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.297,
      "mean_token_accuracy": 0.07279803417623043,
      "num_tokens": 515982.0,
      "step": 235
    },
    {
      "entropy": 7.722148132324219,
      "epoch": 0.015862000594825024,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.2946,
      "mean_token_accuracy": 0.07706804946064949,
      "num_tokens": 526830.0,
      "step": 240
    },
    {
      "entropy": 7.7506262302398685,
      "epoch": 0.016192458940550544,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000122,
      "loss": 7.342,
      "mean_token_accuracy": 0.07662090063095092,
      "num_tokens": 538111.0,
      "step": 245
    },
    {
      "entropy": 7.804240894317627,
      "epoch": 0.016522917286276063,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001245,
      "loss": 7.2915,
      "mean_token_accuracy": 0.07277994081377984,
      "num_tokens": 548286.0,
      "step": 250
    },
    {
      "entropy": 7.752009725570678,
      "epoch": 0.016853375632001587,
      "grad_norm": 1.21875,
      "learning_rate": 0.000127,
      "loss": 7.2752,
      "mean_token_accuracy": 0.07613983601331711,
      "num_tokens": 558638.0,
      "step": 255
    },
    {
      "entropy": 7.705926036834716,
      "epoch": 0.017183833977727107,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0001295,
      "loss": 7.2402,
      "mean_token_accuracy": 0.07607613280415534,
      "num_tokens": 568909.0,
      "step": 260
    },
    {
      "entropy": 7.722606134414673,
      "epoch": 0.01751429232345263,
      "grad_norm": 1.484375,
      "learning_rate": 0.000132,
      "loss": 7.2178,
      "mean_token_accuracy": 0.07890873812139035,
      "num_tokens": 580392.0,
      "step": 265
    },
    {
      "entropy": 7.688320446014404,
      "epoch": 0.01784475066917815,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.2767,
      "mean_token_accuracy": 0.07444539405405522,
      "num_tokens": 590160.0,
      "step": 270
    },
    {
      "entropy": 7.632183170318603,
      "epoch": 0.01817520901490367,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.1822,
      "mean_token_accuracy": 0.07618632428348064,
      "num_tokens": 601054.0,
      "step": 275
    },
    {
      "entropy": 7.7831042289733885,
      "epoch": 0.018505667360629193,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001395,
      "loss": 7.3189,
      "mean_token_accuracy": 0.07360220067203045,
      "num_tokens": 611908.0,
      "step": 280
    },
    {
      "entropy": 7.640690326690674,
      "epoch": 0.018836125706354713,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.258,
      "mean_token_accuracy": 0.07853379994630813,
      "num_tokens": 622259.0,
      "step": 285
    },
    {
      "entropy": 7.618532133102417,
      "epoch": 0.019166584052080236,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001445,
      "loss": 7.1215,
      "mean_token_accuracy": 0.08114137947559356,
      "num_tokens": 632728.0,
      "step": 290
    },
    {
      "entropy": 7.7060990810394285,
      "epoch": 0.019497042397805756,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000147,
      "loss": 7.309,
      "mean_token_accuracy": 0.07869038358330727,
      "num_tokens": 644600.0,
      "step": 295
    },
    {
      "entropy": 7.549208354949951,
      "epoch": 0.01982750074353128,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001495,
      "loss": 7.1684,
      "mean_token_accuracy": 0.08070796430110931,
      "num_tokens": 654004.0,
      "step": 300
    },
    {
      "entropy": 7.680343866348267,
      "epoch": 0.0201579590892568,
      "grad_norm": 1.03125,
      "learning_rate": 0.000152,
      "loss": 7.1505,
      "mean_token_accuracy": 0.08140776604413986,
      "num_tokens": 665275.0,
      "step": 305
    },
    {
      "entropy": 7.586634254455566,
      "epoch": 0.02048841743498232,
      "grad_norm": 1.1875,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.2169,
      "mean_token_accuracy": 0.07984307780861855,
      "num_tokens": 677397.0,
      "step": 310
    },
    {
      "entropy": 7.557528400421143,
      "epoch": 0.020818875780707843,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000157,
      "loss": 7.0796,
      "mean_token_accuracy": 0.08578637950122356,
      "num_tokens": 687975.0,
      "step": 315
    },
    {
      "entropy": 7.519621467590332,
      "epoch": 0.021149334126433363,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001595,
      "loss": 7.0386,
      "mean_token_accuracy": 0.09029009193181992,
      "num_tokens": 699102.0,
      "step": 320
    },
    {
      "entropy": 7.4558220386505125,
      "epoch": 0.021479792472158886,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000162,
      "loss": 7.1014,
      "mean_token_accuracy": 0.07989090159535409,
      "num_tokens": 712007.0,
      "step": 325
    },
    {
      "entropy": 7.526741886138916,
      "epoch": 0.021810250817884406,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.0863,
      "mean_token_accuracy": 0.08748645186424256,
      "num_tokens": 723757.0,
      "step": 330
    },
    {
      "entropy": 7.52968955039978,
      "epoch": 0.022140709163609926,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.0994,
      "mean_token_accuracy": 0.08689600750803947,
      "num_tokens": 734353.0,
      "step": 335
    },
    {
      "entropy": 7.439078617095947,
      "epoch": 0.02247116750933545,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.9892,
      "mean_token_accuracy": 0.0902569204568863,
      "num_tokens": 744673.0,
      "step": 340
    },
    {
      "entropy": 7.432306814193725,
      "epoch": 0.02280162585506097,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.0258,
      "mean_token_accuracy": 0.0794426791369915,
      "num_tokens": 755181.0,
      "step": 345
    },
    {
      "entropy": 7.557726144790649,
      "epoch": 0.023132084200786492,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.0575,
      "mean_token_accuracy": 0.08544816374778748,
      "num_tokens": 765558.0,
      "step": 350
    },
    {
      "entropy": 7.414818954467774,
      "epoch": 0.023462542546512012,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000177,
      "loss": 7.0836,
      "mean_token_accuracy": 0.07729014120995999,
      "num_tokens": 778224.0,
      "step": 355
    },
    {
      "entropy": 7.490838050842285,
      "epoch": 0.023793000892237532,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001795,
      "loss": 6.991,
      "mean_token_accuracy": 0.08522340208292008,
      "num_tokens": 789586.0,
      "step": 360
    },
    {
      "entropy": 7.417488718032837,
      "epoch": 0.024123459237963055,
      "grad_norm": 1.015625,
      "learning_rate": 0.000182,
      "loss": 7.0535,
      "mean_token_accuracy": 0.08176884576678276,
      "num_tokens": 801595.0,
      "step": 365
    },
    {
      "entropy": 7.4084100246429445,
      "epoch": 0.024453917583688575,
      "grad_norm": 1.015625,
      "learning_rate": 0.0001845,
      "loss": 7.0797,
      "mean_token_accuracy": 0.08523592539131641,
      "num_tokens": 812850.0,
      "step": 370
    },
    {
      "entropy": 7.462594747543335,
      "epoch": 0.0247843759294141,
      "grad_norm": 1.140625,
      "learning_rate": 0.000187,
      "loss": 7.078,
      "mean_token_accuracy": 0.08119027838110923,
      "num_tokens": 823125.0,
      "step": 375
    },
    {
      "entropy": 7.436292457580566,
      "epoch": 0.02511483427513962,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001895,
      "loss": 6.9898,
      "mean_token_accuracy": 0.08521195650100707,
      "num_tokens": 834838.0,
      "step": 380
    },
    {
      "entropy": 7.3805746078491214,
      "epoch": 0.02544529262086514,
      "grad_norm": 1.359375,
      "learning_rate": 0.000192,
      "loss": 7.0507,
      "mean_token_accuracy": 0.0866511344909668,
      "num_tokens": 845976.0,
      "step": 385
    },
    {
      "entropy": 7.368652486801148,
      "epoch": 0.02577575096659066,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001945,
      "loss": 7.0367,
      "mean_token_accuracy": 0.08296765461564064,
      "num_tokens": 857193.0,
      "step": 390
    },
    {
      "entropy": 7.454722309112549,
      "epoch": 0.02610620931231618,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.0047,
      "mean_token_accuracy": 0.08462346903979778,
      "num_tokens": 868471.0,
      "step": 395
    },
    {
      "entropy": 7.42730746269226,
      "epoch": 0.026436667658041705,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.9478,
      "mean_token_accuracy": 0.08773355185985565,
      "num_tokens": 879339.0,
      "step": 400
    },
    {
      "entropy": 7.206215667724609,
      "epoch": 0.026767126003767225,
      "grad_norm": 1.078125,
      "learning_rate": 0.000202,
      "loss": 6.8858,
      "mean_token_accuracy": 0.08911687806248665,
      "num_tokens": 890081.0,
      "step": 405
    },
    {
      "entropy": 7.307197189331054,
      "epoch": 0.027097584349492745,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.909,
      "mean_token_accuracy": 0.08829740136861801,
      "num_tokens": 899967.0,
      "step": 410
    },
    {
      "entropy": 7.394805383682251,
      "epoch": 0.027428042695218268,
      "grad_norm": 1.171875,
      "learning_rate": 0.000207,
      "loss": 6.9474,
      "mean_token_accuracy": 0.08440508171916009,
      "num_tokens": 910697.0,
      "step": 415
    },
    {
      "entropy": 7.307165718078613,
      "epoch": 0.027758501040943788,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002095,
      "loss": 7.0184,
      "mean_token_accuracy": 0.08860758394002914,
      "num_tokens": 920879.0,
      "step": 420
    },
    {
      "entropy": 7.332070398330688,
      "epoch": 0.02808895938666931,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000212,
      "loss": 6.93,
      "mean_token_accuracy": 0.0954623855650425,
      "num_tokens": 931566.0,
      "step": 425
    },
    {
      "entropy": 7.345228767395019,
      "epoch": 0.02841941773239483,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002145,
      "loss": 6.868,
      "mean_token_accuracy": 0.08827793225646019,
      "num_tokens": 941736.0,
      "step": 430
    },
    {
      "entropy": 7.245734167098999,
      "epoch": 0.028749876078120355,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.9063,
      "mean_token_accuracy": 0.08495340272784233,
      "num_tokens": 951652.0,
      "step": 435
    },
    {
      "entropy": 7.267821741104126,
      "epoch": 0.029080334423845874,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002195,
      "loss": 6.8298,
      "mean_token_accuracy": 0.08889664858579635,
      "num_tokens": 963242.0,
      "step": 440
    },
    {
      "entropy": 7.126494026184082,
      "epoch": 0.029410792769571394,
      "grad_norm": 1.1875,
      "learning_rate": 0.000222,
      "loss": 6.8716,
      "mean_token_accuracy": 0.08861989453434944,
      "num_tokens": 974078.0,
      "step": 445
    },
    {
      "entropy": 7.203296804428101,
      "epoch": 0.029741251115296918,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0002245,
      "loss": 6.8335,
      "mean_token_accuracy": 0.0920184887945652,
      "num_tokens": 986324.0,
      "step": 450
    },
    {
      "entropy": 7.216691255569458,
      "epoch": 0.030071709461022438,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.8409,
      "mean_token_accuracy": 0.09037284776568413,
      "num_tokens": 997625.0,
      "step": 455
    },
    {
      "entropy": 7.202143144607544,
      "epoch": 0.03040216780674796,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.8316,
      "mean_token_accuracy": 0.09061756655573845,
      "num_tokens": 1007914.0,
      "step": 460
    },
    {
      "entropy": 7.096453285217285,
      "epoch": 0.03073262615247348,
      "grad_norm": 1.0,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.792,
      "mean_token_accuracy": 0.0925156094133854,
      "num_tokens": 1018510.0,
      "step": 465
    },
    {
      "entropy": 7.169570064544677,
      "epoch": 0.031063084498199,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.7677,
      "mean_token_accuracy": 0.09652838632464408,
      "num_tokens": 1028532.0,
      "step": 470
    },
    {
      "entropy": 7.168693256378174,
      "epoch": 0.031393542843924524,
      "grad_norm": 1.09375,
      "learning_rate": 0.000237,
      "loss": 6.9262,
      "mean_token_accuracy": 0.08653468191623688,
      "num_tokens": 1040252.0,
      "step": 475
    },
    {
      "entropy": 7.152310705184936,
      "epoch": 0.03172400118965005,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002395,
      "loss": 6.7222,
      "mean_token_accuracy": 0.09424662217497826,
      "num_tokens": 1050486.0,
      "step": 480
    },
    {
      "entropy": 7.076179075241089,
      "epoch": 0.032054459535375564,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000242,
      "loss": 6.766,
      "mean_token_accuracy": 0.09703566282987594,
      "num_tokens": 1061218.0,
      "step": 485
    },
    {
      "entropy": 7.191670942306518,
      "epoch": 0.03238491788110109,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002445,
      "loss": 6.888,
      "mean_token_accuracy": 0.0887337252497673,
      "num_tokens": 1071077.0,
      "step": 490
    },
    {
      "entropy": 7.052482748031617,
      "epoch": 0.03271537622682661,
      "grad_norm": 1.6328125,
      "learning_rate": 0.000247,
      "loss": 6.8699,
      "mean_token_accuracy": 0.08658004626631736,
      "num_tokens": 1081434.0,
      "step": 495
    },
    {
      "entropy": 7.08971209526062,
      "epoch": 0.03304583457255213,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002495,
      "loss": 6.811,
      "mean_token_accuracy": 0.09111799448728561,
      "num_tokens": 1092237.0,
      "step": 500
    },
    {
      "entropy": 7.195395469665527,
      "epoch": 0.03337629291827765,
      "grad_norm": 1.109375,
      "learning_rate": 0.000252,
      "loss": 6.8992,
      "mean_token_accuracy": 0.0858646959066391,
      "num_tokens": 1102368.0,
      "step": 505
    },
    {
      "entropy": 7.053650522232056,
      "epoch": 0.033706751264003174,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002545,
      "loss": 6.7836,
      "mean_token_accuracy": 0.0907538540661335,
      "num_tokens": 1113174.0,
      "step": 510
    },
    {
      "entropy": 7.080994558334351,
      "epoch": 0.0340372096097287,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000257,
      "loss": 6.7959,
      "mean_token_accuracy": 0.09097289815545082,
      "num_tokens": 1124468.0,
      "step": 515
    },
    {
      "entropy": 7.159065294265747,
      "epoch": 0.03436766795545421,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002595,
      "loss": 6.7495,
      "mean_token_accuracy": 0.08832652196288109,
      "num_tokens": 1134691.0,
      "step": 520
    },
    {
      "entropy": 7.132658243179321,
      "epoch": 0.03469812630117974,
      "grad_norm": 1.03125,
      "learning_rate": 0.000262,
      "loss": 6.9154,
      "mean_token_accuracy": 0.0892325647175312,
      "num_tokens": 1146487.0,
      "step": 525
    },
    {
      "entropy": 7.005402994155884,
      "epoch": 0.03502858464690526,
      "grad_norm": 1.078125,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.7692,
      "mean_token_accuracy": 0.0914691999554634,
      "num_tokens": 1156897.0,
      "step": 530
    },
    {
      "entropy": 7.095370483398438,
      "epoch": 0.035359042992630776,
      "grad_norm": 1.0625,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.6971,
      "mean_token_accuracy": 0.09731558784842491,
      "num_tokens": 1167884.0,
      "step": 535
    },
    {
      "entropy": 6.940491104125977,
      "epoch": 0.0356895013383563,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.7671,
      "mean_token_accuracy": 0.09241313189268112,
      "num_tokens": 1178214.0,
      "step": 540
    },
    {
      "entropy": 7.2427126407623295,
      "epoch": 0.03601995968408182,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.8648,
      "mean_token_accuracy": 0.08732976764440536,
      "num_tokens": 1188139.0,
      "step": 545
    },
    {
      "entropy": 7.035971355438233,
      "epoch": 0.03635041802980734,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002745,
      "loss": 6.7803,
      "mean_token_accuracy": 0.09080611243844032,
      "num_tokens": 1198349.0,
      "step": 550
    },
    {
      "entropy": 6.935536193847656,
      "epoch": 0.03668087637553286,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000277,
      "loss": 6.7616,
      "mean_token_accuracy": 0.09351899325847626,
      "num_tokens": 1212208.0,
      "step": 555
    },
    {
      "entropy": 7.122801494598389,
      "epoch": 0.037011334721258386,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002795,
      "loss": 6.763,
      "mean_token_accuracy": 0.09313179403543473,
      "num_tokens": 1222600.0,
      "step": 560
    },
    {
      "entropy": 6.91089358329773,
      "epoch": 0.03734179306698391,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.6661,
      "mean_token_accuracy": 0.0968889094889164,
      "num_tokens": 1233383.0,
      "step": 565
    },
    {
      "entropy": 7.025641775131225,
      "epoch": 0.037672251412709426,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002845,
      "loss": 6.679,
      "mean_token_accuracy": 0.09819161146879196,
      "num_tokens": 1243571.0,
      "step": 570
    },
    {
      "entropy": 6.8856569766998295,
      "epoch": 0.03800270975843495,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000287,
      "loss": 6.7086,
      "mean_token_accuracy": 0.086531862616539,
      "num_tokens": 1255150.0,
      "step": 575
    },
    {
      "entropy": 6.880285835266113,
      "epoch": 0.03833316810416047,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002895,
      "loss": 6.7201,
      "mean_token_accuracy": 0.095609350502491,
      "num_tokens": 1265994.0,
      "step": 580
    },
    {
      "entropy": 6.986137962341308,
      "epoch": 0.03866362644988599,
      "grad_norm": 1.296875,
      "learning_rate": 0.000292,
      "loss": 6.7123,
      "mean_token_accuracy": 0.09263514503836631,
      "num_tokens": 1276244.0,
      "step": 585
    },
    {
      "entropy": 6.976613140106201,
      "epoch": 0.03899408479561151,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002945,
      "loss": 6.6008,
      "mean_token_accuracy": 0.1000402919948101,
      "num_tokens": 1285479.0,
      "step": 590
    },
    {
      "entropy": 6.9175629138946535,
      "epoch": 0.039324543141337036,
      "grad_norm": 1.125,
      "learning_rate": 0.000297,
      "loss": 6.7273,
      "mean_token_accuracy": 0.09454974755644799,
      "num_tokens": 1295591.0,
      "step": 595
    },
    {
      "entropy": 6.946761512756348,
      "epoch": 0.03965500148706256,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002995,
      "loss": 6.6503,
      "mean_token_accuracy": 0.10151792019605636,
      "num_tokens": 1305516.0,
      "step": 600
    },
    {
      "entropy": 6.922713088989258,
      "epoch": 0.039985459832788076,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000302,
      "loss": 6.647,
      "mean_token_accuracy": 0.09776916652917862,
      "num_tokens": 1316345.0,
      "step": 605
    },
    {
      "entropy": 6.957687187194824,
      "epoch": 0.0403159181785136,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003045,
      "loss": 6.8079,
      "mean_token_accuracy": 0.09032155275344848,
      "num_tokens": 1327332.0,
      "step": 610
    },
    {
      "entropy": 6.881102657318115,
      "epoch": 0.04064637652423912,
      "grad_norm": 1.15625,
      "learning_rate": 0.000307,
      "loss": 6.6599,
      "mean_token_accuracy": 0.09579801037907601,
      "num_tokens": 1337572.0,
      "step": 615
    },
    {
      "entropy": 6.878293704986572,
      "epoch": 0.04097683486996464,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003095,
      "loss": 6.6088,
      "mean_token_accuracy": 0.09545752853155136,
      "num_tokens": 1347712.0,
      "step": 620
    },
    {
      "entropy": 6.984841537475586,
      "epoch": 0.04130729321569016,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000312,
      "loss": 6.6957,
      "mean_token_accuracy": 0.0921433411538601,
      "num_tokens": 1359318.0,
      "step": 625
    },
    {
      "entropy": 6.789507246017456,
      "epoch": 0.041637751561415685,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003145,
      "loss": 6.6235,
      "mean_token_accuracy": 0.09607752040028572,
      "num_tokens": 1370322.0,
      "step": 630
    },
    {
      "entropy": 6.803044176101684,
      "epoch": 0.0419682099071412,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000317,
      "loss": 6.6378,
      "mean_token_accuracy": 0.09525535404682159,
      "num_tokens": 1382660.0,
      "step": 635
    },
    {
      "entropy": 6.93178014755249,
      "epoch": 0.042298668252866725,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003195,
      "loss": 6.6985,
      "mean_token_accuracy": 0.09292855113744736,
      "num_tokens": 1394377.0,
      "step": 640
    },
    {
      "entropy": 6.805371952056885,
      "epoch": 0.04262912659859225,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000322,
      "loss": 6.6399,
      "mean_token_accuracy": 0.09336945563554763,
      "num_tokens": 1404680.0,
      "step": 645
    },
    {
      "entropy": 6.839718198776245,
      "epoch": 0.04295958494431777,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.67,
      "mean_token_accuracy": 0.09462371170520782,
      "num_tokens": 1416715.0,
      "step": 650
    },
    {
      "entropy": 6.81149001121521,
      "epoch": 0.04329004329004329,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.6178,
      "mean_token_accuracy": 0.10017903596162796,
      "num_tokens": 1427354.0,
      "step": 655
    },
    {
      "entropy": 6.818669986724854,
      "epoch": 0.04362050163576881,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.5643,
      "mean_token_accuracy": 0.09899974688887596,
      "num_tokens": 1438554.0,
      "step": 660
    },
    {
      "entropy": 6.8915050506591795,
      "epoch": 0.043950959981494335,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.7589,
      "mean_token_accuracy": 0.0964146949350834,
      "num_tokens": 1449213.0,
      "step": 665
    },
    {
      "entropy": 6.898461866378784,
      "epoch": 0.04428141832721985,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.6459,
      "mean_token_accuracy": 0.09545622617006302,
      "num_tokens": 1460571.0,
      "step": 670
    },
    {
      "entropy": 6.756736993789673,
      "epoch": 0.044611876672945375,
      "grad_norm": 1.53125,
      "learning_rate": 0.000337,
      "loss": 6.7076,
      "mean_token_accuracy": 0.09004327580332756,
      "num_tokens": 1470638.0,
      "step": 675
    },
    {
      "entropy": 6.968370485305786,
      "epoch": 0.0449423350186709,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003395,
      "loss": 6.7096,
      "mean_token_accuracy": 0.09662523344159127,
      "num_tokens": 1481773.0,
      "step": 680
    },
    {
      "entropy": 6.718348550796509,
      "epoch": 0.045272793364396414,
      "grad_norm": 1.109375,
      "learning_rate": 0.000342,
      "loss": 6.638,
      "mean_token_accuracy": 0.09720009118318558,
      "num_tokens": 1491982.0,
      "step": 685
    },
    {
      "entropy": 6.835583591461182,
      "epoch": 0.04560325171012194,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.6416,
      "mean_token_accuracy": 0.09842949360609055,
      "num_tokens": 1503027.0,
      "step": 690
    },
    {
      "entropy": 6.808763265609741,
      "epoch": 0.04593371005584746,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000347,
      "loss": 6.6703,
      "mean_token_accuracy": 0.09374243021011353,
      "num_tokens": 1515977.0,
      "step": 695
    },
    {
      "entropy": 6.725175905227661,
      "epoch": 0.046264168401572985,
      "grad_norm": 1.125,
      "learning_rate": 0.0003495,
      "loss": 6.5534,
      "mean_token_accuracy": 0.09220106229186058,
      "num_tokens": 1527309.0,
      "step": 700
    },
    {
      "entropy": 6.720183277130127,
      "epoch": 0.0465946267472985,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000352,
      "loss": 6.5864,
      "mean_token_accuracy": 0.10298208072781563,
      "num_tokens": 1538233.0,
      "step": 705
    },
    {
      "entropy": 6.7770648956298825,
      "epoch": 0.046925085093024024,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0003545,
      "loss": 6.5643,
      "mean_token_accuracy": 0.09746524840593337,
      "num_tokens": 1549893.0,
      "step": 710
    },
    {
      "entropy": 6.791210126876831,
      "epoch": 0.04725554343874955,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000357,
      "loss": 6.6119,
      "mean_token_accuracy": 0.09718122556805611,
      "num_tokens": 1560926.0,
      "step": 715
    },
    {
      "entropy": 6.679304456710815,
      "epoch": 0.047586001784475064,
      "grad_norm": 1.0,
      "learning_rate": 0.0003595,
      "loss": 6.4565,
      "mean_token_accuracy": 0.10528466776013375,
      "num_tokens": 1571826.0,
      "step": 720
    },
    {
      "entropy": 6.780915212631226,
      "epoch": 0.04791646013020059,
      "grad_norm": 1.046875,
      "learning_rate": 0.000362,
      "loss": 6.6402,
      "mean_token_accuracy": 0.09756145626306534,
      "num_tokens": 1582710.0,
      "step": 725
    },
    {
      "entropy": 6.73593544960022,
      "epoch": 0.04824691847592611,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003645,
      "loss": 6.4899,
      "mean_token_accuracy": 0.09683176577091217,
      "num_tokens": 1593148.0,
      "step": 730
    },
    {
      "entropy": 6.734481763839722,
      "epoch": 0.048577376821651634,
      "grad_norm": 1.140625,
      "learning_rate": 0.000367,
      "loss": 6.5603,
      "mean_token_accuracy": 0.09808081015944481,
      "num_tokens": 1603286.0,
      "step": 735
    },
    {
      "entropy": 6.590597772598267,
      "epoch": 0.04890783516737715,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003695,
      "loss": 6.4828,
      "mean_token_accuracy": 0.1038731187582016,
      "num_tokens": 1613341.0,
      "step": 740
    },
    {
      "entropy": 6.7730015277862545,
      "epoch": 0.049238293513102674,
      "grad_norm": 1.03125,
      "learning_rate": 0.000372,
      "loss": 6.5973,
      "mean_token_accuracy": 0.09209345281124115,
      "num_tokens": 1624733.0,
      "step": 745
    },
    {
      "entropy": 6.66158356666565,
      "epoch": 0.0495687518588282,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0003745,
      "loss": 6.5181,
      "mean_token_accuracy": 0.09734064191579819,
      "num_tokens": 1635369.0,
      "step": 750
    },
    {
      "entropy": 6.763376712799072,
      "epoch": 0.049899210204553714,
      "grad_norm": 1.125,
      "learning_rate": 0.000377,
      "loss": 6.5794,
      "mean_token_accuracy": 0.10017797723412514,
      "num_tokens": 1646391.0,
      "step": 755
    },
    {
      "entropy": 6.685816621780395,
      "epoch": 0.05022966855027924,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003795,
      "loss": 6.6399,
      "mean_token_accuracy": 0.09641192629933357,
      "num_tokens": 1657923.0,
      "step": 760
    },
    {
      "entropy": 6.643207740783692,
      "epoch": 0.05056012689600476,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000382,
      "loss": 6.5121,
      "mean_token_accuracy": 0.09868812561035156,
      "num_tokens": 1669732.0,
      "step": 765
    },
    {
      "entropy": 6.617096710205078,
      "epoch": 0.05089058524173028,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003845,
      "loss": 6.4583,
      "mean_token_accuracy": 0.10043382048606872,
      "num_tokens": 1680646.0,
      "step": 770
    },
    {
      "entropy": 6.679386281967163,
      "epoch": 0.0512210435874558,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.5689,
      "mean_token_accuracy": 0.09984266906976699,
      "num_tokens": 1690855.0,
      "step": 775
    },
    {
      "entropy": 6.724302148818969,
      "epoch": 0.05155150193318132,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.529,
      "mean_token_accuracy": 0.10136074721813201,
      "num_tokens": 1701798.0,
      "step": 780
    },
    {
      "entropy": 6.5292439460754395,
      "epoch": 0.05188196027890685,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.456,
      "mean_token_accuracy": 0.09994935691356659,
      "num_tokens": 1713307.0,
      "step": 785
    },
    {
      "entropy": 6.717782592773437,
      "epoch": 0.05221241862463236,
      "grad_norm": 1.25,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.5878,
      "mean_token_accuracy": 0.09384994134306908,
      "num_tokens": 1724158.0,
      "step": 790
    },
    {
      "entropy": 6.530282402038575,
      "epoch": 0.05254287697035789,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.5416,
      "mean_token_accuracy": 0.10089174509048462,
      "num_tokens": 1734978.0,
      "step": 795
    },
    {
      "entropy": 6.784086656570435,
      "epoch": 0.05287333531608341,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0003995,
      "loss": 6.5628,
      "mean_token_accuracy": 0.09250908866524696,
      "num_tokens": 1745757.0,
      "step": 800
    },
    {
      "entropy": 6.637925291061402,
      "epoch": 0.053203793661808926,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000402,
      "loss": 6.5656,
      "mean_token_accuracy": 0.10313564985990524,
      "num_tokens": 1755370.0,
      "step": 805
    },
    {
      "entropy": 6.652379417419434,
      "epoch": 0.05353425200753445,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004045,
      "loss": 6.5075,
      "mean_token_accuracy": 0.10148903802037239,
      "num_tokens": 1766031.0,
      "step": 810
    },
    {
      "entropy": 6.659540414810181,
      "epoch": 0.05386471035325997,
      "grad_norm": 1.125,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.5279,
      "mean_token_accuracy": 0.10324057936668396,
      "num_tokens": 1778672.0,
      "step": 815
    },
    {
      "entropy": 6.496312141418457,
      "epoch": 0.05419516869898549,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004095,
      "loss": 6.4894,
      "mean_token_accuracy": 0.10615217164158822,
      "num_tokens": 1789912.0,
      "step": 820
    },
    {
      "entropy": 6.570638751983642,
      "epoch": 0.05452562704471101,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000412,
      "loss": 6.4024,
      "mean_token_accuracy": 0.10341350361704826,
      "num_tokens": 1800350.0,
      "step": 825
    },
    {
      "entropy": 6.506610012054443,
      "epoch": 0.054856085390436536,
      "grad_norm": 1.125,
      "learning_rate": 0.0004145,
      "loss": 6.4464,
      "mean_token_accuracy": 0.09816132262349128,
      "num_tokens": 1811353.0,
      "step": 830
    },
    {
      "entropy": 6.599850749969482,
      "epoch": 0.05518654373616206,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000417,
      "loss": 6.5677,
      "mean_token_accuracy": 0.10067287907004356,
      "num_tokens": 1822702.0,
      "step": 835
    },
    {
      "entropy": 6.622064638137817,
      "epoch": 0.055517002081887576,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004195,
      "loss": 6.5611,
      "mean_token_accuracy": 0.09993282034993171,
      "num_tokens": 1833520.0,
      "step": 840
    },
    {
      "entropy": 6.631695938110352,
      "epoch": 0.0558474604276131,
      "grad_norm": 1.046875,
      "learning_rate": 0.000422,
      "loss": 6.4769,
      "mean_token_accuracy": 0.10197674706578255,
      "num_tokens": 1843905.0,
      "step": 845
    },
    {
      "entropy": 6.651054716110229,
      "epoch": 0.05617791877333862,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004245,
      "loss": 6.5903,
      "mean_token_accuracy": 0.09589819759130477,
      "num_tokens": 1854350.0,
      "step": 850
    },
    {
      "entropy": 6.625249004364013,
      "epoch": 0.05650837711906414,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000427,
      "loss": 6.5257,
      "mean_token_accuracy": 0.1002203494310379,
      "num_tokens": 1864279.0,
      "step": 855
    },
    {
      "entropy": 6.585396718978882,
      "epoch": 0.05683883546478966,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004295,
      "loss": 6.5416,
      "mean_token_accuracy": 0.10187701731920243,
      "num_tokens": 1876618.0,
      "step": 860
    },
    {
      "entropy": 6.66397533416748,
      "epoch": 0.057169293810515186,
      "grad_norm": 1.09375,
      "learning_rate": 0.000432,
      "loss": 6.5215,
      "mean_token_accuracy": 0.09983031451702118,
      "num_tokens": 1888689.0,
      "step": 865
    },
    {
      "entropy": 6.483687019348144,
      "epoch": 0.05749975215624071,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004345,
      "loss": 6.4599,
      "mean_token_accuracy": 0.09891335591673851,
      "num_tokens": 1899527.0,
      "step": 870
    },
    {
      "entropy": 6.619280052185059,
      "epoch": 0.057830210501966225,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000437,
      "loss": 6.5642,
      "mean_token_accuracy": 0.09296367540955544,
      "num_tokens": 1910212.0,
      "step": 875
    },
    {
      "entropy": 6.569179058074951,
      "epoch": 0.05816066884769175,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004395,
      "loss": 6.4496,
      "mean_token_accuracy": 0.10345134884119034,
      "num_tokens": 1919938.0,
      "step": 880
    },
    {
      "entropy": 6.494620656967163,
      "epoch": 0.05849112719341727,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000442,
      "loss": 6.4105,
      "mean_token_accuracy": 0.106119804084301,
      "num_tokens": 1929984.0,
      "step": 885
    },
    {
      "entropy": 6.534124374389648,
      "epoch": 0.05882158553914279,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004445,
      "loss": 6.5382,
      "mean_token_accuracy": 0.093291524797678,
      "num_tokens": 1941683.0,
      "step": 890
    },
    {
      "entropy": 6.617558002471924,
      "epoch": 0.05915204388486831,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000447,
      "loss": 6.5158,
      "mean_token_accuracy": 0.09773639366030692,
      "num_tokens": 1952847.0,
      "step": 895
    },
    {
      "entropy": 6.576697492599488,
      "epoch": 0.059482502230593835,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.4391,
      "mean_token_accuracy": 0.10089132189750671,
      "num_tokens": 1963449.0,
      "step": 900
    },
    {
      "entropy": 6.443887376785279,
      "epoch": 0.05981296057631935,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.4521,
      "mean_token_accuracy": 0.10279465019702912,
      "num_tokens": 1973855.0,
      "step": 905
    },
    {
      "entropy": 6.472233676910401,
      "epoch": 0.060143418922044875,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.4577,
      "mean_token_accuracy": 0.09866586327552795,
      "num_tokens": 1985015.0,
      "step": 910
    },
    {
      "entropy": 6.604752397537231,
      "epoch": 0.0604738772677704,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.5311,
      "mean_token_accuracy": 0.10184109881520272,
      "num_tokens": 1996042.0,
      "step": 915
    },
    {
      "entropy": 6.517620611190796,
      "epoch": 0.06080433561349592,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.4881,
      "mean_token_accuracy": 0.09871948435902596,
      "num_tokens": 2008155.0,
      "step": 920
    },
    {
      "entropy": 6.533037567138672,
      "epoch": 0.06113479395922144,
      "grad_norm": 1.046875,
      "learning_rate": 0.000462,
      "loss": 6.4375,
      "mean_token_accuracy": 0.10870096236467361,
      "num_tokens": 2017975.0,
      "step": 925
    },
    {
      "entropy": 6.4154664993286135,
      "epoch": 0.06146525230494696,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004645,
      "loss": 6.3442,
      "mean_token_accuracy": 0.10482841059565544,
      "num_tokens": 2029210.0,
      "step": 930
    },
    {
      "entropy": 6.495858383178711,
      "epoch": 0.061795710650672485,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000467,
      "loss": 6.4541,
      "mean_token_accuracy": 0.10356884673237801,
      "num_tokens": 2039131.0,
      "step": 935
    },
    {
      "entropy": 6.439481210708618,
      "epoch": 0.062126168996398,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004695,
      "loss": 6.5025,
      "mean_token_accuracy": 0.10553577691316604,
      "num_tokens": 2049865.0,
      "step": 940
    },
    {
      "entropy": 6.535756683349609,
      "epoch": 0.062456627342123525,
      "grad_norm": 1.234375,
      "learning_rate": 0.000472,
      "loss": 6.4931,
      "mean_token_accuracy": 0.09843265414237976,
      "num_tokens": 2061151.0,
      "step": 945
    },
    {
      "entropy": 6.403180408477783,
      "epoch": 0.06278708568784905,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004745,
      "loss": 6.3533,
      "mean_token_accuracy": 0.10208684876561165,
      "num_tokens": 2071492.0,
      "step": 950
    },
    {
      "entropy": 6.472384834289551,
      "epoch": 0.06311754403357457,
      "grad_norm": 1.078125,
      "learning_rate": 0.000477,
      "loss": 6.4064,
      "mean_token_accuracy": 0.10374934822320939,
      "num_tokens": 2083291.0,
      "step": 955
    },
    {
      "entropy": 6.539694118499756,
      "epoch": 0.0634480023793001,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004795,
      "loss": 6.5599,
      "mean_token_accuracy": 0.10300241261720658,
      "num_tokens": 2094515.0,
      "step": 960
    },
    {
      "entropy": 6.558036661148071,
      "epoch": 0.0637784607250256,
      "grad_norm": 1.109375,
      "learning_rate": 0.000482,
      "loss": 6.4193,
      "mean_token_accuracy": 0.10273518785834312,
      "num_tokens": 2104774.0,
      "step": 965
    },
    {
      "entropy": 6.342342042922974,
      "epoch": 0.06410891907075113,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004845,
      "loss": 6.3231,
      "mean_token_accuracy": 0.11382870972156525,
      "num_tokens": 2114751.0,
      "step": 970
    },
    {
      "entropy": 6.497040271759033,
      "epoch": 0.06443937741647665,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000487,
      "loss": 6.3843,
      "mean_token_accuracy": 0.10335814580321312,
      "num_tokens": 2126815.0,
      "step": 975
    },
    {
      "entropy": 6.449503803253174,
      "epoch": 0.06476983576220217,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004895,
      "loss": 6.4582,
      "mean_token_accuracy": 0.09764749184250832,
      "num_tokens": 2137951.0,
      "step": 980
    },
    {
      "entropy": 6.487096357345581,
      "epoch": 0.0651002941079277,
      "grad_norm": 1.03125,
      "learning_rate": 0.000492,
      "loss": 6.5674,
      "mean_token_accuracy": 0.1011856272816658,
      "num_tokens": 2149291.0,
      "step": 985
    },
    {
      "entropy": 6.517102432250977,
      "epoch": 0.06543075245365322,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004945,
      "loss": 6.5642,
      "mean_token_accuracy": 0.09843995049595833,
      "num_tokens": 2161351.0,
      "step": 990
    },
    {
      "entropy": 6.437722778320312,
      "epoch": 0.06576121079937874,
      "grad_norm": 1.0,
      "learning_rate": 0.000497,
      "loss": 6.3354,
      "mean_token_accuracy": 0.10348827317357064,
      "num_tokens": 2172832.0,
      "step": 995
    },
    {
      "entropy": 6.447791671752929,
      "epoch": 0.06609166914510425,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995,
      "loss": 6.4564,
      "mean_token_accuracy": 0.1005012884736061,
      "num_tokens": 2183151.0,
      "step": 1000
    }
  ],
  "logging_steps": 5,
  "max_steps": 4000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 3612452161536000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}