eus-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407/checkpoint-11500/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.880103137086378,
  "eval_steps": 500,
  "global_step": 11500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 4.746951723098755,
      "epoch": 0.004297378599054577,
      "grad_norm": 13.0,
      "learning_rate": 2e-06,
      "loss": 14.3802,
      "mean_token_accuracy": 0.0,
      "num_tokens": 10107.0,
      "step": 5
    },
    {
      "entropy": 4.743826389312744,
      "epoch": 0.008594757198109154,
      "grad_norm": 14.125,
      "learning_rate": 4.5e-06,
      "loss": 14.1741,
      "mean_token_accuracy": 0.00023007302079349756,
      "num_tokens": 18391.0,
      "step": 10
    },
    {
      "entropy": 4.808784341812133,
      "epoch": 0.01289213579716373,
      "grad_norm": 17.5,
      "learning_rate": 7e-06,
      "loss": 13.9391,
      "mean_token_accuracy": 0.0,
      "num_tokens": 27061.0,
      "step": 15
    },
    {
      "entropy": 5.067050266265869,
      "epoch": 0.017189514396218308,
      "grad_norm": 28.5,
      "learning_rate": 9.5e-06,
      "loss": 13.3774,
      "mean_token_accuracy": 0.0,
      "num_tokens": 36339.0,
      "step": 20
    },
    {
      "entropy": 7.5024512767791744,
      "epoch": 0.021486892995272882,
      "grad_norm": 19.5,
      "learning_rate": 1.2e-05,
      "loss": 11.6591,
      "mean_token_accuracy": 0.00038444948149845003,
      "num_tokens": 45770.0,
      "step": 25
    },
    {
      "entropy": 10.366794681549072,
      "epoch": 0.02578427159432746,
      "grad_norm": 4.0625,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 10.7655,
      "mean_token_accuracy": 0.00122114981058985,
      "num_tokens": 54575.0,
      "step": 30
    },
    {
      "entropy": 10.686303997039795,
      "epoch": 0.030081650193382038,
      "grad_norm": 3.09375,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.5273,
      "mean_token_accuracy": 0.03648172654211521,
      "num_tokens": 66403.0,
      "step": 35
    },
    {
      "entropy": 10.672397136688232,
      "epoch": 0.034379028792436615,
      "grad_norm": 2.640625,
      "learning_rate": 1.95e-05,
      "loss": 10.1061,
      "mean_token_accuracy": 0.058901159465312956,
      "num_tokens": 76510.0,
      "step": 40
    },
    {
      "entropy": 10.515008449554443,
      "epoch": 0.03867640739149119,
      "grad_norm": 2.296875,
      "learning_rate": 2.2e-05,
      "loss": 9.8287,
      "mean_token_accuracy": 0.05614017099142075,
      "num_tokens": 84836.0,
      "step": 45
    },
    {
      "entropy": 10.411137008666993,
      "epoch": 0.042973785990545764,
      "grad_norm": 1.84375,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.6563,
      "mean_token_accuracy": 0.05396632887423038,
      "num_tokens": 93197.0,
      "step": 50
    },
    {
      "entropy": 10.430077075958252,
      "epoch": 0.047271164589600345,
      "grad_norm": 2.40625,
      "learning_rate": 2.7e-05,
      "loss": 9.5344,
      "mean_token_accuracy": 0.055140935629606244,
      "num_tokens": 101546.0,
      "step": 55
    },
    {
      "entropy": 10.39182996749878,
      "epoch": 0.05156854318865492,
      "grad_norm": 1.8203125,
      "learning_rate": 2.95e-05,
      "loss": 9.5444,
      "mean_token_accuracy": 0.05580790489912033,
      "num_tokens": 111703.0,
      "step": 60
    },
    {
      "entropy": 10.338463401794433,
      "epoch": 0.055865921787709494,
      "grad_norm": 1.7734375,
      "learning_rate": 3.2e-05,
      "loss": 9.4004,
      "mean_token_accuracy": 0.0589165486395359,
      "num_tokens": 119894.0,
      "step": 65
    },
    {
      "entropy": 10.395281219482422,
      "epoch": 0.060163300386764075,
      "grad_norm": 1.921875,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.321,
      "mean_token_accuracy": 0.0635790053755045,
      "num_tokens": 128885.0,
      "step": 70
    },
    {
      "entropy": 10.310928249359131,
      "epoch": 0.06446067898581866,
      "grad_norm": 1.7109375,
      "learning_rate": 3.7e-05,
      "loss": 9.2708,
      "mean_token_accuracy": 0.05928411558270454,
      "num_tokens": 138106.0,
      "step": 75
    },
    {
      "entropy": 10.260056686401366,
      "epoch": 0.06875805758487323,
      "grad_norm": 1.65625,
      "learning_rate": 3.95e-05,
      "loss": 9.2086,
      "mean_token_accuracy": 0.0639796681702137,
      "num_tokens": 146691.0,
      "step": 80
    },
    {
      "entropy": 10.25273675918579,
      "epoch": 0.0730554361839278,
      "grad_norm": 1.671875,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.1145,
      "mean_token_accuracy": 0.07113314382731914,
      "num_tokens": 155792.0,
      "step": 85
    },
    {
      "entropy": 10.214042377471923,
      "epoch": 0.07735281478298238,
      "grad_norm": 1.5234375,
      "learning_rate": 4.45e-05,
      "loss": 9.1291,
      "mean_token_accuracy": 0.0655292171984911,
      "num_tokens": 166944.0,
      "step": 90
    },
    {
      "entropy": 10.106289291381836,
      "epoch": 0.08165019338203695,
      "grad_norm": 1.9296875,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.9072,
      "mean_token_accuracy": 0.0739534068852663,
      "num_tokens": 175303.0,
      "step": 95
    },
    {
      "entropy": 10.038150310516357,
      "epoch": 0.08594757198109153,
      "grad_norm": 1.4375,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.8898,
      "mean_token_accuracy": 0.0677933320403099,
      "num_tokens": 184708.0,
      "step": 100
    },
    {
      "entropy": 10.01686487197876,
      "epoch": 0.09024495058014612,
      "grad_norm": 1.46875,
      "learning_rate": 5.2e-05,
      "loss": 8.7892,
      "mean_token_accuracy": 0.06626052372157573,
      "num_tokens": 193835.0,
      "step": 105
    },
    {
      "entropy": 10.002698516845703,
      "epoch": 0.09454232917920069,
      "grad_norm": 1.546875,
      "learning_rate": 5.45e-05,
      "loss": 8.6916,
      "mean_token_accuracy": 0.07468680031597615,
      "num_tokens": 203344.0,
      "step": 110
    },
    {
      "entropy": 9.896427822113036,
      "epoch": 0.09883970777825526,
      "grad_norm": 1.2578125,
      "learning_rate": 5.7e-05,
      "loss": 8.6808,
      "mean_token_accuracy": 0.06314092017710209,
      "num_tokens": 213048.0,
      "step": 115
    },
    {
      "entropy": 9.836742782592774,
      "epoch": 0.10313708637730984,
      "grad_norm": 1.265625,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.5044,
      "mean_token_accuracy": 0.0604202140122652,
      "num_tokens": 221784.0,
      "step": 120
    },
    {
      "entropy": 9.552239894866943,
      "epoch": 0.10743446497636441,
      "grad_norm": 1.1484375,
      "learning_rate": 6.2e-05,
      "loss": 8.3827,
      "mean_token_accuracy": 0.07485125958919525,
      "num_tokens": 230971.0,
      "step": 125
    },
    {
      "entropy": 9.533222961425782,
      "epoch": 0.11173184357541899,
      "grad_norm": 1.1484375,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.2984,
      "mean_token_accuracy": 0.06787660419940948,
      "num_tokens": 240524.0,
      "step": 130
    },
    {
      "entropy": 9.320106983184814,
      "epoch": 0.11602922217447358,
      "grad_norm": 1.21875,
      "learning_rate": 6.7e-05,
      "loss": 8.2114,
      "mean_token_accuracy": 0.07236036621034145,
      "num_tokens": 249220.0,
      "step": 135
    },
    {
      "entropy": 9.246935367584229,
      "epoch": 0.12032660077352815,
      "grad_norm": 1.1640625,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.205,
      "mean_token_accuracy": 0.06581656485795975,
      "num_tokens": 258934.0,
      "step": 140
    },
    {
      "entropy": 9.093988609313964,
      "epoch": 0.12462397937258272,
      "grad_norm": 1.375,
      "learning_rate": 7.2e-05,
      "loss": 8.1081,
      "mean_token_accuracy": 0.07143516950309277,
      "num_tokens": 267680.0,
      "step": 145
    },
    {
      "entropy": 8.769672107696532,
      "epoch": 0.1289213579716373,
      "grad_norm": 1.1171875,
      "learning_rate": 7.45e-05,
      "loss": 8.0668,
      "mean_token_accuracy": 0.07136639766395092,
      "num_tokens": 276227.0,
      "step": 150
    },
    {
      "entropy": 8.74085454940796,
      "epoch": 0.1332187365706919,
      "grad_norm": 1.0390625,
      "learning_rate": 7.7e-05,
      "loss": 7.9531,
      "mean_token_accuracy": 0.07765112966299056,
      "num_tokens": 286342.0,
      "step": 155
    },
    {
      "entropy": 8.63883180618286,
      "epoch": 0.13751611516974646,
      "grad_norm": 0.91015625,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.9755,
      "mean_token_accuracy": 0.06722700893878937,
      "num_tokens": 294994.0,
      "step": 160
    },
    {
      "entropy": 8.51505479812622,
      "epoch": 0.14181349376880104,
      "grad_norm": 0.92578125,
      "learning_rate": 8.2e-05,
      "loss": 7.9472,
      "mean_token_accuracy": 0.07361817210912705,
      "num_tokens": 303882.0,
      "step": 165
    },
    {
      "entropy": 8.39471378326416,
      "epoch": 0.1461108723678556,
      "grad_norm": 0.78515625,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.9246,
      "mean_token_accuracy": 0.07520690858364106,
      "num_tokens": 312515.0,
      "step": 170
    },
    {
      "entropy": 8.389386463165284,
      "epoch": 0.15040825096691018,
      "grad_norm": 1.0703125,
      "learning_rate": 8.7e-05,
      "loss": 7.9233,
      "mean_token_accuracy": 0.07540233507752418,
      "num_tokens": 320801.0,
      "step": 175
    },
    {
      "entropy": 8.254477882385254,
      "epoch": 0.15470562956596476,
      "grad_norm": 1.140625,
      "learning_rate": 8.95e-05,
      "loss": 7.9058,
      "mean_token_accuracy": 0.08120642676949501,
      "num_tokens": 329382.0,
      "step": 180
    },
    {
      "entropy": 8.332880115509033,
      "epoch": 0.15900300816501933,
      "grad_norm": 0.9296875,
      "learning_rate": 9.2e-05,
      "loss": 7.8804,
      "mean_token_accuracy": 0.07222829759120941,
      "num_tokens": 337894.0,
      "step": 185
    },
    {
      "entropy": 8.286204624176026,
      "epoch": 0.1633003867640739,
      "grad_norm": 1.21875,
      "learning_rate": 9.45e-05,
      "loss": 7.8675,
      "mean_token_accuracy": 0.0751982007175684,
      "num_tokens": 346380.0,
      "step": 190
    },
    {
      "entropy": 8.191093826293946,
      "epoch": 0.16759776536312848,
      "grad_norm": 0.9609375,
      "learning_rate": 9.7e-05,
      "loss": 7.9246,
      "mean_token_accuracy": 0.0726242907345295,
      "num_tokens": 356305.0,
      "step": 195
    },
    {
      "entropy": 8.269799137115479,
      "epoch": 0.17189514396218306,
      "grad_norm": 1.0859375,
      "learning_rate": 9.95e-05,
      "loss": 7.8754,
      "mean_token_accuracy": 0.07030960507690906,
      "num_tokens": 364899.0,
      "step": 200
    },
    {
      "entropy": 8.152498531341553,
      "epoch": 0.17619252256123766,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000102,
      "loss": 7.8478,
      "mean_token_accuracy": 0.07710132971405984,
      "num_tokens": 373663.0,
      "step": 205
    },
    {
      "entropy": 8.213716077804566,
      "epoch": 0.18048990116029223,
      "grad_norm": 1.3125,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.8339,
      "mean_token_accuracy": 0.07559528909623622,
      "num_tokens": 382730.0,
      "step": 210
    },
    {
      "entropy": 8.200652503967286,
      "epoch": 0.1847872797593468,
      "grad_norm": 2.3125,
      "learning_rate": 0.000107,
      "loss": 7.8751,
      "mean_token_accuracy": 0.06925368532538415,
      "num_tokens": 392676.0,
      "step": 215
    },
    {
      "entropy": 7.960573291778564,
      "epoch": 0.18908465835840138,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0001095,
      "loss": 7.8586,
      "mean_token_accuracy": 0.08359932452440262,
      "num_tokens": 401050.0,
      "step": 220
    },
    {
      "entropy": 8.224654483795167,
      "epoch": 0.19338203695745596,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000112,
      "loss": 7.7944,
      "mean_token_accuracy": 0.0739112664014101,
      "num_tokens": 410009.0,
      "step": 225
    },
    {
      "entropy": 8.076114654541016,
      "epoch": 0.19767941555651053,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0001145,
      "loss": 7.784,
      "mean_token_accuracy": 0.07774697132408619,
      "num_tokens": 419302.0,
      "step": 230
    },
    {
      "entropy": 8.134877395629882,
      "epoch": 0.2019767941555651,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.7035,
      "mean_token_accuracy": 0.0846225418150425,
      "num_tokens": 427296.0,
      "step": 235
    },
    {
      "entropy": 8.007131862640382,
      "epoch": 0.20627417275461968,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.8318,
      "mean_token_accuracy": 0.07436586394906045,
      "num_tokens": 436368.0,
      "step": 240
    },
    {
      "entropy": 8.062287092208862,
      "epoch": 0.21057155135367425,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000122,
      "loss": 7.7528,
      "mean_token_accuracy": 0.0802234098315239,
      "num_tokens": 445535.0,
      "step": 245
    },
    {
      "entropy": 8.11290283203125,
      "epoch": 0.21486892995272883,
      "grad_norm": 0.921875,
      "learning_rate": 0.0001245,
      "loss": 7.7292,
      "mean_token_accuracy": 0.07871693558990955,
      "num_tokens": 454769.0,
      "step": 250
    },
    {
      "entropy": 8.053960275650024,
      "epoch": 0.2191663085517834,
      "grad_norm": 1.15625,
      "learning_rate": 0.000127,
      "loss": 7.7547,
      "mean_token_accuracy": 0.08017503693699837,
      "num_tokens": 463975.0,
      "step": 255
    },
    {
      "entropy": 7.906766176223755,
      "epoch": 0.22346368715083798,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001295,
      "loss": 7.7438,
      "mean_token_accuracy": 0.08368083834648132,
      "num_tokens": 472899.0,
      "step": 260
    },
    {
      "entropy": 8.039681100845337,
      "epoch": 0.22776106574989258,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000132,
      "loss": 7.7418,
      "mean_token_accuracy": 0.07471806854009629,
      "num_tokens": 481556.0,
      "step": 265
    },
    {
      "entropy": 8.029216766357422,
      "epoch": 0.23205844434894715,
      "grad_norm": 1.375,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.7123,
      "mean_token_accuracy": 0.08060996606945992,
      "num_tokens": 490253.0,
      "step": 270
    },
    {
      "entropy": 8.084418487548827,
      "epoch": 0.23635582294800173,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.7157,
      "mean_token_accuracy": 0.07992063909769058,
      "num_tokens": 498444.0,
      "step": 275
    },
    {
      "entropy": 7.9231773853302006,
      "epoch": 0.2406532015470563,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001395,
      "loss": 7.664,
      "mean_token_accuracy": 0.08332347348332406,
      "num_tokens": 508330.0,
      "step": 280
    },
    {
      "entropy": 7.949900054931641,
      "epoch": 0.24495058014611087,
      "grad_norm": 1.203125,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.7586,
      "mean_token_accuracy": 0.0789482720196247,
      "num_tokens": 517900.0,
      "step": 285
    },
    {
      "entropy": 7.986289691925049,
      "epoch": 0.24924795874516545,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0001445,
      "loss": 7.7176,
      "mean_token_accuracy": 0.07970602363348007,
      "num_tokens": 527808.0,
      "step": 290
    },
    {
      "entropy": 7.944300651550293,
      "epoch": 0.25354533734422,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000147,
      "loss": 7.7025,
      "mean_token_accuracy": 0.074923125654459,
      "num_tokens": 536931.0,
      "step": 295
    },
    {
      "entropy": 8.035796642303467,
      "epoch": 0.2578427159432746,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001495,
      "loss": 7.7166,
      "mean_token_accuracy": 0.08002732172608376,
      "num_tokens": 545758.0,
      "step": 300
    },
    {
      "entropy": 7.967339134216308,
      "epoch": 0.26214009454232917,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000152,
      "loss": 7.735,
      "mean_token_accuracy": 0.08182198479771614,
      "num_tokens": 555165.0,
      "step": 305
    },
    {
      "entropy": 7.954096126556396,
      "epoch": 0.2664374731413838,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.6505,
      "mean_token_accuracy": 0.08079984113574028,
      "num_tokens": 564719.0,
      "step": 310
    },
    {
      "entropy": 8.011483335494995,
      "epoch": 0.2707348517404383,
      "grad_norm": 1.59375,
      "learning_rate": 0.000157,
      "loss": 7.6203,
      "mean_token_accuracy": 0.0855506956577301,
      "num_tokens": 573572.0,
      "step": 315
    },
    {
      "entropy": 7.915408372879028,
      "epoch": 0.2750322303394929,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001595,
      "loss": 7.7815,
      "mean_token_accuracy": 0.07522618100047111,
      "num_tokens": 581497.0,
      "step": 320
    },
    {
      "entropy": 7.911842298507691,
      "epoch": 0.27932960893854747,
      "grad_norm": 1.28125,
      "learning_rate": 0.000162,
      "loss": 7.6282,
      "mean_token_accuracy": 0.08246761485934258,
      "num_tokens": 591107.0,
      "step": 325
    },
    {
      "entropy": 7.908317279815674,
      "epoch": 0.28362698753760207,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.6727,
      "mean_token_accuracy": 0.07925951480865479,
      "num_tokens": 600241.0,
      "step": 330
    },
    {
      "entropy": 7.992547035217285,
      "epoch": 0.2879243661366566,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.5985,
      "mean_token_accuracy": 0.08225269913673401,
      "num_tokens": 608697.0,
      "step": 335
    },
    {
      "entropy": 7.899400424957276,
      "epoch": 0.2922217447357112,
      "grad_norm": 1.078125,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.5755,
      "mean_token_accuracy": 0.08702421039342881,
      "num_tokens": 617275.0,
      "step": 340
    },
    {
      "entropy": 7.919580078125,
      "epoch": 0.29651912333476577,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.6679,
      "mean_token_accuracy": 0.08152761459350585,
      "num_tokens": 626644.0,
      "step": 345
    },
    {
      "entropy": 7.855055665969848,
      "epoch": 0.30081650193382037,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.5941,
      "mean_token_accuracy": 0.08607768341898918,
      "num_tokens": 635110.0,
      "step": 350
    },
    {
      "entropy": 7.904864168167114,
      "epoch": 0.30511388053287497,
      "grad_norm": 1.140625,
      "learning_rate": 0.000177,
      "loss": 7.731,
      "mean_token_accuracy": 0.08116863258183002,
      "num_tokens": 644746.0,
      "step": 355
    },
    {
      "entropy": 8.003858804702759,
      "epoch": 0.3094112591319295,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001795,
      "loss": 7.615,
      "mean_token_accuracy": 0.08461246490478516,
      "num_tokens": 654281.0,
      "step": 360
    },
    {
      "entropy": 7.807305002212525,
      "epoch": 0.3137086377309841,
      "grad_norm": 1.109375,
      "learning_rate": 0.000182,
      "loss": 7.6127,
      "mean_token_accuracy": 0.08584560602903366,
      "num_tokens": 663174.0,
      "step": 365
    },
    {
      "entropy": 7.868645715713501,
      "epoch": 0.31800601633003867,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001845,
      "loss": 7.5401,
      "mean_token_accuracy": 0.08367656841874123,
      "num_tokens": 672178.0,
      "step": 370
    },
    {
      "entropy": 7.838464975357056,
      "epoch": 0.32230339492909327,
      "grad_norm": 1.09375,
      "learning_rate": 0.000187,
      "loss": 7.5441,
      "mean_token_accuracy": 0.08873862400650978,
      "num_tokens": 681323.0,
      "step": 375
    },
    {
      "entropy": 7.861468744277954,
      "epoch": 0.3266007735281478,
      "grad_norm": 1.015625,
      "learning_rate": 0.0001895,
      "loss": 7.4599,
      "mean_token_accuracy": 0.08782187923789024,
      "num_tokens": 690461.0,
      "step": 380
    },
    {
      "entropy": 7.801782035827637,
      "epoch": 0.3308981521272024,
      "grad_norm": 1.9375,
      "learning_rate": 0.000192,
      "loss": 7.5001,
      "mean_token_accuracy": 0.08662289157509803,
      "num_tokens": 699199.0,
      "step": 385
    },
    {
      "entropy": 7.801443576812744,
      "epoch": 0.33519553072625696,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001945,
      "loss": 7.4921,
      "mean_token_accuracy": 0.09190349280834198,
      "num_tokens": 707949.0,
      "step": 390
    },
    {
      "entropy": 7.831652879714966,
      "epoch": 0.33949290932531156,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.5308,
      "mean_token_accuracy": 0.08901702687144279,
      "num_tokens": 715752.0,
      "step": 395
    },
    {
      "entropy": 7.8149333000183105,
      "epoch": 0.3437902879243661,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.4733,
      "mean_token_accuracy": 0.08961250334978103,
      "num_tokens": 724416.0,
      "step": 400
    },
    {
      "entropy": 7.810607194900513,
      "epoch": 0.3480876665234207,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000202,
      "loss": 7.5157,
      "mean_token_accuracy": 0.08556607961654664,
      "num_tokens": 733116.0,
      "step": 405
    },
    {
      "entropy": 7.813943815231323,
      "epoch": 0.3523850451224753,
      "grad_norm": 1.203125,
      "learning_rate": 0.00020449999999999998,
      "loss": 7.4905,
      "mean_token_accuracy": 0.08932617083191871,
      "num_tokens": 742093.0,
      "step": 410
    },
    {
      "entropy": 7.833597850799561,
      "epoch": 0.35668242372152986,
      "grad_norm": 1.078125,
      "learning_rate": 0.000207,
      "loss": 7.4856,
      "mean_token_accuracy": 0.08866851031780243,
      "num_tokens": 750402.0,
      "step": 415
    },
    {
      "entropy": 7.651254415512085,
      "epoch": 0.36097980232058446,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002095,
      "loss": 7.4363,
      "mean_token_accuracy": 0.09088600948452949,
      "num_tokens": 760961.0,
      "step": 420
    },
    {
      "entropy": 7.848704862594604,
      "epoch": 0.365277180919639,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000212,
      "loss": 7.5138,
      "mean_token_accuracy": 0.08746314346790314,
      "num_tokens": 770554.0,
      "step": 425
    },
    {
      "entropy": 7.62291374206543,
      "epoch": 0.3695745595186936,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002145,
      "loss": 7.4938,
      "mean_token_accuracy": 0.09328901097178459,
      "num_tokens": 779172.0,
      "step": 430
    },
    {
      "entropy": 7.798684120178223,
      "epoch": 0.37387193811774816,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021700000000000002,
      "loss": 7.5411,
      "mean_token_accuracy": 0.0797388345003128,
      "num_tokens": 788040.0,
      "step": 435
    },
    {
      "entropy": 7.799862670898437,
      "epoch": 0.37816931671680276,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0002195,
      "loss": 7.4367,
      "mean_token_accuracy": 0.09049512967467307,
      "num_tokens": 796786.0,
      "step": 440
    },
    {
      "entropy": 7.653696584701538,
      "epoch": 0.3824666953158573,
      "grad_norm": 1.21875,
      "learning_rate": 0.000222,
      "loss": 7.4549,
      "mean_token_accuracy": 0.08962092027068139,
      "num_tokens": 805520.0,
      "step": 445
    },
    {
      "entropy": 7.74529128074646,
      "epoch": 0.3867640739149119,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002245,
      "loss": 7.3698,
      "mean_token_accuracy": 0.09231138080358506,
      "num_tokens": 814939.0,
      "step": 450
    },
    {
      "entropy": 7.6202319145202635,
      "epoch": 0.39106145251396646,
      "grad_norm": 1.171875,
      "learning_rate": 0.00022700000000000002,
      "loss": 7.3903,
      "mean_token_accuracy": 0.08881158530712127,
      "num_tokens": 823862.0,
      "step": 455
    },
    {
      "entropy": 7.745624685287476,
      "epoch": 0.39535883111302106,
      "grad_norm": 1.140625,
      "learning_rate": 0.00022950000000000002,
      "loss": 7.3957,
      "mean_token_accuracy": 0.09272652640938758,
      "num_tokens": 832820.0,
      "step": 460
    },
    {
      "entropy": 7.745703077316284,
      "epoch": 0.39965620971207566,
      "grad_norm": 1.125,
      "learning_rate": 0.00023200000000000003,
      "loss": 7.3724,
      "mean_token_accuracy": 0.09544146358966828,
      "num_tokens": 841538.0,
      "step": 465
    },
    {
      "entropy": 7.645022964477539,
      "epoch": 0.4039535883111302,
      "grad_norm": 1.046875,
      "learning_rate": 0.00023449999999999998,
      "loss": 7.4799,
      "mean_token_accuracy": 0.08852436989545823,
      "num_tokens": 851123.0,
      "step": 470
    },
    {
      "entropy": 7.723429012298584,
      "epoch": 0.4082509669101848,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000237,
      "loss": 7.4569,
      "mean_token_accuracy": 0.09157502874732018,
      "num_tokens": 860357.0,
      "step": 475
    },
    {
      "entropy": 7.702331590652466,
      "epoch": 0.41254834550923936,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002395,
      "loss": 7.3553,
      "mean_token_accuracy": 0.09141395911574364,
      "num_tokens": 869980.0,
      "step": 480
    },
    {
      "entropy": 7.594449472427368,
      "epoch": 0.41684572410829396,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000242,
      "loss": 7.3152,
      "mean_token_accuracy": 0.09529538601636886,
      "num_tokens": 878250.0,
      "step": 485
    },
    {
      "entropy": 7.647209882736206,
      "epoch": 0.4211431027073485,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0002445,
      "loss": 7.432,
      "mean_token_accuracy": 0.08998423144221306,
      "num_tokens": 887624.0,
      "step": 490
    },
    {
      "entropy": 7.620798921585083,
      "epoch": 0.4254404813064031,
      "grad_norm": 1.15625,
      "learning_rate": 0.000247,
      "loss": 7.3811,
      "mean_token_accuracy": 0.09268759191036224,
      "num_tokens": 897120.0,
      "step": 495
    },
    {
      "entropy": 7.560871362686157,
      "epoch": 0.42973785990545765,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002495,
      "loss": 7.3499,
      "mean_token_accuracy": 0.09775018766522407,
      "num_tokens": 906215.0,
      "step": 500
    },
    {
      "epoch": 0.42973785990545765,
      "eval_entropy": 7.452849494444357,
      "eval_loss": 7.401772499084473,
      "eval_mean_token_accuracy": 0.09653489479610512,
      "eval_num_tokens": 906215.0,
      "eval_runtime": 2.6878,
      "eval_samples_per_second": 1320.408,
      "eval_steps_per_second": 165.191,
      "step": 500
    },
    {
      "entropy": 7.597157382965088,
      "epoch": 0.43403523850451226,
      "grad_norm": 1.140625,
      "learning_rate": 0.000252,
      "loss": 7.3332,
      "mean_token_accuracy": 0.09525968879461288,
      "num_tokens": 915181.0,
      "step": 505
    },
    {
      "entropy": 7.594379711151123,
      "epoch": 0.4383326171035668,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002545,
      "loss": 7.3459,
      "mean_token_accuracy": 0.09548510313034057,
      "num_tokens": 924377.0,
      "step": 510
    },
    {
      "entropy": 7.679618883132934,
      "epoch": 0.4426299957026214,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000257,
      "loss": 7.4198,
      "mean_token_accuracy": 0.09123436212539673,
      "num_tokens": 933114.0,
      "step": 515
    },
    {
      "entropy": 7.633022212982178,
      "epoch": 0.44692737430167595,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002595,
      "loss": 7.3843,
      "mean_token_accuracy": 0.09295491799712181,
      "num_tokens": 943306.0,
      "step": 520
    },
    {
      "entropy": 7.551965427398682,
      "epoch": 0.45122475290073055,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000262,
      "loss": 7.2733,
      "mean_token_accuracy": 0.09815134108066559,
      "num_tokens": 951515.0,
      "step": 525
    },
    {
      "entropy": 7.5277495861053465,
      "epoch": 0.45552213149978515,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026450000000000003,
      "loss": 7.4038,
      "mean_token_accuracy": 0.09049785882234573,
      "num_tokens": 962686.0,
      "step": 530
    },
    {
      "entropy": 7.628147220611572,
      "epoch": 0.4598195100988397,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026700000000000004,
      "loss": 7.398,
      "mean_token_accuracy": 0.0952046200633049,
      "num_tokens": 972136.0,
      "step": 535
    },
    {
      "entropy": 7.478239583969116,
      "epoch": 0.4641168886978943,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026950000000000005,
      "loss": 7.2984,
      "mean_token_accuracy": 0.09473967850208283,
      "num_tokens": 981301.0,
      "step": 540
    },
    {
      "entropy": 7.588911247253418,
      "epoch": 0.46841426729694885,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00027200000000000005,
      "loss": 7.3291,
      "mean_token_accuracy": 0.09194630309939385,
      "num_tokens": 990360.0,
      "step": 545
    },
    {
      "entropy": 7.613106918334961,
      "epoch": 0.47271164589600345,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002745,
      "loss": 7.3639,
      "mean_token_accuracy": 0.09320687279105186,
      "num_tokens": 999415.0,
      "step": 550
    },
    {
      "entropy": 7.489257049560547,
      "epoch": 0.477009024495058,
      "grad_norm": 1.15625,
      "learning_rate": 0.000277,
      "loss": 7.3081,
      "mean_token_accuracy": 0.09810379892587662,
      "num_tokens": 1008762.0,
      "step": 555
    },
    {
      "entropy": 7.5401557922363285,
      "epoch": 0.4813064030941126,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002795,
      "loss": 7.2954,
      "mean_token_accuracy": 0.0994754634797573,
      "num_tokens": 1017704.0,
      "step": 560
    },
    {
      "entropy": 7.514191961288452,
      "epoch": 0.48560378169316715,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028199999999999997,
      "loss": 7.2372,
      "mean_token_accuracy": 0.09821939915418625,
      "num_tokens": 1026251.0,
      "step": 565
    },
    {
      "entropy": 7.479329919815063,
      "epoch": 0.48990116029222175,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002845,
      "loss": 7.2328,
      "mean_token_accuracy": 0.09923249781131745,
      "num_tokens": 1036191.0,
      "step": 570
    },
    {
      "entropy": 7.463119411468506,
      "epoch": 0.4941985388912763,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000287,
      "loss": 7.265,
      "mean_token_accuracy": 0.10023235529661179,
      "num_tokens": 1044936.0,
      "step": 575
    },
    {
      "entropy": 7.498543548583984,
      "epoch": 0.4984959174903309,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002895,
      "loss": 7.2826,
      "mean_token_accuracy": 0.09519179984927177,
      "num_tokens": 1053683.0,
      "step": 580
    },
    {
      "entropy": 7.506382083892822,
      "epoch": 0.5027932960893855,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000292,
      "loss": 7.258,
      "mean_token_accuracy": 0.1035758949816227,
      "num_tokens": 1062932.0,
      "step": 585
    },
    {
      "entropy": 7.397946691513061,
      "epoch": 0.50709067468844,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002945,
      "loss": 7.2631,
      "mean_token_accuracy": 0.09526635929942132,
      "num_tokens": 1072313.0,
      "step": 590
    },
    {
      "entropy": 7.519252490997315,
      "epoch": 0.5113880532874946,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000297,
      "loss": 7.2487,
      "mean_token_accuracy": 0.09590291604399681,
      "num_tokens": 1081675.0,
      "step": 595
    },
    {
      "entropy": 7.414077806472778,
      "epoch": 0.5156854318865493,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002995,
      "loss": 7.2657,
      "mean_token_accuracy": 0.09974563643336296,
      "num_tokens": 1091541.0,
      "step": 600
    },
    {
      "entropy": 7.551970100402832,
      "epoch": 0.5199828104856038,
      "grad_norm": 1.140625,
      "learning_rate": 0.000302,
      "loss": 7.2713,
      "mean_token_accuracy": 0.09822080582380295,
      "num_tokens": 1100724.0,
      "step": 605
    },
    {
      "entropy": 7.332079648971558,
      "epoch": 0.5242801890846583,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0003045,
      "loss": 7.2355,
      "mean_token_accuracy": 0.09697570949792862,
      "num_tokens": 1108869.0,
      "step": 610
    },
    {
      "entropy": 7.480484294891357,
      "epoch": 0.5285775676837129,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000307,
      "loss": 7.25,
      "mean_token_accuracy": 0.09792520478367805,
      "num_tokens": 1117314.0,
      "step": 615
    },
    {
      "entropy": 7.362159061431885,
      "epoch": 0.5328749462827675,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003095,
      "loss": 7.1859,
      "mean_token_accuracy": 0.10019350796937943,
      "num_tokens": 1126786.0,
      "step": 620
    },
    {
      "entropy": 7.40272159576416,
      "epoch": 0.5371723248818221,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000312,
      "loss": 7.1913,
      "mean_token_accuracy": 0.09889030531048774,
      "num_tokens": 1136013.0,
      "step": 625
    },
    {
      "entropy": 7.388201570510864,
      "epoch": 0.5414697034808766,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003145,
      "loss": 7.2166,
      "mean_token_accuracy": 0.10013808235526085,
      "num_tokens": 1144970.0,
      "step": 630
    },
    {
      "entropy": 7.420843648910522,
      "epoch": 0.5457670820799312,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000317,
      "loss": 7.1756,
      "mean_token_accuracy": 0.10355055630207062,
      "num_tokens": 1153810.0,
      "step": 635
    },
    {
      "entropy": 7.348835706710815,
      "epoch": 0.5500644606789858,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003195,
      "loss": 7.1665,
      "mean_token_accuracy": 0.10198655724525452,
      "num_tokens": 1162498.0,
      "step": 640
    },
    {
      "entropy": 7.412497043609619,
      "epoch": 0.5543618392780404,
      "grad_norm": 1.09375,
      "learning_rate": 0.000322,
      "loss": 7.2589,
      "mean_token_accuracy": 0.09946419224143029,
      "num_tokens": 1172091.0,
      "step": 645
    },
    {
      "entropy": 7.459282636642456,
      "epoch": 0.5586592178770949,
      "grad_norm": 1.03125,
      "learning_rate": 0.00032450000000000003,
      "loss": 7.2226,
      "mean_token_accuracy": 0.10092289820313453,
      "num_tokens": 1181400.0,
      "step": 650
    },
    {
      "entropy": 7.370707035064697,
      "epoch": 0.5629565964761496,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032700000000000003,
      "loss": 7.1949,
      "mean_token_accuracy": 0.10673824399709701,
      "num_tokens": 1189780.0,
      "step": 655
    },
    {
      "entropy": 7.305479955673218,
      "epoch": 0.5672539750752041,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032950000000000004,
      "loss": 7.1683,
      "mean_token_accuracy": 0.10197578221559525,
      "num_tokens": 1198671.0,
      "step": 660
    },
    {
      "entropy": 7.374381971359253,
      "epoch": 0.5715513536742587,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033200000000000005,
      "loss": 7.1977,
      "mean_token_accuracy": 0.10073041394352913,
      "num_tokens": 1207173.0,
      "step": 665
    },
    {
      "entropy": 7.295852279663086,
      "epoch": 0.5758487322733132,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033450000000000005,
      "loss": 7.1303,
      "mean_token_accuracy": 0.10481933280825614,
      "num_tokens": 1216387.0,
      "step": 670
    },
    {
      "entropy": 7.426829862594604,
      "epoch": 0.5801461108723679,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000337,
      "loss": 7.0964,
      "mean_token_accuracy": 0.1035951815545559,
      "num_tokens": 1224461.0,
      "step": 675
    },
    {
      "entropy": 7.240889406204223,
      "epoch": 0.5844434894714224,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003395,
      "loss": 7.1706,
      "mean_token_accuracy": 0.10345663502812386,
      "num_tokens": 1233774.0,
      "step": 680
    },
    {
      "entropy": 7.37571530342102,
      "epoch": 0.588740868070477,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000342,
      "loss": 7.1333,
      "mean_token_accuracy": 0.10829071477055549,
      "num_tokens": 1242812.0,
      "step": 685
    },
    {
      "entropy": 7.373822784423828,
      "epoch": 0.5930382466695315,
      "grad_norm": 1.25,
      "learning_rate": 0.00034449999999999997,
      "loss": 7.2372,
      "mean_token_accuracy": 0.10122949331998825,
      "num_tokens": 1252872.0,
      "step": 690
    },
    {
      "entropy": 7.192572116851807,
      "epoch": 0.5973356252685862,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000347,
      "loss": 7.028,
      "mean_token_accuracy": 0.10453386306762695,
      "num_tokens": 1260852.0,
      "step": 695
    },
    {
      "entropy": 7.270504331588745,
      "epoch": 0.6016330038676407,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003495,
      "loss": 7.1215,
      "mean_token_accuracy": 0.10755191147327423,
      "num_tokens": 1268925.0,
      "step": 700
    },
    {
      "entropy": 7.358825540542602,
      "epoch": 0.6059303824666953,
      "grad_norm": 1.0625,
      "learning_rate": 0.000352,
      "loss": 7.1678,
      "mean_token_accuracy": 0.09976383447647094,
      "num_tokens": 1278994.0,
      "step": 705
    },
    {
      "entropy": 7.304379320144653,
      "epoch": 0.6102277610657499,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003545,
      "loss": 7.1582,
      "mean_token_accuracy": 0.09768114313483238,
      "num_tokens": 1287698.0,
      "step": 710
    },
    {
      "entropy": 7.309779357910156,
      "epoch": 0.6145251396648045,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000357,
      "loss": 7.1038,
      "mean_token_accuracy": 0.10614442825317383,
      "num_tokens": 1297475.0,
      "step": 715
    },
    {
      "entropy": 7.348983478546143,
      "epoch": 0.618822518263859,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003595,
      "loss": 7.1361,
      "mean_token_accuracy": 0.10617230013012886,
      "num_tokens": 1306836.0,
      "step": 720
    },
    {
      "entropy": 7.270800399780273,
      "epoch": 0.6231198968629136,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000362,
      "loss": 7.1698,
      "mean_token_accuracy": 0.09751353785395622,
      "num_tokens": 1315872.0,
      "step": 725
    },
    {
      "entropy": 7.261401176452637,
      "epoch": 0.6274172754619682,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003645,
      "loss": 7.1272,
      "mean_token_accuracy": 0.10790985673666001,
      "num_tokens": 1324624.0,
      "step": 730
    },
    {
      "entropy": 7.305949640274048,
      "epoch": 0.6317146540610228,
      "grad_norm": 1.15625,
      "learning_rate": 0.000367,
      "loss": 7.0967,
      "mean_token_accuracy": 0.1090677909553051,
      "num_tokens": 1333058.0,
      "step": 735
    },
    {
      "entropy": 7.134536504745483,
      "epoch": 0.6360120326600773,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003695,
      "loss": 7.0539,
      "mean_token_accuracy": 0.10417703315615653,
      "num_tokens": 1342376.0,
      "step": 740
    },
    {
      "entropy": 7.308330726623535,
      "epoch": 0.6403094112591319,
      "grad_norm": 1.15625,
      "learning_rate": 0.000372,
      "loss": 7.1004,
      "mean_token_accuracy": 0.099180655926466,
      "num_tokens": 1351386.0,
      "step": 745
    },
    {
      "entropy": 7.188440799713135,
      "epoch": 0.6446067898581865,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003745,
      "loss": 7.0039,
      "mean_token_accuracy": 0.11299571469426155,
      "num_tokens": 1358958.0,
      "step": 750
    },
    {
      "entropy": 7.3689507961273195,
      "epoch": 0.6489041684572411,
      "grad_norm": 1.1875,
      "learning_rate": 0.000377,
      "loss": 7.1512,
      "mean_token_accuracy": 0.1033362664282322,
      "num_tokens": 1368599.0,
      "step": 755
    },
    {
      "entropy": 7.267584180831909,
      "epoch": 0.6532015470562956,
      "grad_norm": 1.6875,
      "learning_rate": 0.0003795,
      "loss": 7.1894,
      "mean_token_accuracy": 0.10379417687654495,
      "num_tokens": 1378529.0,
      "step": 760
    },
    {
      "entropy": 7.118428087234497,
      "epoch": 0.6574989256553503,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000382,
      "loss": 6.9598,
      "mean_token_accuracy": 0.11283587664365768,
      "num_tokens": 1386993.0,
      "step": 765
    },
    {
      "entropy": 7.2359050750732425,
      "epoch": 0.6617963042544048,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003845,
      "loss": 7.0655,
      "mean_token_accuracy": 0.11571843326091766,
      "num_tokens": 1395790.0,
      "step": 770
    },
    {
      "entropy": 7.0956977844238285,
      "epoch": 0.6660936828534594,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038700000000000003,
      "loss": 7.0727,
      "mean_token_accuracy": 0.10878411009907722,
      "num_tokens": 1405587.0,
      "step": 775
    },
    {
      "entropy": 7.142232990264892,
      "epoch": 0.6703910614525139,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038950000000000003,
      "loss": 7.0347,
      "mean_token_accuracy": 0.1105166994035244,
      "num_tokens": 1414478.0,
      "step": 780
    },
    {
      "entropy": 7.294425201416016,
      "epoch": 0.6746884400515686,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039200000000000004,
      "loss": 7.0946,
      "mean_token_accuracy": 0.112546656280756,
      "num_tokens": 1423791.0,
      "step": 785
    },
    {
      "entropy": 7.22660493850708,
      "epoch": 0.6789858186506231,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039450000000000005,
      "loss": 7.1307,
      "mean_token_accuracy": 0.09776996597647666,
      "num_tokens": 1432955.0,
      "step": 790
    },
    {
      "entropy": 7.262840032577515,
      "epoch": 0.6832831972496777,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00039700000000000005,
      "loss": 7.1067,
      "mean_token_accuracy": 0.1036052331328392,
      "num_tokens": 1441907.0,
      "step": 795
    },
    {
      "entropy": 7.072665357589722,
      "epoch": 0.6875805758487322,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003995,
      "loss": 6.9795,
      "mean_token_accuracy": 0.11090076863765716,
      "num_tokens": 1451062.0,
      "step": 800
    },
    {
      "entropy": 7.22030816078186,
      "epoch": 0.6918779544477869,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000402,
      "loss": 7.0569,
      "mean_token_accuracy": 0.10773894265294075,
      "num_tokens": 1460132.0,
      "step": 805
    },
    {
      "entropy": 7.0464826107025145,
      "epoch": 0.6961753330468414,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004045,
      "loss": 6.95,
      "mean_token_accuracy": 0.11331479921936989,
      "num_tokens": 1469582.0,
      "step": 810
    },
    {
      "entropy": 7.150194263458252,
      "epoch": 0.700472711645896,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00040699999999999997,
      "loss": 7.0273,
      "mean_token_accuracy": 0.10949807912111283,
      "num_tokens": 1479053.0,
      "step": 815
    },
    {
      "entropy": 7.140777254104615,
      "epoch": 0.7047700902449506,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004095,
      "loss": 7.0444,
      "mean_token_accuracy": 0.10801303759217262,
      "num_tokens": 1488189.0,
      "step": 820
    },
    {
      "entropy": 7.048080348968506,
      "epoch": 0.7090674688440052,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000412,
      "loss": 6.909,
      "mean_token_accuracy": 0.11225773021578789,
      "num_tokens": 1497324.0,
      "step": 825
    },
    {
      "entropy": 7.119872570037842,
      "epoch": 0.7133648474430597,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004145,
      "loss": 7.0142,
      "mean_token_accuracy": 0.11574618220329284,
      "num_tokens": 1506543.0,
      "step": 830
    },
    {
      "entropy": 7.156819486618042,
      "epoch": 0.7176622260421143,
      "grad_norm": 1.15625,
      "learning_rate": 0.000417,
      "loss": 6.99,
      "mean_token_accuracy": 0.11488222405314445,
      "num_tokens": 1516737.0,
      "step": 835
    },
    {
      "entropy": 6.952379846572876,
      "epoch": 0.7219596046411689,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004195,
      "loss": 6.8673,
      "mean_token_accuracy": 0.11356691420078277,
      "num_tokens": 1525561.0,
      "step": 840
    },
    {
      "entropy": 7.108857202529907,
      "epoch": 0.7262569832402235,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000422,
      "loss": 6.8969,
      "mean_token_accuracy": 0.11771438643336296,
      "num_tokens": 1533323.0,
      "step": 845
    },
    {
      "entropy": 7.133508110046387,
      "epoch": 0.730554361839278,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004245,
      "loss": 6.9472,
      "mean_token_accuracy": 0.11780929937958717,
      "num_tokens": 1542632.0,
      "step": 850
    },
    {
      "entropy": 7.068312883377075,
      "epoch": 0.7348517404383326,
      "grad_norm": 1.296875,
      "learning_rate": 0.000427,
      "loss": 6.9049,
      "mean_token_accuracy": 0.11859052330255508,
      "num_tokens": 1551236.0,
      "step": 855
    },
    {
      "entropy": 6.930421543121338,
      "epoch": 0.7391491190373872,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004295,
      "loss": 6.9216,
      "mean_token_accuracy": 0.11472166553139687,
      "num_tokens": 1559674.0,
      "step": 860
    },
    {
      "entropy": 7.060778665542602,
      "epoch": 0.7434464976364418,
      "grad_norm": 1.125,
      "learning_rate": 0.000432,
      "loss": 6.9908,
      "mean_token_accuracy": 0.11511028632521629,
      "num_tokens": 1569481.0,
      "step": 865
    },
    {
      "entropy": 7.130613136291504,
      "epoch": 0.7477438762354963,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004345,
      "loss": 7.0149,
      "mean_token_accuracy": 0.11255836263298988,
      "num_tokens": 1578488.0,
      "step": 870
    },
    {
      "entropy": 7.021413326263428,
      "epoch": 0.752041254834551,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000437,
      "loss": 6.9125,
      "mean_token_accuracy": 0.11169529184699059,
      "num_tokens": 1586675.0,
      "step": 875
    },
    {
      "entropy": 7.099101734161377,
      "epoch": 0.7563386334336055,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004395,
      "loss": 7.0387,
      "mean_token_accuracy": 0.10673936307430268,
      "num_tokens": 1595411.0,
      "step": 880
    },
    {
      "entropy": 7.1430614471435545,
      "epoch": 0.7606360120326601,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000442,
      "loss": 7.0139,
      "mean_token_accuracy": 0.1113392397761345,
      "num_tokens": 1604046.0,
      "step": 885
    },
    {
      "entropy": 7.069469738006592,
      "epoch": 0.7649333906317146,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004445,
      "loss": 6.9978,
      "mean_token_accuracy": 0.10678019747138023,
      "num_tokens": 1613759.0,
      "step": 890
    },
    {
      "entropy": 6.96642370223999,
      "epoch": 0.7692307692307693,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000447,
      "loss": 6.9579,
      "mean_token_accuracy": 0.11460690051317216,
      "num_tokens": 1623323.0,
      "step": 895
    },
    {
      "entropy": 7.08248085975647,
      "epoch": 0.7735281478298238,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.9397,
      "mean_token_accuracy": 0.11283179298043251,
      "num_tokens": 1631727.0,
      "step": 900
    },
    {
      "entropy": 7.034572553634644,
      "epoch": 0.7778255264288784,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.8681,
      "mean_token_accuracy": 0.11410197094082833,
      "num_tokens": 1639544.0,
      "step": 905
    },
    {
      "entropy": 7.053156805038452,
      "epoch": 0.7821229050279329,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.9805,
      "mean_token_accuracy": 0.11028798148036004,
      "num_tokens": 1648931.0,
      "step": 910
    },
    {
      "entropy": 6.9188769340515135,
      "epoch": 0.7864202836269876,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.8809,
      "mean_token_accuracy": 0.11601683646440505,
      "num_tokens": 1657688.0,
      "step": 915
    },
    {
      "entropy": 7.098619890213013,
      "epoch": 0.7907176622260421,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.9528,
      "mean_token_accuracy": 0.11290125101804734,
      "num_tokens": 1666879.0,
      "step": 920
    },
    {
      "entropy": 6.971384572982788,
      "epoch": 0.7950150408250967,
      "grad_norm": 1.0625,
      "learning_rate": 0.000462,
      "loss": 6.9249,
      "mean_token_accuracy": 0.11611186563968659,
      "num_tokens": 1676773.0,
      "step": 925
    },
    {
      "entropy": 7.025462627410889,
      "epoch": 0.7993124194241513,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004645,
      "loss": 7.0319,
      "mean_token_accuracy": 0.10499885082244872,
      "num_tokens": 1686144.0,
      "step": 930
    },
    {
      "entropy": 7.038275098800659,
      "epoch": 0.8036097980232059,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000467,
      "loss": 7.0061,
      "mean_token_accuracy": 0.10581193268299102,
      "num_tokens": 1695476.0,
      "step": 935
    },
    {
      "entropy": 6.974601221084595,
      "epoch": 0.8079071766222604,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004695,
      "loss": 6.8898,
      "mean_token_accuracy": 0.11397958695888519,
      "num_tokens": 1704907.0,
      "step": 940
    },
    {
      "entropy": 7.0045435428619385,
      "epoch": 0.812204555221315,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000472,
      "loss": 6.918,
      "mean_token_accuracy": 0.11368482336401939,
      "num_tokens": 1714564.0,
      "step": 945
    },
    {
      "entropy": 7.094973802566528,
      "epoch": 0.8165019338203696,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004745,
      "loss": 6.956,
      "mean_token_accuracy": 0.110804034024477,
      "num_tokens": 1725285.0,
      "step": 950
    },
    {
      "entropy": 6.9100405216217045,
      "epoch": 0.8207993124194242,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000477,
      "loss": 6.9344,
      "mean_token_accuracy": 0.11342336311936378,
      "num_tokens": 1734331.0,
      "step": 955
    },
    {
      "entropy": 7.086512899398803,
      "epoch": 0.8250966910184787,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004795,
      "loss": 6.8794,
      "mean_token_accuracy": 0.11003802865743637,
      "num_tokens": 1742340.0,
      "step": 960
    },
    {
      "entropy": 6.859744882583618,
      "epoch": 0.8293940696175333,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000482,
      "loss": 6.8616,
      "mean_token_accuracy": 0.11626880392432212,
      "num_tokens": 1751725.0,
      "step": 965
    },
    {
      "entropy": 6.893418073654175,
      "epoch": 0.8336914482165879,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004845,
      "loss": 6.7937,
      "mean_token_accuracy": 0.12318155318498611,
      "num_tokens": 1760294.0,
      "step": 970
    },
    {
      "entropy": 6.883913803100586,
      "epoch": 0.8379888268156425,
      "grad_norm": 1.171875,
      "learning_rate": 0.000487,
      "loss": 6.8749,
      "mean_token_accuracy": 0.11642721146345139,
      "num_tokens": 1768912.0,
      "step": 975
    },
    {
      "entropy": 7.098875427246094,
      "epoch": 0.842286205414697,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004895,
      "loss": 7.0701,
      "mean_token_accuracy": 0.11430519297719002,
      "num_tokens": 1778633.0,
      "step": 980
    },
    {
      "entropy": 6.9482495307922365,
      "epoch": 0.8465835840137517,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000492,
      "loss": 6.9373,
      "mean_token_accuracy": 0.11983695551753044,
      "num_tokens": 1787275.0,
      "step": 985
    },
    {
      "entropy": 7.047261524200439,
      "epoch": 0.8508809626128062,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004945,
      "loss": 6.9066,
      "mean_token_accuracy": 0.11325752139091491,
      "num_tokens": 1795994.0,
      "step": 990
    },
    {
      "entropy": 6.881836032867431,
      "epoch": 0.8551783412118608,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000497,
      "loss": 6.8176,
      "mean_token_accuracy": 0.1229711078107357,
      "num_tokens": 1806379.0,
      "step": 995
    },
    {
      "entropy": 6.983829498291016,
      "epoch": 0.8594757198109153,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995,
      "loss": 6.8635,
      "mean_token_accuracy": 0.12418088242411614,
      "num_tokens": 1816135.0,
      "step": 1000
    },
    {
      "epoch": 0.8594757198109153,
      "eval_entropy": 6.5237359571027325,
      "eval_loss": 6.886626243591309,
      "eval_mean_token_accuracy": 0.12066693746627451,
      "eval_num_tokens": 1816135.0,
      "eval_runtime": 2.6304,
      "eval_samples_per_second": 1349.239,
      "eval_steps_per_second": 168.797,
      "step": 1000
    },
    {
      "entropy": 6.860518598556519,
      "epoch": 0.86377309840997,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999998427807679,
      "loss": 6.8687,
      "mean_token_accuracy": 0.11553931757807731,
      "num_tokens": 1824777.0,
      "step": 1005
    },
    {
      "entropy": 6.905518817901611,
      "epoch": 0.8680704770090245,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004999992040780138,
      "loss": 6.9043,
      "mean_token_accuracy": 0.11664655804634094,
      "num_tokens": 1833807.0,
      "step": 1010
    },
    {
      "entropy": 7.013458967208862,
      "epoch": 0.8723678556080791,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999980740669294,
      "loss": 6.866,
      "mean_token_accuracy": 0.11891232207417488,
      "num_tokens": 1843375.0,
      "step": 1015
    },
    {
      "entropy": 6.952661991119385,
      "epoch": 0.8766652342071336,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999964527499823,
      "loss": 6.9331,
      "mean_token_accuracy": 0.10813112482428551,
      "num_tokens": 1853036.0,
      "step": 1020
    },
    {
      "entropy": 6.9601478099823,
      "epoch": 0.8809626128061883,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999943401307127,
      "loss": 6.9041,
      "mean_token_accuracy": 0.10686428099870682,
      "num_tokens": 1862041.0,
      "step": 1025
    },
    {
      "entropy": 6.8647785663604735,
      "epoch": 0.8852599914052428,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999917362137337,
      "loss": 6.7971,
      "mean_token_accuracy": 0.11803380921483039,
      "num_tokens": 1870707.0,
      "step": 1030
    },
    {
      "entropy": 6.97495756149292,
      "epoch": 0.8895573700042974,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999886410047312,
      "loss": 6.8129,
      "mean_token_accuracy": 0.12053435295820236,
      "num_tokens": 1879787.0,
      "step": 1035
    },
    {
      "entropy": 6.89826807975769,
      "epoch": 0.8938547486033519,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999850545104638,
      "loss": 6.8449,
      "mean_token_accuracy": 0.11536444425582885,
      "num_tokens": 1889413.0,
      "step": 1040
    },
    {
      "entropy": 6.934769964218139,
      "epoch": 0.8981521272024066,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999809767387633,
      "loss": 6.8562,
      "mean_token_accuracy": 0.11443335115909577,
      "num_tokens": 1898283.0,
      "step": 1045
    },
    {
      "entropy": 6.856943893432617,
      "epoch": 0.9024495058014611,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999764076985337,
      "loss": 6.8634,
      "mean_token_accuracy": 0.11876284033060074,
      "num_tokens": 1907175.0,
      "step": 1050
    },
    {
      "entropy": 6.854787921905517,
      "epoch": 0.9067468844005157,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999713473997519,
      "loss": 6.9125,
      "mean_token_accuracy": 0.11652960851788521,
      "num_tokens": 1918223.0,
      "step": 1055
    },
    {
      "entropy": 6.947815942764282,
      "epoch": 0.9110442629995703,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999657958534677,
      "loss": 6.8447,
      "mean_token_accuracy": 0.11502521187067032,
      "num_tokens": 1928801.0,
      "step": 1060
    },
    {
      "entropy": 6.816051054000854,
      "epoch": 0.9153416415986249,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999597530718034,
      "loss": 6.8097,
      "mean_token_accuracy": 0.12080107033252716,
      "num_tokens": 1937406.0,
      "step": 1065
    },
    {
      "entropy": 6.9242844581604,
      "epoch": 0.9196390201976794,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499953219067954,
      "loss": 6.7973,
      "mean_token_accuracy": 0.1198230266571045,
      "num_tokens": 1947184.0,
      "step": 1070
    },
    {
      "entropy": 6.883242225646972,
      "epoch": 0.923936398796734,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999461938561873,
      "loss": 6.834,
      "mean_token_accuracy": 0.1205545112490654,
      "num_tokens": 1956293.0,
      "step": 1075
    },
    {
      "entropy": 6.884571552276611,
      "epoch": 0.9282337773957886,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999386774518432,
      "loss": 6.8074,
      "mean_token_accuracy": 0.11995837539434433,
      "num_tokens": 1964791.0,
      "step": 1080
    },
    {
      "entropy": 6.769898319244385,
      "epoch": 0.9325311559948432,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999306698713349,
      "loss": 6.7141,
      "mean_token_accuracy": 0.12293745428323746,
      "num_tokens": 1973754.0,
      "step": 1085
    },
    {
      "entropy": 6.906006813049316,
      "epoch": 0.9368285345938977,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999221711321477,
      "loss": 6.7984,
      "mean_token_accuracy": 0.12007073611021042,
      "num_tokens": 1983035.0,
      "step": 1090
    },
    {
      "entropy": 6.752451467514038,
      "epoch": 0.9411259131929522,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999131812528393,
      "loss": 6.8138,
      "mean_token_accuracy": 0.12244802862405776,
      "num_tokens": 1992584.0,
      "step": 1095
    },
    {
      "entropy": 6.889402627944946,
      "epoch": 0.9454232917920069,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00049990370025304,
      "loss": 6.8571,
      "mean_token_accuracy": 0.11934988871216774,
      "num_tokens": 2001876.0,
      "step": 1100
    },
    {
      "entropy": 6.8836750984191895,
      "epoch": 0.9497206703910615,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998937281534526,
      "loss": 6.7468,
      "mean_token_accuracy": 0.12761817276477813,
      "num_tokens": 2011067.0,
      "step": 1105
    },
    {
      "entropy": 6.86335597038269,
      "epoch": 0.954018048990116,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998832649758521,
      "loss": 6.8229,
      "mean_token_accuracy": 0.11846679598093032,
      "num_tokens": 2020763.0,
      "step": 1110
    },
    {
      "entropy": 6.765384292602539,
      "epoch": 0.9583154275891707,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998723107430862,
      "loss": 6.8129,
      "mean_token_accuracy": 0.1182372123003006,
      "num_tokens": 2029534.0,
      "step": 1115
    },
    {
      "entropy": 6.886447429656982,
      "epoch": 0.9626128061882252,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998608654790741,
      "loss": 6.7573,
      "mean_token_accuracy": 0.11676643267273903,
      "num_tokens": 2039143.0,
      "step": 1120
    },
    {
      "entropy": 6.7921895503997805,
      "epoch": 0.9669101847872797,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499848929208808,
      "loss": 6.7191,
      "mean_token_accuracy": 0.12415711209177971,
      "num_tokens": 2048253.0,
      "step": 1125
    },
    {
      "entropy": 6.730324459075928,
      "epoch": 0.9712075633863343,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998365019583519,
      "loss": 6.7645,
      "mean_token_accuracy": 0.1259891837835312,
      "num_tokens": 2057234.0,
      "step": 1130
    },
    {
      "entropy": 6.922126626968383,
      "epoch": 0.975504941985389,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998235837548417,
      "loss": 6.8125,
      "mean_token_accuracy": 0.12313916236162185,
      "num_tokens": 2065431.0,
      "step": 1135
    },
    {
      "entropy": 6.802910661697387,
      "epoch": 0.9798023205844435,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499810174626486,
      "loss": 6.8096,
      "mean_token_accuracy": 0.12332509085536003,
      "num_tokens": 2074723.0,
      "step": 1140
    },
    {
      "entropy": 6.796639680862427,
      "epoch": 0.984099699183498,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004997962746025646,
      "loss": 6.7022,
      "mean_token_accuracy": 0.12409603446722031,
      "num_tokens": 2084509.0,
      "step": 1145
    },
    {
      "entropy": 6.916227102279663,
      "epoch": 0.9883970777825526,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004997818837134298,
      "loss": 6.8258,
      "mean_token_accuracy": 0.12093790173530579,
      "num_tokens": 2093110.0,
      "step": 1150
    },
    {
      "entropy": 6.6372894763946535,
      "epoch": 0.9926944563816072,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997670019905057,
      "loss": 6.6684,
      "mean_token_accuracy": 0.12281760796904564,
      "num_tokens": 2102355.0,
      "step": 1155
    },
    {
      "entropy": 6.856326389312744,
      "epoch": 0.9969918349806618,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997516294662876,
      "loss": 6.7315,
      "mean_token_accuracy": 0.12142538279294968,
      "num_tokens": 2110418.0,
      "step": 1160
    },
    {
      "entropy": 6.7563766903347435,
      "epoch": 1.0008594757198108,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004997357661743433,
      "loss": 6.7106,
      "mean_token_accuracy": 0.12642807099554274,
      "num_tokens": 2117866.0,
      "step": 1165
    },
    {
      "entropy": 6.752506065368652,
      "epoch": 1.0051568543188656,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997194121493118,
      "loss": 6.5302,
      "mean_token_accuracy": 0.12761134281754494,
      "num_tokens": 2126082.0,
      "step": 1170
    },
    {
      "entropy": 6.704809331893921,
      "epoch": 1.0094542329179201,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997025674269037,
      "loss": 6.5085,
      "mean_token_accuracy": 0.13516685590147973,
      "num_tokens": 2134042.0,
      "step": 1175
    },
    {
      "entropy": 6.653689575195313,
      "epoch": 1.0137516115169747,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996852320439013,
      "loss": 6.591,
      "mean_token_accuracy": 0.12557103708386422,
      "num_tokens": 2142570.0,
      "step": 1180
    },
    {
      "entropy": 6.779953145980835,
      "epoch": 1.0180489901160292,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004996674060381578,
      "loss": 6.5451,
      "mean_token_accuracy": 0.1291184291243553,
      "num_tokens": 2151310.0,
      "step": 1185
    },
    {
      "entropy": 6.66243085861206,
      "epoch": 1.0223463687150838,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996490894485985,
      "loss": 6.5506,
      "mean_token_accuracy": 0.12935295850038528,
      "num_tokens": 2160662.0,
      "step": 1190
    },
    {
      "entropy": 6.731634521484375,
      "epoch": 1.0266437473141383,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996302823152193,
      "loss": 6.5437,
      "mean_token_accuracy": 0.12944689616560937,
      "num_tokens": 2170067.0,
      "step": 1195
    },
    {
      "entropy": 6.652099084854126,
      "epoch": 1.0309411259131929,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996109846790873,
      "loss": 6.4931,
      "mean_token_accuracy": 0.134644240885973,
      "num_tokens": 2178850.0,
      "step": 1200
    },
    {
      "entropy": 6.650523138046265,
      "epoch": 1.0352385045122476,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004995911965823412,
      "loss": 6.5253,
      "mean_token_accuracy": 0.13209154531359674,
      "num_tokens": 2188307.0,
      "step": 1205
    },
    {
      "entropy": 6.697449731826782,
      "epoch": 1.0395358831113022,
      "grad_norm": 1.125,
      "learning_rate": 0.0004995709180681899,
      "loss": 6.5163,
      "mean_token_accuracy": 0.13310288339853288,
      "num_tokens": 2197026.0,
      "step": 1210
    },
    {
      "entropy": 6.669093418121338,
      "epoch": 1.0438332617103567,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499550149180914,
      "loss": 6.4938,
      "mean_token_accuracy": 0.1298208139836788,
      "num_tokens": 2205537.0,
      "step": 1215
    },
    {
      "entropy": 6.640572357177734,
      "epoch": 1.0481306403094113,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995288899658641,
      "loss": 6.5253,
      "mean_token_accuracy": 0.13521900922060012,
      "num_tokens": 2214508.0,
      "step": 1220
    },
    {
      "entropy": 6.663976192474365,
      "epoch": 1.0524280189084658,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995071404694619,
      "loss": 6.6448,
      "mean_token_accuracy": 0.12222216129302979,
      "num_tokens": 2223084.0,
      "step": 1225
    },
    {
      "entropy": 6.801617097854614,
      "epoch": 1.0567253975075204,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994849007391996,
      "loss": 6.5559,
      "mean_token_accuracy": 0.12804892435669898,
      "num_tokens": 2231406.0,
      "step": 1230
    },
    {
      "entropy": 6.55214557647705,
      "epoch": 1.061022776106575,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994621708236401,
      "loss": 6.4763,
      "mean_token_accuracy": 0.1329495020210743,
      "num_tokens": 2239867.0,
      "step": 1235
    },
    {
      "entropy": 6.705474472045898,
      "epoch": 1.0653201547056295,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499438950772416,
      "loss": 6.5366,
      "mean_token_accuracy": 0.12770846113562584,
      "num_tokens": 2248844.0,
      "step": 1240
    },
    {
      "entropy": 6.590055131912232,
      "epoch": 1.0696175333046842,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994152406362311,
      "loss": 6.4625,
      "mean_token_accuracy": 0.13395937606692315,
      "num_tokens": 2257599.0,
      "step": 1245
    },
    {
      "entropy": 6.684833335876465,
      "epoch": 1.0739149119037388,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004993910404668586,
      "loss": 6.5105,
      "mean_token_accuracy": 0.1265903040766716,
      "num_tokens": 2266510.0,
      "step": 1250
    },
    {
      "entropy": 6.589158630371093,
      "epoch": 1.0782122905027933,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499366350317142,
      "loss": 6.5074,
      "mean_token_accuracy": 0.13423952087759972,
      "num_tokens": 2275462.0,
      "step": 1255
    },
    {
      "entropy": 6.62285099029541,
      "epoch": 1.0825096691018479,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004993411702409948,
      "loss": 6.4747,
      "mean_token_accuracy": 0.13087429031729697,
      "num_tokens": 2283826.0,
      "step": 1260
    },
    {
      "entropy": 6.651145029067993,
      "epoch": 1.0868070477009024,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993155002934002,
      "loss": 6.4916,
      "mean_token_accuracy": 0.1317196421325207,
      "num_tokens": 2292967.0,
      "step": 1265
    },
    {
      "entropy": 6.677734279632569,
      "epoch": 1.091104426299957,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004992893405304111,
      "loss": 6.6155,
      "mean_token_accuracy": 0.13076187074184417,
      "num_tokens": 2302336.0,
      "step": 1270
    },
    {
      "entropy": 6.616421175003052,
      "epoch": 1.0954018048990115,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00049926269100915,
      "loss": 6.526,
      "mean_token_accuracy": 0.13683564439415932,
      "num_tokens": 2311465.0,
      "step": 1275
    },
    {
      "entropy": 6.670866680145264,
      "epoch": 1.0996991834980663,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004992355517878087,
      "loss": 6.6418,
      "mean_token_accuracy": 0.125611037760973,
      "num_tokens": 2320281.0,
      "step": 1280
    },
    {
      "entropy": 6.660199451446533,
      "epoch": 1.1039965620971208,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004992079229256484,
      "loss": 6.5372,
      "mean_token_accuracy": 0.12755262702703477,
      "num_tokens": 2329755.0,
      "step": 1285
    },
    {
      "entropy": 6.533046245574951,
      "epoch": 1.1082939406961754,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004991798044829996,
      "loss": 6.4921,
      "mean_token_accuracy": 0.1316668689250946,
      "num_tokens": 2338807.0,
      "step": 1290
    },
    {
      "entropy": 6.721467113494873,
      "epoch": 1.11259131929523,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004991511965212618,
      "loss": 6.5841,
      "mean_token_accuracy": 0.13009289875626565,
      "num_tokens": 2348056.0,
      "step": 1295
    },
    {
      "entropy": 6.537866449356079,
      "epoch": 1.1168886978942845,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004991220991029032,
      "loss": 6.5754,
      "mean_token_accuracy": 0.12793516367673874,
      "num_tokens": 2357780.0,
      "step": 1300
    },
    {
      "entropy": 6.719857215881348,
      "epoch": 1.121186076493339,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000499092512291461,
      "loss": 6.5582,
      "mean_token_accuracy": 0.1265191540122032,
      "num_tokens": 2367060.0,
      "step": 1305
    },
    {
      "entropy": 6.613448905944824,
      "epoch": 1.1254834550923936,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499062436151541,
      "loss": 6.5412,
      "mean_token_accuracy": 0.13018715083599092,
      "num_tokens": 2375751.0,
      "step": 1310
    },
    {
      "entropy": 6.668878126144409,
      "epoch": 1.129780833691448,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004990318707488173,
      "loss": 6.5926,
      "mean_token_accuracy": 0.1261170446872711,
      "num_tokens": 2385013.0,
      "step": 1315
    },
    {
      "entropy": 6.6744931697845455,
      "epoch": 1.1340782122905029,
      "grad_norm": 1.125,
      "learning_rate": 0.0004990008161500327,
      "loss": 6.4939,
      "mean_token_accuracy": 0.13619888350367546,
      "num_tokens": 2392935.0,
      "step": 1320
    },
    {
      "entropy": 6.598589897155762,
      "epoch": 1.1383755908895574,
      "grad_norm": 1.140625,
      "learning_rate": 0.000498969272422998,
      "loss": 6.5151,
      "mean_token_accuracy": 0.1341960161924362,
      "num_tokens": 2401560.0,
      "step": 1325
    },
    {
      "entropy": 6.5921900272369385,
      "epoch": 1.142672969488612,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004989372396365921,
      "loss": 6.4283,
      "mean_token_accuracy": 0.13502056449651717,
      "num_tokens": 2410050.0,
      "step": 1330
    },
    {
      "entropy": 6.658480310440064,
      "epoch": 1.1469703480876665,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004989047178607618,
      "loss": 6.5281,
      "mean_token_accuracy": 0.13080739006400108,
      "num_tokens": 2418980.0,
      "step": 1335
    },
    {
      "entropy": 6.614177894592285,
      "epoch": 1.151267726686721,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004988717071665215,
      "loss": 6.531,
      "mean_token_accuracy": 0.12711252719163896,
      "num_tokens": 2427992.0,
      "step": 1340
    },
    {
      "entropy": 6.590984678268432,
      "epoch": 1.1555651052857756,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004988382076259537,
      "loss": 6.4322,
      "mean_token_accuracy": 0.13690400421619414,
      "num_tokens": 2436368.0,
      "step": 1345
    },
    {
      "entropy": 6.564122438430786,
      "epoch": 1.1598624838848304,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004988042193122077,
      "loss": 6.431,
      "mean_token_accuracy": 0.1385386474430561,
      "num_tokens": 2445499.0,
      "step": 1350
    },
    {
      "entropy": 6.659243583679199,
      "epoch": 1.164159862483885,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004987697422995005,
      "loss": 6.4754,
      "mean_token_accuracy": 0.1324051983654499,
      "num_tokens": 2454312.0,
      "step": 1355
    },
    {
      "entropy": 6.5486212253570555,
      "epoch": 1.1684572410829395,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004987347766631161,
      "loss": 6.5325,
      "mean_token_accuracy": 0.13133806958794594,
      "num_tokens": 2462922.0,
      "step": 1360
    },
    {
      "entropy": 6.615437030792236,
      "epoch": 1.172754619681994,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004986993224794055,
      "loss": 6.583,
      "mean_token_accuracy": 0.12835746854543686,
      "num_tokens": 2472195.0,
      "step": 1365
    },
    {
      "entropy": 6.589010429382324,
      "epoch": 1.1770519982810486,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004986633798257865,
      "loss": 6.4796,
      "mean_token_accuracy": 0.13408638834953307,
      "num_tokens": 2481021.0,
      "step": 1370
    },
    {
      "entropy": 6.539653873443603,
      "epoch": 1.181349376880103,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004986269487807434,
      "loss": 6.4946,
      "mean_token_accuracy": 0.13308287411928177,
      "num_tokens": 2490250.0,
      "step": 1375
    },
    {
      "entropy": 6.634622812271118,
      "epoch": 1.1856467554791577,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000498590029423827,
      "loss": 6.5705,
      "mean_token_accuracy": 0.132661934196949,
      "num_tokens": 2499122.0,
      "step": 1380
    },
    {
      "entropy": 6.700552654266358,
      "epoch": 1.1899441340782122,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004985526218356546,
      "loss": 6.5204,
      "mean_token_accuracy": 0.12895534336566924,
      "num_tokens": 2508454.0,
      "step": 1385
    },
    {
      "entropy": 6.527865076065064,
      "epoch": 1.1942415126772667,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004985147260979093,
      "loss": 6.4634,
      "mean_token_accuracy": 0.14121531173586846,
      "num_tokens": 2517353.0,
      "step": 1390
    },
    {
      "entropy": 6.636118793487549,
      "epoch": 1.1985388912763215,
      "grad_norm": 1.125,
      "learning_rate": 0.0004984763422933402,
      "loss": 6.4827,
      "mean_token_accuracy": 0.13766878992319107,
      "num_tokens": 2526321.0,
      "step": 1395
    },
    {
      "entropy": 6.5368650436401365,
      "epoch": 1.202836269875376,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004984374705057623,
      "loss": 6.4993,
      "mean_token_accuracy": 0.1347013756632805,
      "num_tokens": 2535924.0,
      "step": 1400
    },
    {
      "entropy": 6.610053443908692,
      "epoch": 1.2071336484744306,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004983981108200561,
      "loss": 6.5027,
      "mean_token_accuracy": 0.13288647755980493,
      "num_tokens": 2545606.0,
      "step": 1405
    },
    {
      "entropy": 6.4733387470245365,
      "epoch": 1.2114310270734852,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004983582633221672,
      "loss": 6.477,
      "mean_token_accuracy": 0.13318859785795212,
      "num_tokens": 2554947.0,
      "step": 1410
    },
    {
      "entropy": 6.720026731491089,
      "epoch": 1.2157284056725397,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004983179280991068,
      "loss": 6.6352,
      "mean_token_accuracy": 0.12735798433423043,
      "num_tokens": 2564462.0,
      "step": 1415
    },
    {
      "entropy": 6.607101345062256,
      "epoch": 1.2200257842715942,
      "grad_norm": 1.125,
      "learning_rate": 0.0004982771052389508,
      "loss": 6.4639,
      "mean_token_accuracy": 0.13422766998410224,
      "num_tokens": 2573124.0,
      "step": 1420
    },
    {
      "entropy": 6.607088661193847,
      "epoch": 1.224323162870649,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004982357948308401,
      "loss": 6.56,
      "mean_token_accuracy": 0.12319548428058624,
      "num_tokens": 2581829.0,
      "step": 1425
    },
    {
      "entropy": 6.636404609680175,
      "epoch": 1.2286205414697036,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004981939969649799,
      "loss": 6.427,
      "mean_token_accuracy": 0.1378232181072235,
      "num_tokens": 2590631.0,
      "step": 1430
    },
    {
      "entropy": 6.567480611801147,
      "epoch": 1.232917920068758,
      "grad_norm": 1.125,
      "learning_rate": 0.0004981517117326404,
      "loss": 6.5469,
      "mean_token_accuracy": 0.13434875309467315,
      "num_tokens": 2600684.0,
      "step": 1435
    },
    {
      "entropy": 6.560835504531861,
      "epoch": 1.2372152986678127,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004981089392261553,
      "loss": 6.4626,
      "mean_token_accuracy": 0.13627152964472772,
      "num_tokens": 2609667.0,
      "step": 1440
    },
    {
      "entropy": 6.532538223266601,
      "epoch": 1.2415126772668672,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000498065679538923,
      "loss": 6.5324,
      "mean_token_accuracy": 0.13216254264116287,
      "num_tokens": 2620025.0,
      "step": 1445
    },
    {
      "entropy": 6.637110471725464,
      "epoch": 1.2458100558659218,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004980219327654049,
      "loss": 6.4511,
      "mean_token_accuracy": 0.1365908809006214,
      "num_tokens": 2629032.0,
      "step": 1450
    },
    {
      "entropy": 6.5011430263519285,
      "epoch": 1.2501074344649763,
      "grad_norm": 1.0625,
      "learning_rate": 0.000497977699001127,
      "loss": 6.4337,
      "mean_token_accuracy": 0.14005313366651534,
      "num_tokens": 2638303.0,
      "step": 1455
    },
    {
      "entropy": 6.583467960357666,
      "epoch": 1.2544048130640308,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004979329783426778,
      "loss": 6.473,
      "mean_token_accuracy": 0.1399681955575943,
      "num_tokens": 2647902.0,
      "step": 1460
    },
    {
      "entropy": 6.504309749603271,
      "epoch": 1.2587021916630854,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004978877708877094,
      "loss": 6.5023,
      "mean_token_accuracy": 0.1329456977546215,
      "num_tokens": 2657902.0,
      "step": 1465
    },
    {
      "entropy": 6.5875834941864015,
      "epoch": 1.2629995702621402,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004978420767349368,
      "loss": 6.4405,
      "mean_token_accuracy": 0.13250284716486932,
      "num_tokens": 2667082.0,
      "step": 1470
    },
    {
      "entropy": 6.548440313339233,
      "epoch": 1.2672969488611947,
      "grad_norm": 1.0,
      "learning_rate": 0.0004977958959841379,
      "loss": 6.5259,
      "mean_token_accuracy": 0.13341723382472992,
      "num_tokens": 2676855.0,
      "step": 1475
    },
    {
      "entropy": 6.60371618270874,
      "epoch": 1.2715943274602493,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000497749228736153,
      "loss": 6.4516,
      "mean_token_accuracy": 0.1368909828364849,
      "num_tokens": 2685750.0,
      "step": 1480
    },
    {
      "entropy": 6.530312252044678,
      "epoch": 1.2758917060593038,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004977020750928845,
      "loss": 6.5123,
      "mean_token_accuracy": 0.14152853339910507,
      "num_tokens": 2695272.0,
      "step": 1485
    },
    {
      "entropy": 6.559909868240356,
      "epoch": 1.2801890846583583,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004976544351572973,
      "loss": 6.4417,
      "mean_token_accuracy": 0.13969839066267015,
      "num_tokens": 2704806.0,
      "step": 1490
    },
    {
      "entropy": 6.4414268970489506,
      "epoch": 1.2844864632574131,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004976063090334179,
      "loss": 6.5131,
      "mean_token_accuracy": 0.13546389266848563,
      "num_tokens": 2713521.0,
      "step": 1495
    },
    {
      "entropy": 6.612242555618286,
      "epoch": 1.2887838418564677,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004975576968263346,
      "loss": 6.4793,
      "mean_token_accuracy": 0.12537865340709686,
      "num_tokens": 2721848.0,
      "step": 1500
    },
    {
      "epoch": 1.2887838418564677,
      "eval_entropy": 6.396234018308622,
      "eval_loss": 6.579471111297607,
      "eval_mean_token_accuracy": 0.13534304930887245,
      "eval_num_tokens": 2721848.0,
      "eval_runtime": 2.8309,
      "eval_samples_per_second": 1253.663,
      "eval_steps_per_second": 156.84,
      "step": 1500
    },
    {
      "entropy": 6.469302129745484,
      "epoch": 1.2930812204555222,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000497508598642197,
      "loss": 6.4775,
      "mean_token_accuracy": 0.13551390171051025,
      "num_tokens": 2731473.0,
      "step": 1505
    },
    {
      "entropy": 6.6442944526672365,
      "epoch": 1.2973785990545768,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000497459014588216,
      "loss": 6.5218,
      "mean_token_accuracy": 0.12938549742102623,
      "num_tokens": 2739867.0,
      "step": 1510
    },
    {
      "entropy": 6.515505123138428,
      "epoch": 1.3016759776536313,
      "grad_norm": 1.078125,
      "learning_rate": 0.000497408944772663,
      "loss": 6.447,
      "mean_token_accuracy": 0.13564593568444253,
      "num_tokens": 2748903.0,
      "step": 1515
    },
    {
      "entropy": 6.512873935699463,
      "epoch": 1.3059733562526858,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004973583893048707,
      "loss": 6.4229,
      "mean_token_accuracy": 0.13668794333934783,
      "num_tokens": 2757711.0,
      "step": 1520
    },
    {
      "entropy": 6.633020448684692,
      "epoch": 1.3102707348517404,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004973073482952321,
      "loss": 6.4576,
      "mean_token_accuracy": 0.13406115919351577,
      "num_tokens": 2765633.0,
      "step": 1525
    },
    {
      "entropy": 6.488648509979248,
      "epoch": 1.314568113450795,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004972558218552004,
      "loss": 6.4817,
      "mean_token_accuracy": 0.1359037533402443,
      "num_tokens": 2774495.0,
      "step": 1530
    },
    {
      "entropy": 6.504700994491577,
      "epoch": 1.3188654920498495,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004972038100972885,
      "loss": 6.5048,
      "mean_token_accuracy": 0.1317047119140625,
      "num_tokens": 2782665.0,
      "step": 1535
    },
    {
      "entropy": 6.534765195846558,
      "epoch": 1.323162870648904,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004971513131350697,
      "loss": 6.4356,
      "mean_token_accuracy": 0.1329593062400818,
      "num_tokens": 2791394.0,
      "step": 1540
    },
    {
      "entropy": 6.490444326400757,
      "epoch": 1.3274602492479588,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004970983310831759,
      "loss": 6.4197,
      "mean_token_accuracy": 0.13485073372721673,
      "num_tokens": 2800488.0,
      "step": 1545
    },
    {
      "entropy": 6.544950771331787,
      "epoch": 1.3317576278470133,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004970448640572989,
      "loss": 6.5464,
      "mean_token_accuracy": 0.13310329020023345,
      "num_tokens": 2810116.0,
      "step": 1550
    },
    {
      "entropy": 6.5609852313995365,
      "epoch": 1.336055006446068,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004969909121741895,
      "loss": 6.3531,
      "mean_token_accuracy": 0.13995248079299927,
      "num_tokens": 2819205.0,
      "step": 1555
    },
    {
      "entropy": 6.486909627914429,
      "epoch": 1.3403523850451224,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004969364755516569,
      "loss": 6.428,
      "mean_token_accuracy": 0.13530739918351173,
      "num_tokens": 2828017.0,
      "step": 1560
    },
    {
      "entropy": 6.5232460498809814,
      "epoch": 1.344649763644177,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004968815543085689,
      "loss": 6.4764,
      "mean_token_accuracy": 0.13402760103344918,
      "num_tokens": 2837125.0,
      "step": 1565
    },
    {
      "entropy": 6.532066297531128,
      "epoch": 1.3489471422432318,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004968261485648516,
      "loss": 6.4831,
      "mean_token_accuracy": 0.1357971176505089,
      "num_tokens": 2845438.0,
      "step": 1570
    },
    {
      "entropy": 6.543150091171265,
      "epoch": 1.3532445208422863,
      "grad_norm": 1.03125,
      "learning_rate": 0.000496770258441489,
      "loss": 6.4545,
      "mean_token_accuracy": 0.14239902943372726,
      "num_tokens": 2854389.0,
      "step": 1575
    },
    {
      "entropy": 6.534445333480835,
      "epoch": 1.3575418994413408,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004967138840605228,
      "loss": 6.4277,
      "mean_token_accuracy": 0.13663119375705718,
      "num_tokens": 2863654.0,
      "step": 1580
    },
    {
      "entropy": 6.371426105499268,
      "epoch": 1.3618392780403954,
      "grad_norm": 1.0625,
      "learning_rate": 0.000496657025545052,
      "loss": 6.3269,
      "mean_token_accuracy": 0.13967968374490738,
      "num_tokens": 2872871.0,
      "step": 1585
    },
    {
      "entropy": 6.499322366714478,
      "epoch": 1.36613665663945,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000496599683019233,
      "loss": 6.4312,
      "mean_token_accuracy": 0.13665958493947983,
      "num_tokens": 2881140.0,
      "step": 1590
    },
    {
      "entropy": 6.529193878173828,
      "epoch": 1.3704340352385045,
      "grad_norm": 1.046875,
      "learning_rate": 0.000496541856608279,
      "loss": 6.4037,
      "mean_token_accuracy": 0.13579225316643714,
      "num_tokens": 2889945.0,
      "step": 1595
    },
    {
      "entropy": 6.398252201080322,
      "epoch": 1.374731413837559,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004964835464384595,
      "loss": 6.3424,
      "mean_token_accuracy": 0.14328474253416063,
      "num_tokens": 2898897.0,
      "step": 1600
    },
    {
      "entropy": 6.480287218093872,
      "epoch": 1.3790287924366136,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000496424752637101,
      "loss": 6.3699,
      "mean_token_accuracy": 0.14322275519371033,
      "num_tokens": 2907717.0,
      "step": 1605
    },
    {
      "entropy": 6.459333705902099,
      "epoch": 1.3833261710356681,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004963654753325853,
      "loss": 6.3616,
      "mean_token_accuracy": 0.14588853269815444,
      "num_tokens": 2916213.0,
      "step": 1610
    },
    {
      "entropy": 6.520181608200073,
      "epoch": 1.387623549634723,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004963057146543505,
      "loss": 6.5216,
      "mean_token_accuracy": 0.13179171159863473,
      "num_tokens": 2925706.0,
      "step": 1615
    },
    {
      "entropy": 6.476054716110229,
      "epoch": 1.3919209282337774,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00049624547073289,
      "loss": 6.4326,
      "mean_token_accuracy": 0.13477319702506066,
      "num_tokens": 2934464.0,
      "step": 1620
    },
    {
      "entropy": 6.495903539657593,
      "epoch": 1.396218306832832,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004961847436997526,
      "loss": 6.3254,
      "mean_token_accuracy": 0.14422428011894226,
      "num_tokens": 2944095.0,
      "step": 1625
    },
    {
      "entropy": 6.374484729766846,
      "epoch": 1.4005156854318865,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004961235336875416,
      "loss": 6.358,
      "mean_token_accuracy": 0.14231155291199685,
      "num_tokens": 2953357.0,
      "step": 1630
    },
    {
      "entropy": 6.491524600982666,
      "epoch": 1.404813064030941,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004960618408299154,
      "loss": 6.4899,
      "mean_token_accuracy": 0.1311335600912571,
      "num_tokens": 2963020.0,
      "step": 1635
    },
    {
      "entropy": 6.494795560836792,
      "epoch": 1.4091104426299956,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004959996652615865,
      "loss": 6.3217,
      "mean_token_accuracy": 0.1407495141029358,
      "num_tokens": 2971955.0,
      "step": 1640
    },
    {
      "entropy": 6.465664577484131,
      "epoch": 1.4134078212290504,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004959370071183216,
      "loss": 6.3813,
      "mean_token_accuracy": 0.1411087453365326,
      "num_tokens": 2980662.0,
      "step": 1645
    },
    {
      "entropy": 6.45834002494812,
      "epoch": 1.417705199828105,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004958738665369407,
      "loss": 6.5253,
      "mean_token_accuracy": 0.12874177545309068,
      "num_tokens": 2990038.0,
      "step": 1650
    },
    {
      "entropy": 6.516742801666259,
      "epoch": 1.4220025784271595,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004958102436553179,
      "loss": 6.4256,
      "mean_token_accuracy": 0.1377176247537136,
      "num_tokens": 2999835.0,
      "step": 1655
    },
    {
      "entropy": 6.502577257156372,
      "epoch": 1.426299957026214,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00049574613861238,
      "loss": 6.4241,
      "mean_token_accuracy": 0.13374193757772446,
      "num_tokens": 3009593.0,
      "step": 1660
    },
    {
      "entropy": 6.572561645507813,
      "epoch": 1.4305973356252686,
      "grad_norm": 1.0,
      "learning_rate": 0.0004956815515481069,
      "loss": 6.4608,
      "mean_token_accuracy": 0.14016298577189445,
      "num_tokens": 3019187.0,
      "step": 1665
    },
    {
      "entropy": 6.403193092346191,
      "epoch": 1.4348947142243231,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004956164826035309,
      "loss": 6.357,
      "mean_token_accuracy": 0.13920670598745347,
      "num_tokens": 3027875.0,
      "step": 1670
    },
    {
      "entropy": 6.453193140029907,
      "epoch": 1.4391920928233777,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004955509319207363,
      "loss": 6.3935,
      "mean_token_accuracy": 0.14017415642738343,
      "num_tokens": 3036902.0,
      "step": 1675
    },
    {
      "entropy": 6.406733131408691,
      "epoch": 1.4434894714224322,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004954848996428601,
      "loss": 6.3749,
      "mean_token_accuracy": 0.14780990928411483,
      "num_tokens": 3046653.0,
      "step": 1680
    },
    {
      "entropy": 6.486447191238403,
      "epoch": 1.4477868500214868,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00049541838591409,
      "loss": 6.4719,
      "mean_token_accuracy": 0.13637785688042642,
      "num_tokens": 3056273.0,
      "step": 1685
    },
    {
      "entropy": 6.465806913375855,
      "epoch": 1.4520842286205415,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004953513908796657,
      "loss": 6.3789,
      "mean_token_accuracy": 0.137953469902277,
      "num_tokens": 3065662.0,
      "step": 1690
    },
    {
      "entropy": 6.500221633911133,
      "epoch": 1.456381607219596,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004952839146858773,
      "loss": 6.4145,
      "mean_token_accuracy": 0.14149838909506798,
      "num_tokens": 3073970.0,
      "step": 1695
    },
    {
      "entropy": 6.40098557472229,
      "epoch": 1.4606789858186506,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004952159574800658,
      "loss": 6.409,
      "mean_token_accuracy": 0.1353250853717327,
      "num_tokens": 3082500.0,
      "step": 1700
    },
    {
      "entropy": 6.496115350723267,
      "epoch": 1.4649763644177052,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004951475194106229,
      "loss": 6.3622,
      "mean_token_accuracy": 0.14535047560930253,
      "num_tokens": 3091574.0,
      "step": 1705
    },
    {
      "entropy": 6.4026820182800295,
      "epoch": 1.4692737430167597,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004950786006269898,
      "loss": 6.4757,
      "mean_token_accuracy": 0.13191901370882989,
      "num_tokens": 3102402.0,
      "step": 1710
    },
    {
      "entropy": 6.5063896656036375,
      "epoch": 1.4735711216158143,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004950092012796576,
      "loss": 6.2814,
      "mean_token_accuracy": 0.143421570956707,
      "num_tokens": 3111347.0,
      "step": 1715
    },
    {
      "entropy": 6.346796178817749,
      "epoch": 1.477868500214869,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004949393215201666,
      "loss": 6.3502,
      "mean_token_accuracy": 0.1382603771984577,
      "num_tokens": 3120018.0,
      "step": 1720
    },
    {
      "entropy": 6.488202047348023,
      "epoch": 1.4821658788139236,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004948689615011065,
      "loss": 6.4354,
      "mean_token_accuracy": 0.13543768227100372,
      "num_tokens": 3129669.0,
      "step": 1725
    },
    {
      "entropy": 6.442359161376953,
      "epoch": 1.4864632574129781,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004947981213761154,
      "loss": 6.3728,
      "mean_token_accuracy": 0.13687124252319335,
      "num_tokens": 3139112.0,
      "step": 1730
    },
    {
      "entropy": 6.462133312225342,
      "epoch": 1.4907606360120327,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004947268012998797,
      "loss": 6.3215,
      "mean_token_accuracy": 0.14909568279981614,
      "num_tokens": 3148437.0,
      "step": 1735
    },
    {
      "entropy": 6.405123472213745,
      "epoch": 1.4950580146110872,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000494655001428134,
      "loss": 6.3124,
      "mean_token_accuracy": 0.14245474264025687,
      "num_tokens": 3158165.0,
      "step": 1740
    },
    {
      "entropy": 6.4186896800994875,
      "epoch": 1.4993553932101418,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004945827219176604,
      "loss": 6.4,
      "mean_token_accuracy": 0.14140995144844054,
      "num_tokens": 3167262.0,
      "step": 1745
    },
    {
      "entropy": 6.396775150299073,
      "epoch": 1.5036527718091963,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004945099629262888,
      "loss": 6.3753,
      "mean_token_accuracy": 0.14143237695097924,
      "num_tokens": 3176696.0,
      "step": 1750
    },
    {
      "entropy": 6.565005731582642,
      "epoch": 1.5079501504082509,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004944367246128954,
      "loss": 6.4545,
      "mean_token_accuracy": 0.13297815918922423,
      "num_tokens": 3185857.0,
      "step": 1755
    },
    {
      "entropy": 6.3231383800506595,
      "epoch": 1.5122475290073054,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004943630071374036,
      "loss": 6.2933,
      "mean_token_accuracy": 0.14562440365552903,
      "num_tokens": 3194687.0,
      "step": 1760
    },
    {
      "entropy": 6.474056005477905,
      "epoch": 1.51654490760636,
      "grad_norm": 1.0,
      "learning_rate": 0.0004942888106607828,
      "loss": 6.3535,
      "mean_token_accuracy": 0.1362925410270691,
      "num_tokens": 3204913.0,
      "step": 1765
    },
    {
      "entropy": 6.452696895599365,
      "epoch": 1.5208422862054147,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004942141353450486,
      "loss": 6.3414,
      "mean_token_accuracy": 0.14594482108950615,
      "num_tokens": 3213312.0,
      "step": 1770
    },
    {
      "entropy": 6.347502994537353,
      "epoch": 1.5251396648044693,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004941389813532619,
      "loss": 6.2698,
      "mean_token_accuracy": 0.15359112471342087,
      "num_tokens": 3222992.0,
      "step": 1775
    },
    {
      "entropy": 6.322586011886597,
      "epoch": 1.5294370434035238,
      "grad_norm": 1.03125,
      "learning_rate": 0.000494063348849529,
      "loss": 6.3051,
      "mean_token_accuracy": 0.1480158232152462,
      "num_tokens": 3232836.0,
      "step": 1780
    },
    {
      "entropy": 6.5091650009155275,
      "epoch": 1.5337344220025786,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004939872379990011,
      "loss": 6.4637,
      "mean_token_accuracy": 0.13492583408951758,
      "num_tokens": 3243171.0,
      "step": 1785
    },
    {
      "entropy": 6.4711463928222654,
      "epoch": 1.5380318006016331,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004939106489678739,
      "loss": 6.3839,
      "mean_token_accuracy": 0.14319418519735336,
      "num_tokens": 3251995.0,
      "step": 1790
    },
    {
      "entropy": 6.373999691009521,
      "epoch": 1.5423291792006877,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000493833581923387,
      "loss": 6.326,
      "mean_token_accuracy": 0.14339777454733849,
      "num_tokens": 3260841.0,
      "step": 1795
    },
    {
      "entropy": 6.5138959884643555,
      "epoch": 1.5466265577997422,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004937560370338244,
      "loss": 6.4805,
      "mean_token_accuracy": 0.13039591386914254,
      "num_tokens": 3270979.0,
      "step": 1800
    },
    {
      "entropy": 6.466180372238159,
      "epoch": 1.5509239363987968,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000493678014468513,
      "loss": 6.4021,
      "mean_token_accuracy": 0.143599297106266,
      "num_tokens": 3279848.0,
      "step": 1805
    },
    {
      "entropy": 6.3830766677856445,
      "epoch": 1.5552213149978513,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004935995143978227,
      "loss": 6.3995,
      "mean_token_accuracy": 0.13850793689489366,
      "num_tokens": 3289172.0,
      "step": 1810
    },
    {
      "entropy": 6.427796506881714,
      "epoch": 1.5595186935969059,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004935205369931664,
      "loss": 6.275,
      "mean_token_accuracy": 0.14568763375282287,
      "num_tokens": 3297432.0,
      "step": 1815
    },
    {
      "entropy": 6.4932654857635494,
      "epoch": 1.5638160721959604,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004934410824269992,
      "loss": 6.3083,
      "mean_token_accuracy": 0.14067512452602388,
      "num_tokens": 3307486.0,
      "step": 1820
    },
    {
      "entropy": 6.30113582611084,
      "epoch": 1.568113450795015,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004933611508728182,
      "loss": 6.2921,
      "mean_token_accuracy": 0.1483084164559841,
      "num_tokens": 3316296.0,
      "step": 1825
    },
    {
      "entropy": 6.427493619918823,
      "epoch": 1.5724108293940695,
      "grad_norm": 1.046875,
      "learning_rate": 0.000493280742505162,
      "loss": 6.3306,
      "mean_token_accuracy": 0.1380011484026909,
      "num_tokens": 3326080.0,
      "step": 1830
    },
    {
      "entropy": 6.297157955169678,
      "epoch": 1.576708207993124,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004931998574996102,
      "loss": 6.2414,
      "mean_token_accuracy": 0.14382967874407768,
      "num_tokens": 3334826.0,
      "step": 1835
    },
    {
      "entropy": 6.3207234859466555,
      "epoch": 1.5810055865921788,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004931184960327832,
      "loss": 6.2526,
      "mean_token_accuracy": 0.1502455472946167,
      "num_tokens": 3343261.0,
      "step": 1840
    },
    {
      "entropy": 6.337889814376831,
      "epoch": 1.5853029651912334,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004930366582823421,
      "loss": 6.3009,
      "mean_token_accuracy": 0.140691427141428,
      "num_tokens": 3352513.0,
      "step": 1845
    },
    {
      "entropy": 6.36025972366333,
      "epoch": 1.589600343790288,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004929543444269879,
      "loss": 6.319,
      "mean_token_accuracy": 0.1484783448278904,
      "num_tokens": 3361577.0,
      "step": 1850
    },
    {
      "entropy": 6.440754079818726,
      "epoch": 1.5938977223893425,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000492871554646461,
      "loss": 6.4157,
      "mean_token_accuracy": 0.13772669062018394,
      "num_tokens": 3370591.0,
      "step": 1855
    },
    {
      "entropy": 6.370141792297363,
      "epoch": 1.5981951009883972,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004927882891215413,
      "loss": 6.3196,
      "mean_token_accuracy": 0.1388826124370098,
      "num_tokens": 3379761.0,
      "step": 1860
    },
    {
      "entropy": 6.4343994617462155,
      "epoch": 1.6024924795874518,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004927045480340475,
      "loss": 6.4095,
      "mean_token_accuracy": 0.13787330240011214,
      "num_tokens": 3388974.0,
      "step": 1865
    },
    {
      "entropy": 6.3982750415802006,
      "epoch": 1.6067898581865063,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004926203315668363,
      "loss": 6.3234,
      "mean_token_accuracy": 0.14442344531416892,
      "num_tokens": 3398339.0,
      "step": 1870
    },
    {
      "entropy": 6.38454155921936,
      "epoch": 1.6110872367855609,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004925356399038032,
      "loss": 6.2867,
      "mean_token_accuracy": 0.13828113302588463,
      "num_tokens": 3408292.0,
      "step": 1875
    },
    {
      "entropy": 6.368298768997192,
      "epoch": 1.6153846153846154,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004924504732298808,
      "loss": 6.2622,
      "mean_token_accuracy": 0.14790038838982583,
      "num_tokens": 3417057.0,
      "step": 1880
    },
    {
      "entropy": 6.436639022827149,
      "epoch": 1.61968199398367,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004923648317310391,
      "loss": 6.381,
      "mean_token_accuracy": 0.14184708818793296,
      "num_tokens": 3425830.0,
      "step": 1885
    },
    {
      "entropy": 6.369312191009522,
      "epoch": 1.6239793725827245,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004922787155942849,
      "loss": 6.411,
      "mean_token_accuracy": 0.13301676735281945,
      "num_tokens": 3435513.0,
      "step": 1890
    },
    {
      "entropy": 6.379889011383057,
      "epoch": 1.628276751181779,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004921921250076611,
      "loss": 6.3129,
      "mean_token_accuracy": 0.14551183134317397,
      "num_tokens": 3444684.0,
      "step": 1895
    },
    {
      "entropy": 6.3351727485656735,
      "epoch": 1.6325741297808336,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004921050601602475,
      "loss": 6.3717,
      "mean_token_accuracy": 0.14514745026826859,
      "num_tokens": 3453454.0,
      "step": 1900
    },
    {
      "entropy": 6.364632034301758,
      "epoch": 1.6368715083798882,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004920175212421587,
      "loss": 6.3024,
      "mean_token_accuracy": 0.1404518447816372,
      "num_tokens": 3463228.0,
      "step": 1905
    },
    {
      "entropy": 6.370801830291748,
      "epoch": 1.6411688869789427,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004919295084445445,
      "loss": 6.2039,
      "mean_token_accuracy": 0.15301429331302643,
      "num_tokens": 3472131.0,
      "step": 1910
    },
    {
      "entropy": 6.300609064102173,
      "epoch": 1.6454662655779975,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004918410219595899,
      "loss": 6.2529,
      "mean_token_accuracy": 0.15238626301288605,
      "num_tokens": 3480642.0,
      "step": 1915
    },
    {
      "entropy": 6.482364797592163,
      "epoch": 1.649763644177052,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000491752061980514,
      "loss": 6.2502,
      "mean_token_accuracy": 0.14723040014505387,
      "num_tokens": 3489346.0,
      "step": 1920
    },
    {
      "entropy": 6.245130395889282,
      "epoch": 1.6540610227761066,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004916626287015697,
      "loss": 6.2956,
      "mean_token_accuracy": 0.14454371482133865,
      "num_tokens": 3498473.0,
      "step": 1925
    },
    {
      "entropy": 6.452545595169068,
      "epoch": 1.658358401375161,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004915727223180436,
      "loss": 6.3046,
      "mean_token_accuracy": 0.14169235825538634,
      "num_tokens": 3507415.0,
      "step": 1930
    },
    {
      "entropy": 6.395853805541992,
      "epoch": 1.6626557799742159,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004914823430262554,
      "loss": 6.4133,
      "mean_token_accuracy": 0.13193509206175805,
      "num_tokens": 3516873.0,
      "step": 1935
    },
    {
      "entropy": 6.325657653808594,
      "epoch": 1.6669531585732704,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004913914910235573,
      "loss": 6.2705,
      "mean_token_accuracy": 0.14801762700080873,
      "num_tokens": 3525047.0,
      "step": 1940
    },
    {
      "entropy": 6.363438558578491,
      "epoch": 1.671250537172325,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004913001665083337,
      "loss": 6.2905,
      "mean_token_accuracy": 0.14471901580691338,
      "num_tokens": 3534354.0,
      "step": 1945
    },
    {
      "entropy": 6.308793354034424,
      "epoch": 1.6755479157713795,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004912083696800008,
      "loss": 6.3216,
      "mean_token_accuracy": 0.1346129782497883,
      "num_tokens": 3543830.0,
      "step": 1950
    },
    {
      "entropy": 6.393164157867432,
      "epoch": 1.679845294370434,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004911161007390063,
      "loss": 6.2271,
      "mean_token_accuracy": 0.14336094558238982,
      "num_tokens": 3552314.0,
      "step": 1955
    },
    {
      "entropy": 6.246313667297363,
      "epoch": 1.6841426729694886,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004910233598868287,
      "loss": 6.3015,
      "mean_token_accuracy": 0.14284233823418618,
      "num_tokens": 3561656.0,
      "step": 1960
    },
    {
      "entropy": 6.327203702926636,
      "epoch": 1.6884400515685432,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004909301473259769,
      "loss": 6.2888,
      "mean_token_accuracy": 0.14548132047057152,
      "num_tokens": 3571784.0,
      "step": 1965
    },
    {
      "entropy": 6.413244581222534,
      "epoch": 1.6927374301675977,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004908364632599899,
      "loss": 6.2576,
      "mean_token_accuracy": 0.1450193226337433,
      "num_tokens": 3580626.0,
      "step": 1970
    },
    {
      "entropy": 6.291885328292847,
      "epoch": 1.6970348087666522,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004907423078934362,
      "loss": 6.2453,
      "mean_token_accuracy": 0.148175410926342,
      "num_tokens": 3589916.0,
      "step": 1975
    },
    {
      "entropy": 6.36896767616272,
      "epoch": 1.7013321873657068,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004906476814319134,
      "loss": 6.284,
      "mean_token_accuracy": 0.14784513786435127,
      "num_tokens": 3599128.0,
      "step": 1980
    },
    {
      "entropy": 6.303310966491699,
      "epoch": 1.7056295659647613,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004905525840820481,
      "loss": 6.2935,
      "mean_token_accuracy": 0.14577944353222846,
      "num_tokens": 3608764.0,
      "step": 1985
    },
    {
      "entropy": 6.450990772247314,
      "epoch": 1.709926944563816,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004904570160514948,
      "loss": 6.3469,
      "mean_token_accuracy": 0.13999350219964982,
      "num_tokens": 3619082.0,
      "step": 1990
    },
    {
      "entropy": 6.30384087562561,
      "epoch": 1.7142243231628707,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004903609775489358,
      "loss": 6.3176,
      "mean_token_accuracy": 0.1439313054084778,
      "num_tokens": 3628695.0,
      "step": 1995
    },
    {
      "entropy": 6.3533117294311525,
      "epoch": 1.7185217017619252,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004902644687840809,
      "loss": 6.2817,
      "mean_token_accuracy": 0.14347996339201927,
      "num_tokens": 3637599.0,
      "step": 2000
    },
    {
      "epoch": 1.7185217017619252,
      "eval_entropy": 6.091574696807174,
      "eval_loss": 6.368002414703369,
      "eval_mean_token_accuracy": 0.1463287656807953,
      "eval_num_tokens": 3637599.0,
      "eval_runtime": 2.6343,
      "eval_samples_per_second": 1347.23,
      "eval_steps_per_second": 168.546,
      "step": 2000
    },
    {
      "entropy": 6.318877649307251,
      "epoch": 1.7228190803609797,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004901674899676667,
      "loss": 6.2771,
      "mean_token_accuracy": 0.14787266850471498,
      "num_tokens": 3647406.0,
      "step": 2005
    },
    {
      "entropy": 6.291737365722656,
      "epoch": 1.7271164589600345,
      "grad_norm": 1.125,
      "learning_rate": 0.0004900700413114561,
      "loss": 6.1647,
      "mean_token_accuracy": 0.15139409750699998,
      "num_tokens": 3656531.0,
      "step": 2010
    },
    {
      "entropy": 6.29096736907959,
      "epoch": 1.731413837559089,
      "grad_norm": 1.03125,
      "learning_rate": 0.000489972123028238,
      "loss": 6.2599,
      "mean_token_accuracy": 0.14592444598674775,
      "num_tokens": 3664922.0,
      "step": 2015
    },
    {
      "entropy": 6.352511978149414,
      "epoch": 1.7357112161581436,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004898737353318268,
      "loss": 6.1849,
      "mean_token_accuracy": 0.1494704321026802,
      "num_tokens": 3673283.0,
      "step": 2020
    },
    {
      "entropy": 6.282111120223999,
      "epoch": 1.7400085947571982,
      "grad_norm": 1.109375,
      "learning_rate": 0.000489774878437062,
      "loss": 6.3298,
      "mean_token_accuracy": 0.14433063864707946,
      "num_tokens": 3681760.0,
      "step": 2025
    },
    {
      "entropy": 6.375459289550781,
      "epoch": 1.7443059733562527,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004896755525598074,
      "loss": 6.144,
      "mean_token_accuracy": 0.15006723552942275,
      "num_tokens": 3689408.0,
      "step": 2030
    },
    {
      "entropy": 6.231076049804687,
      "epoch": 1.7486033519553073,
      "grad_norm": 1.125,
      "learning_rate": 0.0004895757579169511,
      "loss": 6.2602,
      "mean_token_accuracy": 0.1471289351582527,
      "num_tokens": 3697904.0,
      "step": 2035
    },
    {
      "entropy": 6.453372764587402,
      "epoch": 1.7529007305543618,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004894754947264047,
      "loss": 6.2916,
      "mean_token_accuracy": 0.1484273798763752,
      "num_tokens": 3706704.0,
      "step": 2040
    },
    {
      "entropy": 6.315323305130005,
      "epoch": 1.7571981091534163,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000489374763207103,
      "loss": 6.3484,
      "mean_token_accuracy": 0.14002273380756378,
      "num_tokens": 3715690.0,
      "step": 2045
    },
    {
      "entropy": 6.346850919723511,
      "epoch": 1.761495487752471,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004892735635790033,
      "loss": 6.1477,
      "mean_token_accuracy": 0.15207798033952713,
      "num_tokens": 3724835.0,
      "step": 2050
    },
    {
      "entropy": 6.233658790588379,
      "epoch": 1.7657928663515254,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000489171896063085,
      "loss": 6.1926,
      "mean_token_accuracy": 0.14881692826747894,
      "num_tokens": 3733977.0,
      "step": 2055
    },
    {
      "entropy": 6.367916107177734,
      "epoch": 1.77009024495058,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004890697608813495,
      "loss": 6.3022,
      "mean_token_accuracy": 0.13871972635388374,
      "num_tokens": 3742665.0,
      "step": 2060
    },
    {
      "entropy": 6.434072065353393,
      "epoch": 1.7743876235496348,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004889671582568193,
      "loss": 6.3778,
      "mean_token_accuracy": 0.1384774811565876,
      "num_tokens": 3751647.0,
      "step": 2065
    },
    {
      "entropy": 6.281269264221192,
      "epoch": 1.7786850021486893,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004888640884135374,
      "loss": 6.2769,
      "mean_token_accuracy": 0.14352479204535484,
      "num_tokens": 3760852.0,
      "step": 2070
    },
    {
      "entropy": 6.328350830078125,
      "epoch": 1.7829823807477438,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004887605515765671,
      "loss": 6.2042,
      "mean_token_accuracy": 0.15113454610109328,
      "num_tokens": 3768640.0,
      "step": 2075
    },
    {
      "entropy": 6.2967079162597654,
      "epoch": 1.7872797593467986,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004886565479719914,
      "loss": 6.2419,
      "mean_token_accuracy": 0.1427251249551773,
      "num_tokens": 3776504.0,
      "step": 2080
    },
    {
      "entropy": 6.399069786071777,
      "epoch": 1.7915771379458532,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004885520778269128,
      "loss": 6.2894,
      "mean_token_accuracy": 0.14642380475997924,
      "num_tokens": 3786353.0,
      "step": 2085
    },
    {
      "entropy": 6.313576030731201,
      "epoch": 1.7958745165449077,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004884471413694523,
      "loss": 6.3197,
      "mean_token_accuracy": 0.14385596364736558,
      "num_tokens": 3795902.0,
      "step": 2090
    },
    {
      "entropy": 6.347762823104858,
      "epoch": 1.8001718951439623,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004883417388287491,
      "loss": 6.2388,
      "mean_token_accuracy": 0.1416109524667263,
      "num_tokens": 3805986.0,
      "step": 2095
    },
    {
      "entropy": 6.318545532226563,
      "epoch": 1.8044692737430168,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004882358704349603,
      "loss": 6.3591,
      "mean_token_accuracy": 0.1477951005101204,
      "num_tokens": 3814915.0,
      "step": 2100
    },
    {
      "entropy": 6.37258472442627,
      "epoch": 1.8087666523420713,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004881295364192601,
      "loss": 6.2406,
      "mean_token_accuracy": 0.1547567516565323,
      "num_tokens": 3823966.0,
      "step": 2105
    },
    {
      "entropy": 6.420681047439575,
      "epoch": 1.813064030941126,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004880227370138394,
      "loss": 6.2861,
      "mean_token_accuracy": 0.14362874031066894,
      "num_tokens": 3832775.0,
      "step": 2110
    },
    {
      "entropy": 6.228458261489868,
      "epoch": 1.8173614095401804,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004879154724519057,
      "loss": 6.2121,
      "mean_token_accuracy": 0.1502902552485466,
      "num_tokens": 3842808.0,
      "step": 2115
    },
    {
      "entropy": 6.378505086898803,
      "epoch": 1.821658788139235,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004878077429676816,
      "loss": 6.3469,
      "mean_token_accuracy": 0.1420361042022705,
      "num_tokens": 3853303.0,
      "step": 2120
    },
    {
      "entropy": 6.311634635925293,
      "epoch": 1.8259561667382895,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004876995487964054,
      "loss": 6.25,
      "mean_token_accuracy": 0.13848483711481094,
      "num_tokens": 3862462.0,
      "step": 2125
    },
    {
      "entropy": 6.302492427825928,
      "epoch": 1.830253545337344,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048759089017432996,
      "loss": 6.3741,
      "mean_token_accuracy": 0.14119420051574708,
      "num_tokens": 3871596.0,
      "step": 2130
    },
    {
      "entropy": 6.385835075378418,
      "epoch": 1.8345509239363988,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004874817673387222,
      "loss": 6.2773,
      "mean_token_accuracy": 0.1481023609638214,
      "num_tokens": 3881276.0,
      "step": 2135
    },
    {
      "entropy": 6.259650325775146,
      "epoch": 1.8388483025354534,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048737218052786275,
      "loss": 6.3435,
      "mean_token_accuracy": 0.13825829476118087,
      "num_tokens": 3891610.0,
      "step": 2140
    },
    {
      "entropy": 6.423104095458984,
      "epoch": 1.843145681134508,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048726212998104554,
      "loss": 6.288,
      "mean_token_accuracy": 0.1408744402229786,
      "num_tokens": 3900584.0,
      "step": 2145
    },
    {
      "entropy": 6.276949644088745,
      "epoch": 1.8474430597335625,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004871516159385768,
      "loss": 6.2314,
      "mean_token_accuracy": 0.1494441844522953,
      "num_tokens": 3910208.0,
      "step": 2150
    },
    {
      "entropy": 6.17384672164917,
      "epoch": 1.8517404383326173,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004870406386417752,
      "loss": 6.1413,
      "mean_token_accuracy": 0.1555544376373291,
      "num_tokens": 3918424.0,
      "step": 2155
    },
    {
      "entropy": 6.253190803527832,
      "epoch": 1.8560378169316718,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004869291983329707,
      "loss": 6.0808,
      "mean_token_accuracy": 0.1601550966501236,
      "num_tokens": 3926206.0,
      "step": 2160
    },
    {
      "entropy": 6.334660243988037,
      "epoch": 1.8603351955307263,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004868172952555044,
      "loss": 6.182,
      "mean_token_accuracy": 0.13937367051839827,
      "num_tokens": 3935769.0,
      "step": 2165
    },
    {
      "entropy": 6.230923652648926,
      "epoch": 1.864632574129781,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004867049296537278,
      "loss": 6.1661,
      "mean_token_accuracy": 0.1482834830880165,
      "num_tokens": 3945118.0,
      "step": 2170
    },
    {
      "entropy": 6.328389406204224,
      "epoch": 1.8689299527288354,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004865921017730027,
      "loss": 6.2661,
      "mean_token_accuracy": 0.1476903736591339,
      "num_tokens": 3954012.0,
      "step": 2175
    },
    {
      "entropy": 6.291385698318481,
      "epoch": 1.87322733132789,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048647881185969995,
      "loss": 6.2742,
      "mean_token_accuracy": 0.14401593804359436,
      "num_tokens": 3964239.0,
      "step": 2180
    },
    {
      "entropy": 6.288094854354858,
      "epoch": 1.8775247099269445,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004863650601611994,
      "loss": 6.1621,
      "mean_token_accuracy": 0.15318569391965867,
      "num_tokens": 3973694.0,
      "step": 2185
    },
    {
      "entropy": 6.297281074523926,
      "epoch": 1.881822088525999,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048625084692588937,
      "loss": 6.2134,
      "mean_token_accuracy": 0.15158866047859193,
      "num_tokens": 3982706.0,
      "step": 2190
    },
    {
      "entropy": 6.197356653213501,
      "epoch": 1.8861194671250536,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048613617240316593,
      "loss": 6.174,
      "mean_token_accuracy": 0.15385224521160126,
      "num_tokens": 3990934.0,
      "step": 2195
    },
    {
      "entropy": 6.360083818435669,
      "epoch": 1.8904168457241082,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004860210368434323,
      "loss": 6.2316,
      "mean_token_accuracy": 0.15272461101412774,
      "num_tokens": 3999864.0,
      "step": 2200
    },
    {
      "entropy": 6.193952560424805,
      "epoch": 1.8947142243231627,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048590544049809857,
      "loss": 6.2365,
      "mean_token_accuracy": 0.1530345231294632,
      "num_tokens": 4008273.0,
      "step": 2205
    },
    {
      "entropy": 6.370530462265014,
      "epoch": 1.8990116029222175,
      "grad_norm": 1.03125,
      "learning_rate": 0.000485789383619581,
      "loss": 6.249,
      "mean_token_accuracy": 0.15136862844228743,
      "num_tokens": 4017697.0,
      "step": 2210
    },
    {
      "entropy": 6.214098501205444,
      "epoch": 1.903308981521272,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004856728664613015,
      "loss": 6.2656,
      "mean_token_accuracy": 0.1445502944290638,
      "num_tokens": 4026775.0,
      "step": 2215
    },
    {
      "entropy": 6.3401977062225345,
      "epoch": 1.9076063601203266,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048555588927768674,
      "loss": 6.2475,
      "mean_token_accuracy": 0.14767130538821222,
      "num_tokens": 4036476.0,
      "step": 2220
    },
    {
      "entropy": 6.289152050018311,
      "epoch": 1.9119037387193811,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004854384523241683,
      "loss": 6.2226,
      "mean_token_accuracy": 0.14828546568751336,
      "num_tokens": 4045221.0,
      "step": 2225
    },
    {
      "entropy": 6.2168541431427,
      "epoch": 1.916201117318436,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048532055585718143,
      "loss": 6.1471,
      "mean_token_accuracy": 0.1514877662062645,
      "num_tokens": 4053754.0,
      "step": 2230
    },
    {
      "entropy": 6.268859577178955,
      "epoch": 1.9204984959174904,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048520220013416505,
      "loss": 6.1911,
      "mean_token_accuracy": 0.15194396525621415,
      "num_tokens": 4061730.0,
      "step": 2235
    },
    {
      "entropy": 6.280244159698486,
      "epoch": 1.924795874516545,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004850833854135607,
      "loss": 6.2515,
      "mean_token_accuracy": 0.1469673454761505,
      "num_tokens": 4070501.0,
      "step": 2240
    },
    {
      "entropy": 6.289591121673584,
      "epoch": 1.9290932531155995,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004849641119548122,
      "loss": 6.3447,
      "mean_token_accuracy": 0.1468745768070221,
      "num_tokens": 4079621.0,
      "step": 2245
    },
    {
      "entropy": 6.453436183929443,
      "epoch": 1.933390631714654,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000484844380018365,
      "loss": 6.3098,
      "mean_token_accuracy": 0.1452730752527714,
      "num_tokens": 4090106.0,
      "step": 2250
    },
    {
      "entropy": 6.334768152236938,
      "epoch": 1.9376880103137086,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000484724189865666,
      "loss": 6.2396,
      "mean_token_accuracy": 0.14894529432058334,
      "num_tokens": 4099269.0,
      "step": 2255
    },
    {
      "entropy": 6.102892255783081,
      "epoch": 1.9419853889127632,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004846035417591624,
      "loss": 6.1739,
      "mean_token_accuracy": 0.15417957305908203,
      "num_tokens": 4108414.0,
      "step": 2260
    },
    {
      "entropy": 6.362226295471191,
      "epoch": 1.9462827675118177,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004844824359623014,
      "loss": 6.2842,
      "mean_token_accuracy": 0.14404889196157455,
      "num_tokens": 4117731.0,
      "step": 2265
    },
    {
      "entropy": 6.320440101623535,
      "epoch": 1.9505801461108723,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048436087273952966,
      "loss": 6.2651,
      "mean_token_accuracy": 0.1450080007314682,
      "num_tokens": 4127194.0,
      "step": 2270
    },
    {
      "entropy": 6.2928619384765625,
      "epoch": 1.9548775247099268,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048423885235629265,
      "loss": 6.2306,
      "mean_token_accuracy": 0.15602879226207733,
      "num_tokens": 4135594.0,
      "step": 2275
    },
    {
      "entropy": 6.247224378585815,
      "epoch": 1.9591749033089814,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004841163750790342,
      "loss": 6.2688,
      "mean_token_accuracy": 0.14728586077690126,
      "num_tokens": 4145027.0,
      "step": 2280
    },
    {
      "entropy": 6.255398988723755,
      "epoch": 1.9634722819080361,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048399344117519555,
      "loss": 6.1311,
      "mean_token_accuracy": 0.152347894012928,
      "num_tokens": 4153754.0,
      "step": 2285
    },
    {
      "entropy": 6.192139434814453,
      "epoch": 1.9677696605070907,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048387005091321544,
      "loss": 6.1927,
      "mean_token_accuracy": 0.15041765347123146,
      "num_tokens": 4162765.0,
      "step": 2290
    },
    {
      "entropy": 6.372733211517334,
      "epoch": 1.9720670391061452,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048374620456252877,
      "loss": 6.2137,
      "mean_token_accuracy": 0.14727154970169068,
      "num_tokens": 4171589.0,
      "step": 2295
    },
    {
      "entropy": 6.218329286575317,
      "epoch": 1.9763644177052,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048362190239356644,
      "loss": 6.2076,
      "mean_token_accuracy": 0.14930796027183532,
      "num_tokens": 4181817.0,
      "step": 2300
    },
    {
      "entropy": 6.213005685806275,
      "epoch": 1.9806617963042545,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048349714467775474,
      "loss": 6.1767,
      "mean_token_accuracy": 0.14748584479093552,
      "num_tokens": 4191350.0,
      "step": 2305
    },
    {
      "entropy": 6.370962333679199,
      "epoch": 1.984959174903309,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048337193168751464,
      "loss": 6.2288,
      "mean_token_accuracy": 0.1438884899020195,
      "num_tokens": 4199888.0,
      "step": 2310
    },
    {
      "entropy": 6.2559257507324215,
      "epoch": 1.9892565535023636,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004832462636962613,
      "loss": 6.1964,
      "mean_token_accuracy": 0.14573389440774917,
      "num_tokens": 4209509.0,
      "step": 2315
    },
    {
      "entropy": 6.252245616912842,
      "epoch": 1.9935539321014182,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004831201409784034,
      "loss": 6.1734,
      "mean_token_accuracy": 0.15078805685043334,
      "num_tokens": 4218496.0,
      "step": 2320
    },
    {
      "entropy": 6.245168590545655,
      "epoch": 1.9978513107004727,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004829935638093424,
      "loss": 6.1689,
      "mean_token_accuracy": 0.15246346592903137,
      "num_tokens": 4227504.0,
      "step": 2325
    },
    {
      "entropy": 6.250787470075819,
      "epoch": 2.0017189514396216,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004828665324654724,
      "loss": 6.0796,
      "mean_token_accuracy": 0.1526774697833591,
      "num_tokens": 4235338.0,
      "step": 2330
    },
    {
      "entropy": 6.299807786941528,
      "epoch": 2.006016330038676,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004827390472241791,
      "loss": 5.8319,
      "mean_token_accuracy": 0.1594364270567894,
      "num_tokens": 4244905.0,
      "step": 2335
    },
    {
      "entropy": 6.186063194274903,
      "epoch": 2.010313708637731,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004826111083638392,
      "loss": 5.9311,
      "mean_token_accuracy": 0.15949834063649176,
      "num_tokens": 4254533.0,
      "step": 2340
    },
    {
      "entropy": 6.186539793014527,
      "epoch": 2.0146110872367857,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048248271616382,
      "loss": 5.9071,
      "mean_token_accuracy": 0.1573925197124481,
      "num_tokens": 4264023.0,
      "step": 2345
    },
    {
      "entropy": 6.196358966827392,
      "epoch": 2.0189084658358403,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048235387090447894,
      "loss": 5.9306,
      "mean_token_accuracy": 0.15745894163846968,
      "num_tokens": 4273298.0,
      "step": 2350
    },
    {
      "entropy": 6.200879001617432,
      "epoch": 2.023205844434895,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048222457286716235,
      "loss": 5.8942,
      "mean_token_accuracy": 0.16266611367464065,
      "num_tokens": 4283244.0,
      "step": 2355
    },
    {
      "entropy": 6.158419275283814,
      "epoch": 2.0275032230339494,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048209482233420564,
      "loss": 5.8262,
      "mean_token_accuracy": 0.16634698510169982,
      "num_tokens": 4291677.0,
      "step": 2360
    },
    {
      "entropy": 6.176835346221924,
      "epoch": 2.031800601633004,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000481964619588932,
      "loss": 5.9018,
      "mean_token_accuracy": 0.16285468339920045,
      "num_tokens": 4300822.0,
      "step": 2365
    },
    {
      "entropy": 6.165877914428711,
      "epoch": 2.0360979802320585,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004818339649156523,
      "loss": 5.9042,
      "mean_token_accuracy": 0.1610295593738556,
      "num_tokens": 4310149.0,
      "step": 2370
    },
    {
      "entropy": 6.042674922943116,
      "epoch": 2.040395358831113,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048170285859966395,
      "loss": 5.8132,
      "mean_token_accuracy": 0.17229840457439421,
      "num_tokens": 4319109.0,
      "step": 2375
    },
    {
      "entropy": 6.18833737373352,
      "epoch": 2.0446927374301676,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048157130092725087,
      "loss": 5.7992,
      "mean_token_accuracy": 0.16624504029750825,
      "num_tokens": 4327921.0,
      "step": 2380
    },
    {
      "entropy": 6.128254032135009,
      "epoch": 2.048990116029222,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004814392921856824,
      "loss": 5.9541,
      "mean_token_accuracy": 0.16383159458637236,
      "num_tokens": 4338026.0,
      "step": 2385
    },
    {
      "entropy": 6.095228052139282,
      "epoch": 2.0532874946282766,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004813068326632128,
      "loss": 5.7884,
      "mean_token_accuracy": 0.168255914747715,
      "num_tokens": 4347794.0,
      "step": 2390
    },
    {
      "entropy": 6.24313006401062,
      "epoch": 2.057584873227331,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004811739226490809,
      "loss": 5.9717,
      "mean_token_accuracy": 0.15837548822164535,
      "num_tokens": 4357249.0,
      "step": 2395
    },
    {
      "entropy": 6.085185909271241,
      "epoch": 2.0618822518263857,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048104056243350896,
      "loss": 5.9248,
      "mean_token_accuracy": 0.1550598293542862,
      "num_tokens": 4366053.0,
      "step": 2400
    },
    {
      "entropy": 6.192628812789917,
      "epoch": 2.0661796304254403,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004809067523077023,
      "loss": 5.9337,
      "mean_token_accuracy": 0.16046450436115264,
      "num_tokens": 4375543.0,
      "step": 2405
    },
    {
      "entropy": 6.059787702560425,
      "epoch": 2.0704770090244953,
      "grad_norm": 1.125,
      "learning_rate": 0.00048077249256384884,
      "loss": 5.8025,
      "mean_token_accuracy": 0.16983114629983903,
      "num_tokens": 4384332.0,
      "step": 2410
    },
    {
      "entropy": 6.084091567993164,
      "epoch": 2.07477438762355,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004806377834951182,
      "loss": 5.9162,
      "mean_token_accuracy": 0.15598006695508956,
      "num_tokens": 4393670.0,
      "step": 2415
    },
    {
      "entropy": 6.219714832305908,
      "epoch": 2.0790717662226044,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048050262539566104,
      "loss": 5.9082,
      "mean_token_accuracy": 0.16180012375116348,
      "num_tokens": 4402763.0,
      "step": 2420
    },
    {
      "entropy": 6.109860849380493,
      "epoch": 2.083369144821659,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004803670185606087,
      "loss": 5.8379,
      "mean_token_accuracy": 0.16459062099456787,
      "num_tokens": 4411863.0,
      "step": 2425
    },
    {
      "entropy": 6.097035360336304,
      "epoch": 2.0876665234207135,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004802309632860724,
      "loss": 5.9035,
      "mean_token_accuracy": 0.16392969191074372,
      "num_tokens": 4421110.0,
      "step": 2430
    },
    {
      "entropy": 6.110820770263672,
      "epoch": 2.091963902019768,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048009445986914236,
      "loss": 5.9159,
      "mean_token_accuracy": 0.15644611269235612,
      "num_tokens": 4430249.0,
      "step": 2435
    },
    {
      "entropy": 6.130997562408448,
      "epoch": 2.0962612806188226,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047995750860788756,
      "loss": 5.8785,
      "mean_token_accuracy": 0.15592641681432723,
      "num_tokens": 4439686.0,
      "step": 2440
    },
    {
      "entropy": 6.098659610748291,
      "epoch": 2.100558659217877,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004798201098013547,
      "loss": 5.8752,
      "mean_token_accuracy": 0.1653648167848587,
      "num_tokens": 4448645.0,
      "step": 2445
    },
    {
      "entropy": 6.098980140686035,
      "epoch": 2.1048560378169316,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047968226374956797,
      "loss": 5.8269,
      "mean_token_accuracy": 0.1635419562458992,
      "num_tokens": 4456870.0,
      "step": 2450
    },
    {
      "entropy": 6.087638330459595,
      "epoch": 2.109153416415986,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047954397075352794,
      "loss": 5.8799,
      "mean_token_accuracy": 0.16591449379920958,
      "num_tokens": 4466287.0,
      "step": 2455
    },
    {
      "entropy": 6.110076379776001,
      "epoch": 2.1134507950150407,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047940523111521136,
      "loss": 5.7739,
      "mean_token_accuracy": 0.16818707287311555,
      "num_tokens": 4474461.0,
      "step": 2460
    },
    {
      "entropy": 6.054969310760498,
      "epoch": 2.1177481736140953,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004792660451375701,
      "loss": 5.8418,
      "mean_token_accuracy": 0.16488105356693267,
      "num_tokens": 4483002.0,
      "step": 2465
    },
    {
      "entropy": 6.0769671440124515,
      "epoch": 2.12204555221315,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047912641312453064,
      "loss": 5.8564,
      "mean_token_accuracy": 0.16275329440832137,
      "num_tokens": 4492061.0,
      "step": 2470
    },
    {
      "entropy": 6.130937194824218,
      "epoch": 2.1263429308122044,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047898633538099363,
      "loss": 5.9163,
      "mean_token_accuracy": 0.15475746542215346,
      "num_tokens": 4501829.0,
      "step": 2475
    },
    {
      "entropy": 6.041531896591186,
      "epoch": 2.130640309411259,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004788458122128327,
      "loss": 5.9373,
      "mean_token_accuracy": 0.1593800500035286,
      "num_tokens": 4511539.0,
      "step": 2480
    },
    {
      "entropy": 6.092942094802856,
      "epoch": 2.134937688010314,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047870484392689434,
      "loss": 5.806,
      "mean_token_accuracy": 0.1652669280767441,
      "num_tokens": 4520425.0,
      "step": 2485
    },
    {
      "entropy": 6.10265851020813,
      "epoch": 2.1392350666093685,
      "grad_norm": 1.15625,
      "learning_rate": 0.000478563430830997,
      "loss": 5.8856,
      "mean_token_accuracy": 0.15846171379089355,
      "num_tokens": 4529474.0,
      "step": 2490
    },
    {
      "entropy": 6.099829435348511,
      "epoch": 2.143532445208423,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047842157323393035,
      "loss": 5.84,
      "mean_token_accuracy": 0.1621888354420662,
      "num_tokens": 4538082.0,
      "step": 2495
    },
    {
      "entropy": 6.131585359573364,
      "epoch": 2.1478298238074776,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004782792714454547,
      "loss": 6.0238,
      "mean_token_accuracy": 0.15583711564540864,
      "num_tokens": 4547340.0,
      "step": 2500
    },
    {
      "epoch": 2.1478298238074776,
      "eval_entropy": 6.039603040025041,
      "eval_loss": 6.27166223526001,
      "eval_mean_token_accuracy": 0.1527276904666209,
      "eval_num_tokens": 4547340.0,
      "eval_runtime": 2.6239,
      "eval_samples_per_second": 1352.543,
      "eval_steps_per_second": 169.211,
      "step": 2500
    },
    {
      "entropy": 6.112734031677246,
      "epoch": 2.152127202406532,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004781365257763002,
      "loss": 5.8688,
      "mean_token_accuracy": 0.16214950904250144,
      "num_tokens": 4556415.0,
      "step": 2505
    },
    {
      "entropy": 6.020659351348877,
      "epoch": 2.1564245810055866,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047799333653816633,
      "loss": 5.7514,
      "mean_token_accuracy": 0.17052749693393707,
      "num_tokens": 4565156.0,
      "step": 2510
    },
    {
      "entropy": 6.076334428787232,
      "epoch": 2.160721959604641,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047784970404372124,
      "loss": 5.8509,
      "mean_token_accuracy": 0.1637386590242386,
      "num_tokens": 4574678.0,
      "step": 2515
    },
    {
      "entropy": 6.012679958343506,
      "epoch": 2.1650193382036957,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047770562860660083,
      "loss": 5.8575,
      "mean_token_accuracy": 0.15799981504678726,
      "num_tokens": 4583253.0,
      "step": 2520
    },
    {
      "entropy": 6.172503328323364,
      "epoch": 2.1693167168027503,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004775611105414083,
      "loss": 5.9464,
      "mean_token_accuracy": 0.15288824141025542,
      "num_tokens": 4594042.0,
      "step": 2525
    },
    {
      "entropy": 6.031742715835572,
      "epoch": 2.173614095401805,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004774161501637133,
      "loss": 5.9014,
      "mean_token_accuracy": 0.16449007391929626,
      "num_tokens": 4603128.0,
      "step": 2530
    },
    {
      "entropy": 6.089631795883179,
      "epoch": 2.1779114740008594,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004772707477900514,
      "loss": 5.8645,
      "mean_token_accuracy": 0.1612997442483902,
      "num_tokens": 4611537.0,
      "step": 2535
    },
    {
      "entropy": 6.1750177383422855,
      "epoch": 2.182208852599914,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004771249037379232,
      "loss": 5.9786,
      "mean_token_accuracy": 0.1547987088561058,
      "num_tokens": 4622481.0,
      "step": 2540
    },
    {
      "entropy": 6.0418188095092775,
      "epoch": 2.1865062311989685,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004769786183257939,
      "loss": 5.8752,
      "mean_token_accuracy": 0.16891982555389404,
      "num_tokens": 4631259.0,
      "step": 2545
    },
    {
      "entropy": 6.038510131835937,
      "epoch": 2.190803609798023,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004768318918730924,
      "loss": 5.8046,
      "mean_token_accuracy": 0.1702958658337593,
      "num_tokens": 4640266.0,
      "step": 2550
    },
    {
      "entropy": 6.085853242874146,
      "epoch": 2.195100988397078,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047668472470021044,
      "loss": 5.8902,
      "mean_token_accuracy": 0.15410513281822205,
      "num_tokens": 4649520.0,
      "step": 2555
    },
    {
      "entropy": 6.106590795516968,
      "epoch": 2.1993983669961326,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004765371171285025,
      "loss": 5.8194,
      "mean_token_accuracy": 0.16876625418663024,
      "num_tokens": 4658501.0,
      "step": 2560
    },
    {
      "entropy": 5.971577548980713,
      "epoch": 2.203695745595187,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047638906948028445,
      "loss": 5.8927,
      "mean_token_accuracy": 0.15966321527957916,
      "num_tokens": 4667567.0,
      "step": 2565
    },
    {
      "entropy": 6.086361265182495,
      "epoch": 2.2079931241942417,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047624058207883317,
      "loss": 5.8803,
      "mean_token_accuracy": 0.1661355510354042,
      "num_tokens": 4676618.0,
      "step": 2570
    },
    {
      "entropy": 6.191226530075073,
      "epoch": 2.212290502793296,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047609165524838576,
      "loss": 5.9549,
      "mean_token_accuracy": 0.1582304894924164,
      "num_tokens": 4685967.0,
      "step": 2575
    },
    {
      "entropy": 5.970425224304199,
      "epoch": 2.2165878813923507,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004759422893141389,
      "loss": 5.8483,
      "mean_token_accuracy": 0.1639987826347351,
      "num_tokens": 4694568.0,
      "step": 2580
    },
    {
      "entropy": 6.077778053283692,
      "epoch": 2.2208852599914053,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004757924846022482,
      "loss": 5.9149,
      "mean_token_accuracy": 0.16335686296224594,
      "num_tokens": 4703648.0,
      "step": 2585
    },
    {
      "entropy": 6.03687014579773,
      "epoch": 2.22518263859046,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047564224143982714,
      "loss": 5.7617,
      "mean_token_accuracy": 0.16921258717775345,
      "num_tokens": 4712444.0,
      "step": 2590
    },
    {
      "entropy": 6.101342964172363,
      "epoch": 2.2294800171895144,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047549156015494676,
      "loss": 5.9223,
      "mean_token_accuracy": 0.1618524432182312,
      "num_tokens": 4722034.0,
      "step": 2595
    },
    {
      "entropy": 5.988636541366577,
      "epoch": 2.233777395788569,
      "grad_norm": 1.125,
      "learning_rate": 0.00047534044107663484,
      "loss": 5.9315,
      "mean_token_accuracy": 0.16420270651578903,
      "num_tokens": 4731344.0,
      "step": 2600
    },
    {
      "entropy": 6.179015493392944,
      "epoch": 2.2380747743876235,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047518888453487496,
      "loss": 5.8325,
      "mean_token_accuracy": 0.16798994094133377,
      "num_tokens": 4739302.0,
      "step": 2605
    },
    {
      "entropy": 5.9993438720703125,
      "epoch": 2.242372152986678,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004750368908606061,
      "loss": 5.9712,
      "mean_token_accuracy": 0.1615530237555504,
      "num_tokens": 4748848.0,
      "step": 2610
    },
    {
      "entropy": 6.1699446678161625,
      "epoch": 2.2466695315857326,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047488446038572164,
      "loss": 6.0177,
      "mean_token_accuracy": 0.1542590841650963,
      "num_tokens": 4758194.0,
      "step": 2615
    },
    {
      "entropy": 6.096167087554932,
      "epoch": 2.250966910184787,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004747315934430688,
      "loss": 5.9141,
      "mean_token_accuracy": 0.1600179821252823,
      "num_tokens": 4768081.0,
      "step": 2620
    },
    {
      "entropy": 6.005567502975464,
      "epoch": 2.2552642887838417,
      "grad_norm": 1.171875,
      "learning_rate": 0.000474578290366448,
      "loss": 5.8429,
      "mean_token_accuracy": 0.16374872177839278,
      "num_tokens": 4776471.0,
      "step": 2625
    },
    {
      "entropy": 6.133287858963013,
      "epoch": 2.259561667382896,
      "grad_norm": 1.125,
      "learning_rate": 0.0004744245514906117,
      "loss": 5.8587,
      "mean_token_accuracy": 0.15926120728254317,
      "num_tokens": 4784403.0,
      "step": 2630
    },
    {
      "entropy": 5.980273675918579,
      "epoch": 2.263859045981951,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047427037715126426,
      "loss": 5.816,
      "mean_token_accuracy": 0.16032798886299132,
      "num_tokens": 4792779.0,
      "step": 2635
    },
    {
      "entropy": 6.001334095001221,
      "epoch": 2.2681564245810057,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004741157676850608,
      "loss": 5.8348,
      "mean_token_accuracy": 0.1639099195599556,
      "num_tokens": 4801426.0,
      "step": 2640
    },
    {
      "entropy": 6.066247367858887,
      "epoch": 2.2724538031800603,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047396072342960663,
      "loss": 5.8534,
      "mean_token_accuracy": 0.162247434258461,
      "num_tokens": 4810329.0,
      "step": 2645
    },
    {
      "entropy": 6.035758686065674,
      "epoch": 2.276751181779115,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047380524472345645,
      "loss": 5.9372,
      "mean_token_accuracy": 0.1603607341647148,
      "num_tokens": 4819544.0,
      "step": 2650
    },
    {
      "entropy": 6.040545797348022,
      "epoch": 2.2810485603781694,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004736493319061134,
      "loss": 5.9218,
      "mean_token_accuracy": 0.16008897721767426,
      "num_tokens": 4828113.0,
      "step": 2655
    },
    {
      "entropy": 6.024325180053711,
      "epoch": 2.285345938977224,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004734929853180291,
      "loss": 5.9024,
      "mean_token_accuracy": 0.1628168672323227,
      "num_tokens": 4836989.0,
      "step": 2660
    },
    {
      "entropy": 6.152593612670898,
      "epoch": 2.2896433175762785,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047333620530060175,
      "loss": 5.9454,
      "mean_token_accuracy": 0.1576353132724762,
      "num_tokens": 4847103.0,
      "step": 2665
    },
    {
      "entropy": 6.021273136138916,
      "epoch": 2.293940696175333,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004731789921961764,
      "loss": 5.9643,
      "mean_token_accuracy": 0.15701263695955275,
      "num_tokens": 4856238.0,
      "step": 2670
    },
    {
      "entropy": 6.13275842666626,
      "epoch": 2.2982380747743876,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004730213463480434,
      "loss": 5.8596,
      "mean_token_accuracy": 0.1687108978629112,
      "num_tokens": 4864608.0,
      "step": 2675
    },
    {
      "entropy": 6.022077465057373,
      "epoch": 2.302535453373442,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047286326810043857,
      "loss": 5.8102,
      "mean_token_accuracy": 0.16301658600568772,
      "num_tokens": 4873889.0,
      "step": 2680
    },
    {
      "entropy": 6.018551015853882,
      "epoch": 2.3068328319724967,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047270475779854137,
      "loss": 5.8724,
      "mean_token_accuracy": 0.16501303017139435,
      "num_tokens": 4882902.0,
      "step": 2685
    },
    {
      "entropy": 6.1423060417175295,
      "epoch": 2.311130210571551,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047254581578847507,
      "loss": 5.9,
      "mean_token_accuracy": 0.16175711005926133,
      "num_tokens": 4892390.0,
      "step": 2690
    },
    {
      "entropy": 5.996535396575927,
      "epoch": 2.3154275891706058,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004723864424173055,
      "loss": 6.0089,
      "mean_token_accuracy": 0.15623133033514022,
      "num_tokens": 4901625.0,
      "step": 2695
    },
    {
      "entropy": 6.0932330131530765,
      "epoch": 2.3197249677696608,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004722266380330403,
      "loss": 5.7842,
      "mean_token_accuracy": 0.17170698791742325,
      "num_tokens": 4910804.0,
      "step": 2700
    },
    {
      "entropy": 6.0394035339355465,
      "epoch": 2.3240223463687153,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047206640298462857,
      "loss": 5.8615,
      "mean_token_accuracy": 0.16718332320451737,
      "num_tokens": 4920441.0,
      "step": 2705
    },
    {
      "entropy": 6.0346497058868405,
      "epoch": 2.32831972496777,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047190573762195945,
      "loss": 5.9206,
      "mean_token_accuracy": 0.1583698570728302,
      "num_tokens": 4930204.0,
      "step": 2710
    },
    {
      "entropy": 6.057749891281128,
      "epoch": 2.3326171035668244,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047174464229586186,
      "loss": 6.0217,
      "mean_token_accuracy": 0.15289230942726134,
      "num_tokens": 4941191.0,
      "step": 2715
    },
    {
      "entropy": 6.169630098342895,
      "epoch": 2.336914482165879,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004715831173581036,
      "loss": 5.9768,
      "mean_token_accuracy": 0.15604398101568223,
      "num_tokens": 4951825.0,
      "step": 2720
    },
    {
      "entropy": 6.009410858154297,
      "epoch": 2.3412118607649335,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047142116316139073,
      "loss": 5.9168,
      "mean_token_accuracy": 0.1644278198480606,
      "num_tokens": 4960632.0,
      "step": 2725
    },
    {
      "entropy": 6.152362537384033,
      "epoch": 2.345509239363988,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004712587800593663,
      "loss": 5.932,
      "mean_token_accuracy": 0.16140228062868117,
      "num_tokens": 4969455.0,
      "step": 2730
    },
    {
      "entropy": 6.026926755905151,
      "epoch": 2.3498066179630426,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004710959684066102,
      "loss": 5.8494,
      "mean_token_accuracy": 0.16802738159894942,
      "num_tokens": 4978997.0,
      "step": 2735
    },
    {
      "entropy": 6.05658655166626,
      "epoch": 2.354103996562097,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047093272855863803,
      "loss": 5.9097,
      "mean_token_accuracy": 0.16449163407087325,
      "num_tokens": 4988305.0,
      "step": 2740
    },
    {
      "entropy": 5.999591207504272,
      "epoch": 2.3584013751611517,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004707690608719003,
      "loss": 5.8536,
      "mean_token_accuracy": 0.17064749151468278,
      "num_tokens": 4997022.0,
      "step": 2745
    },
    {
      "entropy": 6.097990083694458,
      "epoch": 2.362698753760206,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004706049657037818,
      "loss": 5.9126,
      "mean_token_accuracy": 0.1574753388762474,
      "num_tokens": 5005664.0,
      "step": 2750
    },
    {
      "entropy": 6.010551643371582,
      "epoch": 2.3669961323592608,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004704404434126009,
      "loss": 5.8714,
      "mean_token_accuracy": 0.15295194536447526,
      "num_tokens": 5014769.0,
      "step": 2755
    },
    {
      "entropy": 6.082769918441772,
      "epoch": 2.3712935109583153,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047027549435760843,
      "loss": 5.9393,
      "mean_token_accuracy": 0.1611029714345932,
      "num_tokens": 5024060.0,
      "step": 2760
    },
    {
      "entropy": 6.1582804203033445,
      "epoch": 2.37559088955737,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004701101188989872,
      "loss": 5.9804,
      "mean_token_accuracy": 0.155602391064167,
      "num_tokens": 5033046.0,
      "step": 2765
    },
    {
      "entropy": 5.99857497215271,
      "epoch": 2.3798882681564244,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046994431739785114,
      "loss": 5.8236,
      "mean_token_accuracy": 0.1750484824180603,
      "num_tokens": 5040894.0,
      "step": 2770
    },
    {
      "entropy": 6.079819822311402,
      "epoch": 2.384185646755479,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046977809021624454,
      "loss": 5.9986,
      "mean_token_accuracy": 0.159964120388031,
      "num_tokens": 5050961.0,
      "step": 2775
    },
    {
      "entropy": 6.148093128204346,
      "epoch": 2.3884830253545335,
      "grad_norm": 1.125,
      "learning_rate": 0.0004696114377171409,
      "loss": 5.923,
      "mean_token_accuracy": 0.15658502727746965,
      "num_tokens": 5060226.0,
      "step": 2780
    },
    {
      "entropy": 6.005710458755493,
      "epoch": 2.3927804039535885,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004694443602644429,
      "loss": 5.891,
      "mean_token_accuracy": 0.15860626548528672,
      "num_tokens": 5069225.0,
      "step": 2785
    },
    {
      "entropy": 6.099031782150268,
      "epoch": 2.397077782552643,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004692768582229808,
      "loss": 5.872,
      "mean_token_accuracy": 0.1625506117939949,
      "num_tokens": 5078386.0,
      "step": 2790
    },
    {
      "entropy": 5.995739078521728,
      "epoch": 2.4013751611516976,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046910893195851213,
      "loss": 5.8067,
      "mean_token_accuracy": 0.1610507681965828,
      "num_tokens": 5087161.0,
      "step": 2795
    },
    {
      "entropy": 6.081733751296997,
      "epoch": 2.405672539750752,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046894058183772074,
      "loss": 5.9655,
      "mean_token_accuracy": 0.16117437928915024,
      "num_tokens": 5096613.0,
      "step": 2800
    },
    {
      "entropy": 6.038573598861694,
      "epoch": 2.4099699183498067,
      "grad_norm": 1.203125,
      "learning_rate": 0.000468771808228216,
      "loss": 5.9261,
      "mean_token_accuracy": 0.1605885148048401,
      "num_tokens": 5106534.0,
      "step": 2805
    },
    {
      "entropy": 6.009059953689575,
      "epoch": 2.414267296948861,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046860261149853197,
      "loss": 5.9486,
      "mean_token_accuracy": 0.15776654332876205,
      "num_tokens": 5115975.0,
      "step": 2810
    },
    {
      "entropy": 6.048645448684693,
      "epoch": 2.4185646755479158,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004684329920181268,
      "loss": 5.8462,
      "mean_token_accuracy": 0.16311540752649306,
      "num_tokens": 5124635.0,
      "step": 2815
    },
    {
      "entropy": 6.0368053913116455,
      "epoch": 2.4228620541469703,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046826295015738154,
      "loss": 5.8119,
      "mean_token_accuracy": 0.17295191586017608,
      "num_tokens": 5133226.0,
      "step": 2820
    },
    {
      "entropy": 5.987456226348877,
      "epoch": 2.427159432746025,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004680924862875996,
      "loss": 5.8974,
      "mean_token_accuracy": 0.16351867467164993,
      "num_tokens": 5142257.0,
      "step": 2825
    },
    {
      "entropy": 6.084688806533814,
      "epoch": 2.4314568113450794,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046792160078100605,
      "loss": 5.9038,
      "mean_token_accuracy": 0.16701326072216033,
      "num_tokens": 5150752.0,
      "step": 2830
    },
    {
      "entropy": 6.010568571090698,
      "epoch": 2.435754189944134,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046775029401074653,
      "loss": 5.7857,
      "mean_token_accuracy": 0.17085398733615875,
      "num_tokens": 5160237.0,
      "step": 2835
    },
    {
      "entropy": 6.090199661254883,
      "epoch": 2.4400515685431885,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046757856635088645,
      "loss": 5.8873,
      "mean_token_accuracy": 0.1663772866129875,
      "num_tokens": 5169752.0,
      "step": 2840
    },
    {
      "entropy": 5.990356922149658,
      "epoch": 2.444348947142243,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004674064181764105,
      "loss": 5.9324,
      "mean_token_accuracy": 0.16507768481969834,
      "num_tokens": 5178892.0,
      "step": 2845
    },
    {
      "entropy": 6.05471134185791,
      "epoch": 2.448646325741298,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046723384986322147,
      "loss": 5.9014,
      "mean_token_accuracy": 0.16048227846622468,
      "num_tokens": 5188468.0,
      "step": 2850
    },
    {
      "entropy": 5.975317525863647,
      "epoch": 2.4529437043403526,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004670608617881395,
      "loss": 5.8258,
      "mean_token_accuracy": 0.16768420934677125,
      "num_tokens": 5197565.0,
      "step": 2855
    },
    {
      "entropy": 5.986767673492432,
      "epoch": 2.457241082939407,
      "grad_norm": 1.125,
      "learning_rate": 0.0004668874543289014,
      "loss": 5.8118,
      "mean_token_accuracy": 0.17526791095733643,
      "num_tokens": 5205791.0,
      "step": 2860
    },
    {
      "entropy": 5.987915706634522,
      "epoch": 2.4615384615384617,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046671362786415986,
      "loss": 5.8336,
      "mean_token_accuracy": 0.1755102038383484,
      "num_tokens": 5214773.0,
      "step": 2865
    },
    {
      "entropy": 5.960792493820191,
      "epoch": 2.465835840137516,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046653938277348237,
      "loss": 5.8568,
      "mean_token_accuracy": 0.17312622368335723,
      "num_tokens": 5223734.0,
      "step": 2870
    },
    {
      "entropy": 6.149091148376465,
      "epoch": 2.4701332187365708,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004663647194373505,
      "loss": 5.9354,
      "mean_token_accuracy": 0.16000303328037263,
      "num_tokens": 5231742.0,
      "step": 2875
    },
    {
      "entropy": 5.962928104400635,
      "epoch": 2.4744305973356253,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046618963823715913,
      "loss": 5.8985,
      "mean_token_accuracy": 0.16055782586336137,
      "num_tokens": 5241673.0,
      "step": 2880
    },
    {
      "entropy": 6.057924270629883,
      "epoch": 2.47872797593468,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046601413955521575,
      "loss": 5.8658,
      "mean_token_accuracy": 0.16624459624290466,
      "num_tokens": 5250082.0,
      "step": 2885
    },
    {
      "entropy": 6.010685443878174,
      "epoch": 2.4830253545337344,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004658382237747393,
      "loss": 5.9297,
      "mean_token_accuracy": 0.16147278249263763,
      "num_tokens": 5259680.0,
      "step": 2890
    },
    {
      "entropy": 6.041595315933227,
      "epoch": 2.487322733132789,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046566189127985946,
      "loss": 5.8996,
      "mean_token_accuracy": 0.16485785096883773,
      "num_tokens": 5269561.0,
      "step": 2895
    },
    {
      "entropy": 6.085468482971192,
      "epoch": 2.4916201117318435,
      "grad_norm": 1.046875,
      "learning_rate": 0.000465485142455616,
      "loss": 5.867,
      "mean_token_accuracy": 0.16827538460493088,
      "num_tokens": 5278659.0,
      "step": 2900
    },
    {
      "entropy": 5.905921554565429,
      "epoch": 2.495917490330898,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046530797768795765,
      "loss": 5.8331,
      "mean_token_accuracy": 0.17265328168869018,
      "num_tokens": 5287619.0,
      "step": 2905
    },
    {
      "entropy": 6.0317473888397215,
      "epoch": 2.5002148689299526,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046513039736374153,
      "loss": 5.9676,
      "mean_token_accuracy": 0.1589129939675331,
      "num_tokens": 5297334.0,
      "step": 2910
    },
    {
      "entropy": 6.07217264175415,
      "epoch": 2.504512247529007,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004649524018707319,
      "loss": 5.8606,
      "mean_token_accuracy": 0.16698037683963776,
      "num_tokens": 5306208.0,
      "step": 2915
    },
    {
      "entropy": 5.981895017623901,
      "epoch": 2.5088096261280617,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046477399159759996,
      "loss": 5.8081,
      "mean_token_accuracy": 0.16871434301137925,
      "num_tokens": 5314754.0,
      "step": 2920
    },
    {
      "entropy": 5.913084983825684,
      "epoch": 2.5131070047271162,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046459516693392246,
      "loss": 5.8211,
      "mean_token_accuracy": 0.16699100136756898,
      "num_tokens": 5324000.0,
      "step": 2925
    },
    {
      "entropy": 6.0547689437866214,
      "epoch": 2.517404383326171,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004644159282701808,
      "loss": 5.8884,
      "mean_token_accuracy": 0.16294697523117066,
      "num_tokens": 5332478.0,
      "step": 2930
    },
    {
      "entropy": 6.078831005096435,
      "epoch": 2.5217017619252258,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046423627599776076,
      "loss": 5.9459,
      "mean_token_accuracy": 0.15489169508218764,
      "num_tokens": 5341635.0,
      "step": 2935
    },
    {
      "entropy": 6.072408723831177,
      "epoch": 2.5259991405242803,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000464056210508951,
      "loss": 5.957,
      "mean_token_accuracy": 0.15589358359575273,
      "num_tokens": 5350144.0,
      "step": 2940
    },
    {
      "entropy": 5.971312141418457,
      "epoch": 2.530296519123335,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004638757321969426,
      "loss": 5.8671,
      "mean_token_accuracy": 0.16721404045820237,
      "num_tokens": 5358788.0,
      "step": 2945
    },
    {
      "entropy": 6.0780237197875975,
      "epoch": 2.5345938977223894,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046369484145582815,
      "loss": 5.9583,
      "mean_token_accuracy": 0.16172538995742797,
      "num_tokens": 5368057.0,
      "step": 2950
    },
    {
      "entropy": 5.921319103240966,
      "epoch": 2.538891276321444,
      "grad_norm": 1.125,
      "learning_rate": 0.00046351353868060054,
      "loss": 5.8036,
      "mean_token_accuracy": 0.16979097872972487,
      "num_tokens": 5376739.0,
      "step": 2955
    },
    {
      "entropy": 6.045459365844726,
      "epoch": 2.5431886549204985,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046333182426715273,
      "loss": 5.9206,
      "mean_token_accuracy": 0.16544167399406434,
      "num_tokens": 5385967.0,
      "step": 2960
    },
    {
      "entropy": 6.095268678665161,
      "epoch": 2.547486033519553,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046314969861227626,
      "loss": 5.9374,
      "mean_token_accuracy": 0.15496064722537994,
      "num_tokens": 5395192.0,
      "step": 2965
    },
    {
      "entropy": 5.993819046020508,
      "epoch": 2.5517834121186076,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004629671621136608,
      "loss": 5.9091,
      "mean_token_accuracy": 0.16285971105098723,
      "num_tokens": 5404694.0,
      "step": 2970
    },
    {
      "entropy": 6.065847873687744,
      "epoch": 2.556080790717662,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004627842151698931,
      "loss": 5.9263,
      "mean_token_accuracy": 0.16298119574785233,
      "num_tokens": 5413102.0,
      "step": 2975
    },
    {
      "entropy": 6.028053331375122,
      "epoch": 2.5603781693167167,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046260085818045625,
      "loss": 5.9072,
      "mean_token_accuracy": 0.1584775522351265,
      "num_tokens": 5423339.0,
      "step": 2980
    },
    {
      "entropy": 6.078606796264649,
      "epoch": 2.5646755479157712,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004624170915457284,
      "loss": 5.8959,
      "mean_token_accuracy": 0.16520549952983857,
      "num_tokens": 5432377.0,
      "step": 2985
    },
    {
      "entropy": 5.980776977539063,
      "epoch": 2.5689729265148262,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046223291566698264,
      "loss": 5.8141,
      "mean_token_accuracy": 0.1644345499575138,
      "num_tokens": 5441038.0,
      "step": 2990
    },
    {
      "entropy": 6.00663743019104,
      "epoch": 2.5732703051138808,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004620483309463855,
      "loss": 5.8539,
      "mean_token_accuracy": 0.16262196451425553,
      "num_tokens": 5449557.0,
      "step": 2995
    },
    {
      "entropy": 6.075806379318237,
      "epoch": 2.5775676837129353,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004618633377869961,
      "loss": 5.969,
      "mean_token_accuracy": 0.1565746933221817,
      "num_tokens": 5458931.0,
      "step": 3000
    },
    {
      "epoch": 2.5775676837129353,
      "eval_entropy": 5.9535146479134085,
      "eval_loss": 6.191389560699463,
      "eval_mean_token_accuracy": 0.15702831731723235,
      "eval_num_tokens": 5458931.0,
      "eval_runtime": 2.6385,
      "eval_samples_per_second": 1345.059,
      "eval_steps_per_second": 168.275,
      "step": 3000
    },
    {
      "entropy": 6.015251874923706,
      "epoch": 2.58186506231199,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004616779365927656,
      "loss": 5.7933,
      "mean_token_accuracy": 0.17440636605024337,
      "num_tokens": 5468539.0,
      "step": 3005
    },
    {
      "entropy": 5.870741844177246,
      "epoch": 2.5861624409110444,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004614921277685361,
      "loss": 5.7313,
      "mean_token_accuracy": 0.1767333447933197,
      "num_tokens": 5475710.0,
      "step": 3010
    },
    {
      "entropy": 5.986069059371948,
      "epoch": 2.590459819510099,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046130591172003976,
      "loss": 5.876,
      "mean_token_accuracy": 0.16390150636434556,
      "num_tokens": 5484597.0,
      "step": 3015
    },
    {
      "entropy": 6.0947503566741945,
      "epoch": 2.5947571981091535,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004611192888538981,
      "loss": 5.9718,
      "mean_token_accuracy": 0.15505678951740265,
      "num_tokens": 5493213.0,
      "step": 3020
    },
    {
      "entropy": 6.071368312835693,
      "epoch": 2.599054576708208,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046093225957762084,
      "loss": 5.9669,
      "mean_token_accuracy": 0.15736154466867447,
      "num_tokens": 5502556.0,
      "step": 3025
    },
    {
      "entropy": 6.0553131103515625,
      "epoch": 2.6033519553072626,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004607448242996051,
      "loss": 5.8744,
      "mean_token_accuracy": 0.16855462789535522,
      "num_tokens": 5511779.0,
      "step": 3030
    },
    {
      "entropy": 6.043158531188965,
      "epoch": 2.607649333906317,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004605569834291347,
      "loss": 5.8432,
      "mean_token_accuracy": 0.1685672163963318,
      "num_tokens": 5520836.0,
      "step": 3035
    },
    {
      "entropy": 5.961936092376709,
      "epoch": 2.6119467125053717,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046036873737637904,
      "loss": 5.8813,
      "mean_token_accuracy": 0.1675259366631508,
      "num_tokens": 5529285.0,
      "step": 3040
    },
    {
      "entropy": 6.0443174839019775,
      "epoch": 2.6162440911044262,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004601800865523921,
      "loss": 5.883,
      "mean_token_accuracy": 0.16329028606414794,
      "num_tokens": 5538160.0,
      "step": 3045
    },
    {
      "entropy": 5.952384805679321,
      "epoch": 2.620541469703481,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045999103136911204,
      "loss": 5.8701,
      "mean_token_accuracy": 0.1625372052192688,
      "num_tokens": 5547355.0,
      "step": 3050
    },
    {
      "entropy": 6.0581084251403805,
      "epoch": 2.6248388483025353,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045980157223935965,
      "loss": 5.8848,
      "mean_token_accuracy": 0.1610598996281624,
      "num_tokens": 5557299.0,
      "step": 3055
    },
    {
      "entropy": 5.965495777130127,
      "epoch": 2.62913622690159,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045961170957683806,
      "loss": 5.8156,
      "mean_token_accuracy": 0.1695963054895401,
      "num_tokens": 5565469.0,
      "step": 3060
    },
    {
      "entropy": 5.990341711044311,
      "epoch": 2.6334336055006444,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045942144379613147,
      "loss": 5.935,
      "mean_token_accuracy": 0.15782787054777145,
      "num_tokens": 5574740.0,
      "step": 3065
    },
    {
      "entropy": 6.084517765045166,
      "epoch": 2.637730984099699,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045923077531270426,
      "loss": 5.9137,
      "mean_token_accuracy": 0.16793479323387145,
      "num_tokens": 5583438.0,
      "step": 3070
    },
    {
      "entropy": 5.985931777954102,
      "epoch": 2.6420283626987535,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004590397045429001,
      "loss": 5.8725,
      "mean_token_accuracy": 0.16909938603639602,
      "num_tokens": 5592389.0,
      "step": 3075
    },
    {
      "entropy": 5.992589473724365,
      "epoch": 2.646325741297808,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00045884823190394134,
      "loss": 5.7798,
      "mean_token_accuracy": 0.17591131776571273,
      "num_tokens": 5601598.0,
      "step": 3080
    },
    {
      "entropy": 5.99040412902832,
      "epoch": 2.650623119896863,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004586563578139275,
      "loss": 5.8858,
      "mean_token_accuracy": 0.1604371339082718,
      "num_tokens": 5610498.0,
      "step": 3085
    },
    {
      "entropy": 5.977429723739624,
      "epoch": 2.6549204984959176,
      "grad_norm": 1.40625,
      "learning_rate": 0.00045846408269183505,
      "loss": 5.7831,
      "mean_token_accuracy": 0.17481062114238738,
      "num_tokens": 5620082.0,
      "step": 3090
    },
    {
      "entropy": 6.044168376922608,
      "epoch": 2.659217877094972,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045827140695751603,
      "loss": 5.8733,
      "mean_token_accuracy": 0.16334586590528488,
      "num_tokens": 5630291.0,
      "step": 3095
    },
    {
      "entropy": 5.9464648246765135,
      "epoch": 2.6635152556940267,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004580783310316971,
      "loss": 5.874,
      "mean_token_accuracy": 0.16866674721240998,
      "num_tokens": 5638784.0,
      "step": 3100
    },
    {
      "entropy": 5.949322175979614,
      "epoch": 2.6678126342930812,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045788485533597895,
      "loss": 5.7265,
      "mean_token_accuracy": 0.17600842714309692,
      "num_tokens": 5647968.0,
      "step": 3105
    },
    {
      "entropy": 6.025244140625,
      "epoch": 2.672110012892136,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045769098029283526,
      "loss": 5.9623,
      "mean_token_accuracy": 0.1540097177028656,
      "num_tokens": 5657543.0,
      "step": 3110
    },
    {
      "entropy": 6.02379412651062,
      "epoch": 2.6764073914911903,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004574967063256115,
      "loss": 5.8573,
      "mean_token_accuracy": 0.16828878223896027,
      "num_tokens": 5666535.0,
      "step": 3115
    },
    {
      "entropy": 5.992211484909058,
      "epoch": 2.680704770090245,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045730203385852447,
      "loss": 5.9371,
      "mean_token_accuracy": 0.16220323592424393,
      "num_tokens": 5676273.0,
      "step": 3120
    },
    {
      "entropy": 5.9227536678314205,
      "epoch": 2.6850021486892994,
      "grad_norm": 1.109375,
      "learning_rate": 0.000457106963316661,
      "loss": 5.8682,
      "mean_token_accuracy": 0.16720343232154847,
      "num_tokens": 5684888.0,
      "step": 3125
    },
    {
      "entropy": 6.045551300048828,
      "epoch": 2.689299527288354,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045691149512597717,
      "loss": 5.9047,
      "mean_token_accuracy": 0.16129783689975738,
      "num_tokens": 5693626.0,
      "step": 3130
    },
    {
      "entropy": 6.0508557796478275,
      "epoch": 2.6935969058874085,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00045671562971329736,
      "loss": 5.8291,
      "mean_token_accuracy": 0.16866300702095033,
      "num_tokens": 5702542.0,
      "step": 3135
    },
    {
      "entropy": 5.952105188369751,
      "epoch": 2.6978942844864635,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045651936750631337,
      "loss": 5.8452,
      "mean_token_accuracy": 0.16578937768936158,
      "num_tokens": 5711440.0,
      "step": 3140
    },
    {
      "entropy": 6.055763578414917,
      "epoch": 2.702191663085518,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045632270893358333,
      "loss": 5.9322,
      "mean_token_accuracy": 0.16412092000246048,
      "num_tokens": 5721495.0,
      "step": 3145
    },
    {
      "entropy": 6.03286509513855,
      "epoch": 2.7064890416845726,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004561256544245312,
      "loss": 5.9443,
      "mean_token_accuracy": 0.15768667608499526,
      "num_tokens": 5730664.0,
      "step": 3150
    },
    {
      "entropy": 5.962044620513916,
      "epoch": 2.710786420283627,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000455928204409445,
      "loss": 5.8253,
      "mean_token_accuracy": 0.17572951316833496,
      "num_tokens": 5740229.0,
      "step": 3155
    },
    {
      "entropy": 5.970143127441406,
      "epoch": 2.7150837988826817,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045573035931947684,
      "loss": 5.8324,
      "mean_token_accuracy": 0.1705564573407173,
      "num_tokens": 5748549.0,
      "step": 3160
    },
    {
      "entropy": 6.028087854385376,
      "epoch": 2.7193811774817362,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004555321195866411,
      "loss": 5.7649,
      "mean_token_accuracy": 0.1665789157152176,
      "num_tokens": 5757603.0,
      "step": 3165
    },
    {
      "entropy": 5.972912311553955,
      "epoch": 2.723678556080791,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004553334856438143,
      "loss": 5.9561,
      "mean_token_accuracy": 0.15752369910478592,
      "num_tokens": 5767520.0,
      "step": 3170
    },
    {
      "entropy": 6.066450452804565,
      "epoch": 2.7279759346798453,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045513445792473356,
      "loss": 5.9355,
      "mean_token_accuracy": 0.15854838043451308,
      "num_tokens": 5776778.0,
      "step": 3175
    },
    {
      "entropy": 6.077041244506836,
      "epoch": 2.7322733132789,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004549350368639958,
      "loss": 5.9885,
      "mean_token_accuracy": 0.15888068675994874,
      "num_tokens": 5785652.0,
      "step": 3180
    },
    {
      "entropy": 6.099004125595092,
      "epoch": 2.7365706918779544,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045473522289705693,
      "loss": 5.9167,
      "mean_token_accuracy": 0.16159161776304246,
      "num_tokens": 5795766.0,
      "step": 3185
    },
    {
      "entropy": 5.919166326522827,
      "epoch": 2.740868070477009,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045453501646023085,
      "loss": 5.9589,
      "mean_token_accuracy": 0.16025417894124985,
      "num_tokens": 5804504.0,
      "step": 3190
    },
    {
      "entropy": 6.0491148948669435,
      "epoch": 2.7451654490760635,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00045433441799068837,
      "loss": 5.8771,
      "mean_token_accuracy": 0.16158021241426468,
      "num_tokens": 5814161.0,
      "step": 3195
    },
    {
      "entropy": 5.993768501281738,
      "epoch": 2.749462827675118,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004541334279264562,
      "loss": 5.7725,
      "mean_token_accuracy": 0.17674892395734787,
      "num_tokens": 5822235.0,
      "step": 3200
    },
    {
      "entropy": 5.940905618667602,
      "epoch": 2.7537602062741726,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045393204670641656,
      "loss": 5.8042,
      "mean_token_accuracy": 0.16777520775794982,
      "num_tokens": 5831572.0,
      "step": 3205
    },
    {
      "entropy": 5.931118297576904,
      "epoch": 2.758057584873227,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004537302747703055,
      "loss": 5.813,
      "mean_token_accuracy": 0.17369199246168138,
      "num_tokens": 5839694.0,
      "step": 3210
    },
    {
      "entropy": 6.1083742618560795,
      "epoch": 2.7623549634722817,
      "grad_norm": 1.25,
      "learning_rate": 0.00045352811255871216,
      "loss": 5.9466,
      "mean_token_accuracy": 0.15793339610099794,
      "num_tokens": 5849131.0,
      "step": 3215
    },
    {
      "entropy": 6.030238151550293,
      "epoch": 2.7666523420713363,
      "grad_norm": 0.984375,
      "learning_rate": 0.00045332556051307804,
      "loss": 5.8486,
      "mean_token_accuracy": 0.16480160504579544,
      "num_tokens": 5858861.0,
      "step": 3220
    },
    {
      "entropy": 6.031745100021363,
      "epoch": 2.770949720670391,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045312261907569585,
      "loss": 5.8528,
      "mean_token_accuracy": 0.1642756223678589,
      "num_tokens": 5867585.0,
      "step": 3225
    },
    {
      "entropy": 5.9777271270751955,
      "epoch": 2.775247099269446,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045291928868970867,
      "loss": 5.8738,
      "mean_token_accuracy": 0.16010442823171617,
      "num_tokens": 5876256.0,
      "step": 3230
    },
    {
      "entropy": 5.9892865180969235,
      "epoch": 2.7795444778685003,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004527155697991087,
      "loss": 5.9467,
      "mean_token_accuracy": 0.15326642394065856,
      "num_tokens": 5885302.0,
      "step": 3235
    },
    {
      "entropy": 6.024959230422974,
      "epoch": 2.783841856467555,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004525114628487365,
      "loss": 5.9677,
      "mean_token_accuracy": 0.15789304226636885,
      "num_tokens": 5895066.0,
      "step": 3240
    },
    {
      "entropy": 6.062315177917481,
      "epoch": 2.7881392350666094,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045230696828428026,
      "loss": 5.9293,
      "mean_token_accuracy": 0.16333307176828385,
      "num_tokens": 5903258.0,
      "step": 3245
    },
    {
      "entropy": 5.992421674728393,
      "epoch": 2.792436613665664,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004521020865522742,
      "loss": 5.8251,
      "mean_token_accuracy": 0.16470370143651963,
      "num_tokens": 5911714.0,
      "step": 3250
    },
    {
      "entropy": 5.975275611877441,
      "epoch": 2.7967339922647185,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045189681810009827,
      "loss": 5.8962,
      "mean_token_accuracy": 0.15860241800546646,
      "num_tokens": 5920432.0,
      "step": 3255
    },
    {
      "entropy": 6.068870687484742,
      "epoch": 2.801031370863773,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045169116337597653,
      "loss": 5.9161,
      "mean_token_accuracy": 0.15756052285432814,
      "num_tokens": 5929202.0,
      "step": 3260
    },
    {
      "entropy": 6.080472087860107,
      "epoch": 2.8053287494628276,
      "grad_norm": 1.28125,
      "learning_rate": 0.000451485122828977,
      "loss": 5.9589,
      "mean_token_accuracy": 0.15693847239017486,
      "num_tokens": 5938034.0,
      "step": 3265
    },
    {
      "entropy": 5.944226551055908,
      "epoch": 2.809626128061882,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045127869690900956,
      "loss": 5.7934,
      "mean_token_accuracy": 0.17031458169221877,
      "num_tokens": 5946944.0,
      "step": 3270
    },
    {
      "entropy": 5.920538330078125,
      "epoch": 2.8139235066609367,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045107188606682613,
      "loss": 5.892,
      "mean_token_accuracy": 0.16842835694551467,
      "num_tokens": 5956475.0,
      "step": 3275
    },
    {
      "entropy": 6.080301523208618,
      "epoch": 2.8182208852599913,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004508646907540188,
      "loss": 5.8798,
      "mean_token_accuracy": 0.15972289741039275,
      "num_tokens": 5965814.0,
      "step": 3280
    },
    {
      "entropy": 5.996534824371338,
      "epoch": 2.8225182638590463,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004506571114230195,
      "loss": 5.9075,
      "mean_token_accuracy": 0.16237858533859253,
      "num_tokens": 5973850.0,
      "step": 3285
    },
    {
      "entropy": 5.90730504989624,
      "epoch": 2.826815642458101,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045044914852709824,
      "loss": 5.8495,
      "mean_token_accuracy": 0.16205417662858962,
      "num_tokens": 5982987.0,
      "step": 3290
    },
    {
      "entropy": 6.092641592025757,
      "epoch": 2.8311130210571553,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004502408025203631,
      "loss": 5.8606,
      "mean_token_accuracy": 0.1686910644173622,
      "num_tokens": 5992227.0,
      "step": 3295
    },
    {
      "entropy": 5.963792896270752,
      "epoch": 2.83541039965621,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004500320738577584,
      "loss": 5.8338,
      "mean_token_accuracy": 0.16970308721065522,
      "num_tokens": 6000243.0,
      "step": 3300
    },
    {
      "entropy": 5.981186723709106,
      "epoch": 2.8397077782552644,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044982296299506407,
      "loss": 5.8322,
      "mean_token_accuracy": 0.1722040668129921,
      "num_tokens": 6009771.0,
      "step": 3305
    },
    {
      "entropy": 6.045083284378052,
      "epoch": 2.844005156854319,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004496134703888948,
      "loss": 5.905,
      "mean_token_accuracy": 0.1612884059548378,
      "num_tokens": 6018683.0,
      "step": 3310
    },
    {
      "entropy": 5.9778515815734865,
      "epoch": 2.8483025354533735,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044940359649669846,
      "loss": 5.7719,
      "mean_token_accuracy": 0.17384362816810608,
      "num_tokens": 6027422.0,
      "step": 3315
    },
    {
      "entropy": 5.9929955959320065,
      "epoch": 2.852599914052428,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044919334177675595,
      "loss": 5.8633,
      "mean_token_accuracy": 0.16016940772533417,
      "num_tokens": 6035670.0,
      "step": 3320
    },
    {
      "entropy": 5.966201400756836,
      "epoch": 2.8568972926514826,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044898270668817955,
      "loss": 5.7873,
      "mean_token_accuracy": 0.17351211607456207,
      "num_tokens": 6044092.0,
      "step": 3325
    },
    {
      "entropy": 5.956835079193115,
      "epoch": 2.861194671250537,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000448771691690912,
      "loss": 5.8567,
      "mean_token_accuracy": 0.1599970579147339,
      "num_tokens": 6053970.0,
      "step": 3330
    },
    {
      "entropy": 5.994864130020142,
      "epoch": 2.8654920498495917,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004485602972457257,
      "loss": 5.8204,
      "mean_token_accuracy": 0.16462148129940032,
      "num_tokens": 6062965.0,
      "step": 3335
    },
    {
      "entropy": 5.992863368988037,
      "epoch": 2.8697894284486463,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044834852381422165,
      "loss": 5.9008,
      "mean_token_accuracy": 0.1637433260679245,
      "num_tokens": 6072420.0,
      "step": 3340
    },
    {
      "entropy": 5.955929565429687,
      "epoch": 2.874086807047701,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044813637185882836,
      "loss": 5.7987,
      "mean_token_accuracy": 0.16683831959962844,
      "num_tokens": 6080915.0,
      "step": 3345
    },
    {
      "entropy": 6.066890859603882,
      "epoch": 2.8783841856467554,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044792384184280106,
      "loss": 5.9239,
      "mean_token_accuracy": 0.16229262053966523,
      "num_tokens": 6090453.0,
      "step": 3350
    },
    {
      "entropy": 5.943010091781616,
      "epoch": 2.88268156424581,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044771093423022013,
      "loss": 5.9791,
      "mean_token_accuracy": 0.15519970804452896,
      "num_tokens": 6099390.0,
      "step": 3355
    },
    {
      "entropy": 5.986198377609253,
      "epoch": 2.8869789428448644,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004474976494859909,
      "loss": 5.8692,
      "mean_token_accuracy": 0.17025581151247024,
      "num_tokens": 6108677.0,
      "step": 3360
    },
    {
      "entropy": 6.007058477401733,
      "epoch": 2.891276321443919,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004472839880758419,
      "loss": 5.8018,
      "mean_token_accuracy": 0.16929676234722138,
      "num_tokens": 6117151.0,
      "step": 3365
    },
    {
      "entropy": 6.097532176971436,
      "epoch": 2.8955737000429735,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004470699504663242,
      "loss": 5.9198,
      "mean_token_accuracy": 0.1557392880320549,
      "num_tokens": 6127167.0,
      "step": 3370
    },
    {
      "entropy": 5.9331482410430905,
      "epoch": 2.899871078642028,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004468555371248104,
      "loss": 5.811,
      "mean_token_accuracy": 0.16359427124261855,
      "num_tokens": 6136487.0,
      "step": 3375
    },
    {
      "entropy": 6.0199809074401855,
      "epoch": 2.904168457241083,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004466407485194937,
      "loss": 5.9189,
      "mean_token_accuracy": 0.16040531545877457,
      "num_tokens": 6145334.0,
      "step": 3380
    },
    {
      "entropy": 6.026473665237427,
      "epoch": 2.9084658358401376,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004464255851193864,
      "loss": 5.8414,
      "mean_token_accuracy": 0.1662326857447624,
      "num_tokens": 6155062.0,
      "step": 3385
    },
    {
      "entropy": 5.972359037399292,
      "epoch": 2.912763214439192,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004462100473943194,
      "loss": 5.8159,
      "mean_token_accuracy": 0.17152023315429688,
      "num_tokens": 6164313.0,
      "step": 3390
    },
    {
      "entropy": 5.973877334594727,
      "epoch": 2.9170605930382467,
      "grad_norm": 1.140625,
      "learning_rate": 0.000445994135814941,
      "loss": 5.8574,
      "mean_token_accuracy": 0.16147627383470536,
      "num_tokens": 6173513.0,
      "step": 3395
    },
    {
      "entropy": 5.96266450881958,
      "epoch": 2.9213579716373013,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044577785085271566,
      "loss": 5.87,
      "mean_token_accuracy": 0.16429724544286728,
      "num_tokens": 6182000.0,
      "step": 3400
    },
    {
      "entropy": 6.0837915420532225,
      "epoch": 2.925655350236356,
      "grad_norm": 1.125,
      "learning_rate": 0.0004455611929799235,
      "loss": 5.8837,
      "mean_token_accuracy": 0.15638401508331298,
      "num_tokens": 6191887.0,
      "step": 3405
    },
    {
      "entropy": 5.879487800598144,
      "epoch": 2.9299527288354104,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004453441626696585,
      "loss": 5.9357,
      "mean_token_accuracy": 0.15588561296463013,
      "num_tokens": 6202897.0,
      "step": 3410
    },
    {
      "entropy": 6.037405920028687,
      "epoch": 2.934250107434465,
      "grad_norm": 1.125,
      "learning_rate": 0.00044512676039582823,
      "loss": 5.8192,
      "mean_token_accuracy": 0.1726475328207016,
      "num_tokens": 6211811.0,
      "step": 3415
    },
    {
      "entropy": 6.025434732437134,
      "epoch": 2.9385474860335195,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004449089866331524,
      "loss": 5.83,
      "mean_token_accuracy": 0.17189965546131133,
      "num_tokens": 6219896.0,
      "step": 3420
    },
    {
      "entropy": 5.89447660446167,
      "epoch": 2.942844864632574,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004446908418571617,
      "loss": 5.8145,
      "mean_token_accuracy": 0.171225206553936,
      "num_tokens": 6228212.0,
      "step": 3425
    },
    {
      "entropy": 5.967305994033813,
      "epoch": 2.9471422432316285,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004444723265441973,
      "loss": 5.9781,
      "mean_token_accuracy": 0.15706060230731964,
      "num_tokens": 6238133.0,
      "step": 3430
    },
    {
      "entropy": 5.97793173789978,
      "epoch": 2.9514396218306835,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004442534411714092,
      "loss": 5.8927,
      "mean_token_accuracy": 0.15664541125297546,
      "num_tokens": 6247331.0,
      "step": 3435
    },
    {
      "entropy": 6.087286186218262,
      "epoch": 2.955737000429738,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044403418621675555,
      "loss": 5.8851,
      "mean_token_accuracy": 0.1622688055038452,
      "num_tokens": 6255280.0,
      "step": 3440
    },
    {
      "entropy": 5.984484243392944,
      "epoch": 2.9600343790287926,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004438145621590017,
      "loss": 5.8304,
      "mean_token_accuracy": 0.16720051914453507,
      "num_tokens": 6264752.0,
      "step": 3445
    },
    {
      "entropy": 5.910861301422119,
      "epoch": 2.964331757627847,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044359456947771857,
      "loss": 5.7809,
      "mean_token_accuracy": 0.16352355778217315,
      "num_tokens": 6273258.0,
      "step": 3450
    },
    {
      "entropy": 5.880699491500854,
      "epoch": 2.9686291362269017,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004433742086532824,
      "loss": 5.6995,
      "mean_token_accuracy": 0.18464418053627013,
      "num_tokens": 6281584.0,
      "step": 3455
    },
    {
      "entropy": 5.95316252708435,
      "epoch": 2.9729265148259563,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044315348016687317,
      "loss": 5.8224,
      "mean_token_accuracy": 0.1641302987933159,
      "num_tokens": 6290016.0,
      "step": 3460
    },
    {
      "entropy": 5.891777420043946,
      "epoch": 2.977223893425011,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004429323845004736,
      "loss": 5.734,
      "mean_token_accuracy": 0.17175184190273285,
      "num_tokens": 6298569.0,
      "step": 3465
    },
    {
      "entropy": 5.903365993499756,
      "epoch": 2.9815212720240654,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044271092213686824,
      "loss": 5.7661,
      "mean_token_accuracy": 0.17182787805795668,
      "num_tokens": 6307684.0,
      "step": 3470
    },
    {
      "entropy": 6.026407480239868,
      "epoch": 2.98581865062312,
      "grad_norm": 1.0,
      "learning_rate": 0.00044248909355964247,
      "loss": 5.9057,
      "mean_token_accuracy": 0.15918205678462982,
      "num_tokens": 6317767.0,
      "step": 3475
    },
    {
      "entropy": 6.092149639129639,
      "epoch": 2.9901160292221745,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044226689925318117,
      "loss": 5.9344,
      "mean_token_accuracy": 0.1575716406106949,
      "num_tokens": 6327457.0,
      "step": 3480
    },
    {
      "entropy": 5.980013179779053,
      "epoch": 2.994413407821229,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044204433970266785,
      "loss": 5.7377,
      "mean_token_accuracy": 0.18050013333559037,
      "num_tokens": 6335747.0,
      "step": 3485
    },
    {
      "entropy": 5.940433025360107,
      "epoch": 2.9987107864202835,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004418214153940837,
      "loss": 5.8196,
      "mean_token_accuracy": 0.1658599093556404,
      "num_tokens": 6344750.0,
      "step": 3490
    },
    {
      "entropy": 5.953445116678874,
      "epoch": 3.002578427159433,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00044159812681420624,
      "loss": 5.7485,
      "mean_token_accuracy": 0.16718086269166735,
      "num_tokens": 6354779.0,
      "step": 3495
    },
    {
      "entropy": 5.959359550476075,
      "epoch": 3.0068758057584875,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004413744744506086,
      "loss": 5.5163,
      "mean_token_accuracy": 0.17318416386842728,
      "num_tokens": 6363809.0,
      "step": 3500
    },
    {
      "epoch": 3.0068758057584875,
      "eval_entropy": 5.808071028004895,
      "eval_loss": 6.120314598083496,
      "eval_mean_token_accuracy": 0.16263622614378864,
      "eval_num_tokens": 6363809.0,
      "eval_runtime": 2.6272,
      "eval_samples_per_second": 1350.889,
      "eval_steps_per_second": 169.004,
      "step": 3500
    },
    {
      "entropy": 5.933870029449463,
      "epoch": 3.011173184357542,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044115045879165806,
      "loss": 5.5839,
      "mean_token_accuracy": 0.175676928460598,
      "num_tokens": 6373082.0,
      "step": 3505
    },
    {
      "entropy": 5.9161820888519285,
      "epoch": 3.0154705629565965,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044092608032651515,
      "loss": 5.5454,
      "mean_token_accuracy": 0.17348759919404982,
      "num_tokens": 6381286.0,
      "step": 3510
    },
    {
      "entropy": 5.909457683563232,
      "epoch": 3.019767941555651,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044070133954513305,
      "loss": 5.5092,
      "mean_token_accuracy": 0.18552801311016082,
      "num_tokens": 6390217.0,
      "step": 3515
    },
    {
      "entropy": 5.969851350784301,
      "epoch": 3.0240653201547056,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004404762369382555,
      "loss": 5.5226,
      "mean_token_accuracy": 0.17632414549589157,
      "num_tokens": 6399276.0,
      "step": 3520
    },
    {
      "entropy": 5.897236680984497,
      "epoch": 3.02836269875376,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044025077299741683,
      "loss": 5.525,
      "mean_token_accuracy": 0.18381931334733964,
      "num_tokens": 6407981.0,
      "step": 3525
    },
    {
      "entropy": 5.844799900054932,
      "epoch": 3.0326600773528147,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044002494821494007,
      "loss": 5.5114,
      "mean_token_accuracy": 0.18218347877264024,
      "num_tokens": 6416159.0,
      "step": 3530
    },
    {
      "entropy": 5.806324815750122,
      "epoch": 3.0369574559518693,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043979876308393635,
      "loss": 5.5665,
      "mean_token_accuracy": 0.17806299924850463,
      "num_tokens": 6424564.0,
      "step": 3535
    },
    {
      "entropy": 5.99275279045105,
      "epoch": 3.041254834550924,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004395722180983036,
      "loss": 5.589,
      "mean_token_accuracy": 0.17690781652927398,
      "num_tokens": 6434163.0,
      "step": 3540
    },
    {
      "entropy": 5.7671958923339846,
      "epoch": 3.0455522131499784,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043934531375272535,
      "loss": 5.417,
      "mean_token_accuracy": 0.19137026965618134,
      "num_tokens": 6443372.0,
      "step": 3545
    },
    {
      "entropy": 5.868448972702026,
      "epoch": 3.049849591749033,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043911805054267015,
      "loss": 5.5048,
      "mean_token_accuracy": 0.18185081630945205,
      "num_tokens": 6452638.0,
      "step": 3550
    },
    {
      "entropy": 5.985547065734863,
      "epoch": 3.0541469703480875,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043889042896439004,
      "loss": 5.5345,
      "mean_token_accuracy": 0.17991046011447906,
      "num_tokens": 6461319.0,
      "step": 3555
    },
    {
      "entropy": 5.893108177185058,
      "epoch": 3.0584443489471425,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043866244951491946,
      "loss": 5.4336,
      "mean_token_accuracy": 0.19298922568559645,
      "num_tokens": 6469506.0,
      "step": 3560
    },
    {
      "entropy": 5.863250494003296,
      "epoch": 3.062741727546197,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043843411269207445,
      "loss": 5.5134,
      "mean_token_accuracy": 0.18835846781730653,
      "num_tokens": 6478404.0,
      "step": 3565
    },
    {
      "entropy": 5.778197145462036,
      "epoch": 3.0670391061452515,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004382054189944514,
      "loss": 5.44,
      "mean_token_accuracy": 0.18331104516983032,
      "num_tokens": 6487447.0,
      "step": 3570
    },
    {
      "entropy": 5.819038820266724,
      "epoch": 3.071336484744306,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004379763689214259,
      "loss": 5.4877,
      "mean_token_accuracy": 0.17835415601730348,
      "num_tokens": 6496738.0,
      "step": 3575
    },
    {
      "entropy": 5.899236726760864,
      "epoch": 3.0756338633433606,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004377469629731518,
      "loss": 5.5141,
      "mean_token_accuracy": 0.17553481161594392,
      "num_tokens": 6505848.0,
      "step": 3580
    },
    {
      "entropy": 5.843332052230835,
      "epoch": 3.079931241942415,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004375172016505599,
      "loss": 5.4782,
      "mean_token_accuracy": 0.17995121628046035,
      "num_tokens": 6515731.0,
      "step": 3585
    },
    {
      "entropy": 5.891195487976074,
      "epoch": 3.0842286205414697,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004372870854553572,
      "loss": 5.543,
      "mean_token_accuracy": 0.1805647909641266,
      "num_tokens": 6524914.0,
      "step": 3590
    },
    {
      "entropy": 5.884959983825683,
      "epoch": 3.0885259991405243,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004370566148900255,
      "loss": 5.5398,
      "mean_token_accuracy": 0.1865120217204094,
      "num_tokens": 6533712.0,
      "step": 3595
    },
    {
      "entropy": 5.859037780761719,
      "epoch": 3.092823377739579,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043682579045782024,
      "loss": 5.6039,
      "mean_token_accuracy": 0.17290965765714644,
      "num_tokens": 6543313.0,
      "step": 3600
    },
    {
      "entropy": 5.807743120193481,
      "epoch": 3.0971207563386334,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004365946126627699,
      "loss": 5.4985,
      "mean_token_accuracy": 0.18919602334499358,
      "num_tokens": 6551634.0,
      "step": 3605
    },
    {
      "entropy": 5.85457592010498,
      "epoch": 3.101418134937688,
      "grad_norm": 1.125,
      "learning_rate": 0.00043636308200967433,
      "loss": 5.4938,
      "mean_token_accuracy": 0.18733223974704744,
      "num_tokens": 6560695.0,
      "step": 3610
    },
    {
      "entropy": 5.76024374961853,
      "epoch": 3.1057155135367425,
      "grad_norm": 1.125,
      "learning_rate": 0.0004361311990041039,
      "loss": 5.3974,
      "mean_token_accuracy": 0.18700604140758514,
      "num_tokens": 6569086.0,
      "step": 3615
    },
    {
      "entropy": 5.808054399490357,
      "epoch": 3.110012892135797,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043589896415239843,
      "loss": 5.4793,
      "mean_token_accuracy": 0.18851352334022523,
      "num_tokens": 6578287.0,
      "step": 3620
    },
    {
      "entropy": 5.828329801559448,
      "epoch": 3.1143102707348516,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043566637796166595,
      "loss": 5.5526,
      "mean_token_accuracy": 0.17525139153003694,
      "num_tokens": 6587015.0,
      "step": 3625
    },
    {
      "entropy": 5.867431020736694,
      "epoch": 3.118607649333906,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043543344093978186,
      "loss": 5.601,
      "mean_token_accuracy": 0.1718316853046417,
      "num_tokens": 6596187.0,
      "step": 3630
    },
    {
      "entropy": 5.864822340011597,
      "epoch": 3.122905027932961,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043520015359538745,
      "loss": 5.4591,
      "mean_token_accuracy": 0.18958659023046492,
      "num_tokens": 6605226.0,
      "step": 3635
    },
    {
      "entropy": 5.794321203231812,
      "epoch": 3.1272024065320156,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004349665164378891,
      "loss": 5.5008,
      "mean_token_accuracy": 0.18200621455907823,
      "num_tokens": 6613232.0,
      "step": 3640
    },
    {
      "entropy": 5.789428567886352,
      "epoch": 3.13149978513107,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043473252997745684,
      "loss": 5.5113,
      "mean_token_accuracy": 0.18076512068510056,
      "num_tokens": 6622247.0,
      "step": 3645
    },
    {
      "entropy": 5.893785858154297,
      "epoch": 3.1357971637301247,
      "grad_norm": 1.984375,
      "learning_rate": 0.00043449819472502366,
      "loss": 5.4675,
      "mean_token_accuracy": 0.18034528195858002,
      "num_tokens": 6630883.0,
      "step": 3650
    },
    {
      "entropy": 5.818866157531739,
      "epoch": 3.1400945423291793,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004342635111922841,
      "loss": 5.5915,
      "mean_token_accuracy": 0.17699178606271743,
      "num_tokens": 6639399.0,
      "step": 3655
    },
    {
      "entropy": 5.8470982074737545,
      "epoch": 3.144391920928234,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004340284798916931,
      "loss": 5.5204,
      "mean_token_accuracy": 0.17858008593320845,
      "num_tokens": 6649288.0,
      "step": 3660
    },
    {
      "entropy": 5.844093990325928,
      "epoch": 3.1486892995272884,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004337931013364653,
      "loss": 5.4288,
      "mean_token_accuracy": 0.18451717346906663,
      "num_tokens": 6658670.0,
      "step": 3665
    },
    {
      "entropy": 5.8302186012268065,
      "epoch": 3.152986678126343,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000433557376040573,
      "loss": 5.5096,
      "mean_token_accuracy": 0.1836533859372139,
      "num_tokens": 6667302.0,
      "step": 3670
    },
    {
      "entropy": 5.885258197784424,
      "epoch": 3.1572840567253975,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043332130451874645,
      "loss": 5.5686,
      "mean_token_accuracy": 0.18251876533031464,
      "num_tokens": 6677393.0,
      "step": 3675
    },
    {
      "entropy": 5.848063898086548,
      "epoch": 3.161581435324452,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043308488728647127,
      "loss": 5.539,
      "mean_token_accuracy": 0.17765716165304185,
      "num_tokens": 6686727.0,
      "step": 3680
    },
    {
      "entropy": 5.796309280395508,
      "epoch": 3.1658788139235066,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004328481248599882,
      "loss": 5.4597,
      "mean_token_accuracy": 0.1872706487774849,
      "num_tokens": 6696116.0,
      "step": 3685
    },
    {
      "entropy": 5.8640275478363035,
      "epoch": 3.170176192522561,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004326110177562918,
      "loss": 5.5685,
      "mean_token_accuracy": 0.17389472126960753,
      "num_tokens": 6704640.0,
      "step": 3690
    },
    {
      "entropy": 5.78633599281311,
      "epoch": 3.1744735711216157,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043237356649312926,
      "loss": 5.4412,
      "mean_token_accuracy": 0.18975964039564133,
      "num_tokens": 6713663.0,
      "step": 3695
    },
    {
      "entropy": 5.780803060531616,
      "epoch": 3.17877094972067,
      "grad_norm": 1.125,
      "learning_rate": 0.0004321357715889991,
      "loss": 5.5799,
      "mean_token_accuracy": 0.17476617842912673,
      "num_tokens": 6722965.0,
      "step": 3700
    },
    {
      "entropy": 5.826023006439209,
      "epoch": 3.1830683283197247,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004318976335631505,
      "loss": 5.5202,
      "mean_token_accuracy": 0.18648175597190858,
      "num_tokens": 6732776.0,
      "step": 3705
    },
    {
      "entropy": 5.850239086151123,
      "epoch": 3.1873657069187797,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043165915293558155,
      "loss": 5.51,
      "mean_token_accuracy": 0.17653508335351945,
      "num_tokens": 6741309.0,
      "step": 3710
    },
    {
      "entropy": 5.831938314437866,
      "epoch": 3.1916630855178343,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004314203302270388,
      "loss": 5.5491,
      "mean_token_accuracy": 0.17970845699310303,
      "num_tokens": 6750584.0,
      "step": 3715
    },
    {
      "entropy": 5.865281438827514,
      "epoch": 3.195960464116889,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004311811659590154,
      "loss": 5.5361,
      "mean_token_accuracy": 0.1788480058312416,
      "num_tokens": 6759344.0,
      "step": 3720
    },
    {
      "entropy": 5.934193706512451,
      "epoch": 3.2002578427159434,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004309416606537507,
      "loss": 5.6777,
      "mean_token_accuracy": 0.17327607274055482,
      "num_tokens": 6770345.0,
      "step": 3725
    },
    {
      "entropy": 5.842852067947388,
      "epoch": 3.204555221314998,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043070181483422843,
      "loss": 5.5815,
      "mean_token_accuracy": 0.17351661771535873,
      "num_tokens": 6779991.0,
      "step": 3730
    },
    {
      "entropy": 5.782969617843628,
      "epoch": 3.2088525999140525,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000430461629024176,
      "loss": 5.5133,
      "mean_token_accuracy": 0.1805244117975235,
      "num_tokens": 6788972.0,
      "step": 3735
    },
    {
      "entropy": 5.787181186676025,
      "epoch": 3.213149978513107,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004302211037480634,
      "loss": 5.4534,
      "mean_token_accuracy": 0.18245744109153747,
      "num_tokens": 6796967.0,
      "step": 3740
    },
    {
      "entropy": 5.786061096191406,
      "epoch": 3.2174473571121616,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004299802395311015,
      "loss": 5.5499,
      "mean_token_accuracy": 0.18429655432701111,
      "num_tokens": 6805961.0,
      "step": 3745
    },
    {
      "entropy": 5.716186952590943,
      "epoch": 3.221744735711216,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004297390368992414,
      "loss": 5.4619,
      "mean_token_accuracy": 0.18117133677005767,
      "num_tokens": 6814657.0,
      "step": 3750
    },
    {
      "entropy": 5.7997174739837645,
      "epoch": 3.2260421143102707,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042949749637917353,
      "loss": 5.487,
      "mean_token_accuracy": 0.18614423125982285,
      "num_tokens": 6823095.0,
      "step": 3755
    },
    {
      "entropy": 5.865487480163575,
      "epoch": 3.230339492909325,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004292556184983256,
      "loss": 5.5216,
      "mean_token_accuracy": 0.18153394907712936,
      "num_tokens": 6832195.0,
      "step": 3760
    },
    {
      "entropy": 5.8879210472106935,
      "epoch": 3.2346368715083798,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004290134037848623,
      "loss": 5.6524,
      "mean_token_accuracy": 0.17485911697149276,
      "num_tokens": 6840922.0,
      "step": 3765
    },
    {
      "entropy": 5.814740896224976,
      "epoch": 3.2389342501074343,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042877085276768386,
      "loss": 5.496,
      "mean_token_accuracy": 0.18620195537805556,
      "num_tokens": 6849182.0,
      "step": 3770
    },
    {
      "entropy": 5.8426494121551515,
      "epoch": 3.243231628706489,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042852796597642455,
      "loss": 5.4881,
      "mean_token_accuracy": 0.18392587453126907,
      "num_tokens": 6857932.0,
      "step": 3775
    },
    {
      "entropy": 5.930465126037598,
      "epoch": 3.247529007305544,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004282847439414522,
      "loss": 5.6207,
      "mean_token_accuracy": 0.16870806813240052,
      "num_tokens": 6867283.0,
      "step": 3780
    },
    {
      "entropy": 5.864709043502808,
      "epoch": 3.2518263859045984,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004280411871938664,
      "loss": 5.608,
      "mean_token_accuracy": 0.17931383848190308,
      "num_tokens": 6876123.0,
      "step": 3785
    },
    {
      "entropy": 5.877500772476196,
      "epoch": 3.256123764503653,
      "grad_norm": 1.25,
      "learning_rate": 0.0004277972962654979,
      "loss": 5.5472,
      "mean_token_accuracy": 0.17602882534265518,
      "num_tokens": 6885239.0,
      "step": 3790
    },
    {
      "entropy": 5.858795881271362,
      "epoch": 3.2604211431027075,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004275530716889069,
      "loss": 5.5937,
      "mean_token_accuracy": 0.171337129175663,
      "num_tokens": 6895061.0,
      "step": 3795
    },
    {
      "entropy": 5.898361396789551,
      "epoch": 3.264718521701762,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004273085139973822,
      "loss": 5.6602,
      "mean_token_accuracy": 0.16613161861896514,
      "num_tokens": 6903828.0,
      "step": 3800
    },
    {
      "entropy": 5.880637073516846,
      "epoch": 3.2690159003008166,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004270636237249401,
      "loss": 5.5461,
      "mean_token_accuracy": 0.17980466932058334,
      "num_tokens": 6912805.0,
      "step": 3805
    },
    {
      "entropy": 5.835609292984008,
      "epoch": 3.273313278899871,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042681840140632314,
      "loss": 5.6146,
      "mean_token_accuracy": 0.17179694026708603,
      "num_tokens": 6922165.0,
      "step": 3810
    },
    {
      "entropy": 5.856044054031372,
      "epoch": 3.2776106574989257,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004265728475769989,
      "loss": 5.5718,
      "mean_token_accuracy": 0.17612663954496383,
      "num_tokens": 6931677.0,
      "step": 3815
    },
    {
      "entropy": 5.847190523147583,
      "epoch": 3.28190803609798,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004263269627731586,
      "loss": 5.5376,
      "mean_token_accuracy": 0.1813596546649933,
      "num_tokens": 6940486.0,
      "step": 3820
    },
    {
      "entropy": 5.766942977905273,
      "epoch": 3.2862054146970348,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004260807475317164,
      "loss": 5.5387,
      "mean_token_accuracy": 0.1778130680322647,
      "num_tokens": 6948990.0,
      "step": 3825
    },
    {
      "entropy": 5.867075777053833,
      "epoch": 3.2905027932960893,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004258342023903081,
      "loss": 5.6853,
      "mean_token_accuracy": 0.1678419291973114,
      "num_tokens": 6959311.0,
      "step": 3830
    },
    {
      "entropy": 5.880772256851197,
      "epoch": 3.294800171895144,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042558732788728975,
      "loss": 5.458,
      "mean_token_accuracy": 0.18143573701381682,
      "num_tokens": 6968619.0,
      "step": 3835
    },
    {
      "entropy": 5.764143800735473,
      "epoch": 3.2990975504941984,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042534012456173643,
      "loss": 5.524,
      "mean_token_accuracy": 0.17863315343856812,
      "num_tokens": 6977469.0,
      "step": 3840
    },
    {
      "entropy": 5.848656415939331,
      "epoch": 3.303394929093253,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042509259295344157,
      "loss": 5.4816,
      "mean_token_accuracy": 0.17849756330251693,
      "num_tokens": 6986772.0,
      "step": 3845
    },
    {
      "entropy": 5.847779750823975,
      "epoch": 3.3076923076923075,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00042484473360291514,
      "loss": 5.5227,
      "mean_token_accuracy": 0.1756258174777031,
      "num_tokens": 6993937.0,
      "step": 3850
    },
    {
      "entropy": 5.738813400268555,
      "epoch": 3.311989686291362,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042459654705138294,
      "loss": 5.5679,
      "mean_token_accuracy": 0.1823994942009449,
      "num_tokens": 7003222.0,
      "step": 3855
    },
    {
      "entropy": 5.774629783630371,
      "epoch": 3.316287064890417,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004243480338407853,
      "loss": 5.525,
      "mean_token_accuracy": 0.18862827271223068,
      "num_tokens": 7012055.0,
      "step": 3860
    },
    {
      "entropy": 5.843663358688355,
      "epoch": 3.3205844434894716,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004240991945137755,
      "loss": 5.5162,
      "mean_token_accuracy": 0.18519170731306075,
      "num_tokens": 7021036.0,
      "step": 3865
    },
    {
      "entropy": 5.8147810935974125,
      "epoch": 3.324881822088526,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042385002961371944,
      "loss": 5.5325,
      "mean_token_accuracy": 0.18135716915130615,
      "num_tokens": 7030450.0,
      "step": 3870
    },
    {
      "entropy": 5.889339017868042,
      "epoch": 3.3291792006875807,
      "grad_norm": 1.25,
      "learning_rate": 0.0004236005396846935,
      "loss": 5.6222,
      "mean_token_accuracy": 0.17458055168390274,
      "num_tokens": 7039740.0,
      "step": 3875
    },
    {
      "entropy": 5.856438732147216,
      "epoch": 3.333476579286635,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042335072527148406,
      "loss": 5.6147,
      "mean_token_accuracy": 0.17771665155887603,
      "num_tokens": 7050430.0,
      "step": 3880
    },
    {
      "entropy": 5.803697156906128,
      "epoch": 3.3377739578856898,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004231005869195859,
      "loss": 5.5804,
      "mean_token_accuracy": 0.17390600442886353,
      "num_tokens": 7059477.0,
      "step": 3885
    },
    {
      "entropy": 5.7714502811431885,
      "epoch": 3.3420713364847443,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004228501251752011,
      "loss": 5.5353,
      "mean_token_accuracy": 0.18197794258594513,
      "num_tokens": 7067805.0,
      "step": 3890
    },
    {
      "entropy": 5.799749040603638,
      "epoch": 3.346368715083799,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042259934058523814,
      "loss": 5.5407,
      "mean_token_accuracy": 0.17531825453042985,
      "num_tokens": 7077606.0,
      "step": 3895
    },
    {
      "entropy": 5.843122863769532,
      "epoch": 3.3506660936828534,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042234823369731027,
      "loss": 5.5208,
      "mean_token_accuracy": 0.1831410214304924,
      "num_tokens": 7085647.0,
      "step": 3900
    },
    {
      "entropy": 5.75871639251709,
      "epoch": 3.354963472281908,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042209680505973465,
      "loss": 5.5217,
      "mean_token_accuracy": 0.18487536758184434,
      "num_tokens": 7095298.0,
      "step": 3905
    },
    {
      "entropy": 5.7882904529571535,
      "epoch": 3.3592608508809625,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004218450552215308,
      "loss": 5.5837,
      "mean_token_accuracy": 0.1783793792128563,
      "num_tokens": 7105207.0,
      "step": 3910
    },
    {
      "entropy": 5.808549976348877,
      "epoch": 3.363558229480017,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004215929847324199,
      "loss": 5.6568,
      "mean_token_accuracy": 0.17325044721364974,
      "num_tokens": 7114833.0,
      "step": 3915
    },
    {
      "entropy": 5.890059137344361,
      "epoch": 3.3678556080790716,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000421340594142823,
      "loss": 5.447,
      "mean_token_accuracy": 0.1935821980237961,
      "num_tokens": 7123608.0,
      "step": 3920
    },
    {
      "entropy": 5.772169685363769,
      "epoch": 3.3721529866781266,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042108788400386035,
      "loss": 5.5202,
      "mean_token_accuracy": 0.18366121798753737,
      "num_tokens": 7132250.0,
      "step": 3925
    },
    {
      "entropy": 5.777109003067016,
      "epoch": 3.376450365277181,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004208348548673498,
      "loss": 5.6247,
      "mean_token_accuracy": 0.17855982035398482,
      "num_tokens": 7142086.0,
      "step": 3930
    },
    {
      "entropy": 5.884216690063477,
      "epoch": 3.3807477438762357,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000420581507285806,
      "loss": 5.571,
      "mean_token_accuracy": 0.16913529187440873,
      "num_tokens": 7152434.0,
      "step": 3935
    },
    {
      "entropy": 5.7801055908203125,
      "epoch": 3.38504512247529,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004203278418124386,
      "loss": 5.5064,
      "mean_token_accuracy": 0.18806122690439225,
      "num_tokens": 7163041.0,
      "step": 3940
    },
    {
      "entropy": 5.74062147140503,
      "epoch": 3.3893425010743448,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004200738590011518,
      "loss": 5.4675,
      "mean_token_accuracy": 0.1898912951350212,
      "num_tokens": 7171875.0,
      "step": 3945
    },
    {
      "entropy": 5.762895250320435,
      "epoch": 3.3936398796733993,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041981955940654245,
      "loss": 5.6127,
      "mean_token_accuracy": 0.17519034296274186,
      "num_tokens": 7180803.0,
      "step": 3950
    },
    {
      "entropy": 5.843291091918945,
      "epoch": 3.397937258272454,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004195649435838992,
      "loss": 5.6222,
      "mean_token_accuracy": 0.16873589307069778,
      "num_tokens": 7190661.0,
      "step": 3955
    },
    {
      "entropy": 5.777584600448608,
      "epoch": 3.4022346368715084,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004193100120892013,
      "loss": 5.4614,
      "mean_token_accuracy": 0.1874678000807762,
      "num_tokens": 7199357.0,
      "step": 3960
    },
    {
      "entropy": 5.876493167877197,
      "epoch": 3.406532015470563,
      "grad_norm": 1.125,
      "learning_rate": 0.0004190547654791172,
      "loss": 5.6328,
      "mean_token_accuracy": 0.1726130172610283,
      "num_tokens": 7209856.0,
      "step": 3965
    },
    {
      "entropy": 5.870272159576416,
      "epoch": 3.4108293940696175,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041879920431100347,
      "loss": 5.6018,
      "mean_token_accuracy": 0.17143317461013793,
      "num_tokens": 7218778.0,
      "step": 3970
    },
    {
      "entropy": 5.822156190872192,
      "epoch": 3.415126772668672,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004185433291429036,
      "loss": 5.6106,
      "mean_token_accuracy": 0.18186013400554657,
      "num_tokens": 7228442.0,
      "step": 3975
    },
    {
      "entropy": 5.845083904266358,
      "epoch": 3.4194241512677266,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041828714053354665,
      "loss": 5.5896,
      "mean_token_accuracy": 0.17571255266666413,
      "num_tokens": 7238724.0,
      "step": 3980
    },
    {
      "entropy": 5.775502252578735,
      "epoch": 3.423721529866781,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004180306390423462,
      "loss": 5.5535,
      "mean_token_accuracy": 0.1877201244235039,
      "num_tokens": 7247844.0,
      "step": 3985
    },
    {
      "entropy": 5.8070330142974855,
      "epoch": 3.4280189084658357,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041777382522939884,
      "loss": 5.6153,
      "mean_token_accuracy": 0.1745489463210106,
      "num_tokens": 7257260.0,
      "step": 3990
    },
    {
      "entropy": 5.832814073562622,
      "epoch": 3.4323162870648902,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041751669965548344,
      "loss": 5.6241,
      "mean_token_accuracy": 0.17142881453037262,
      "num_tokens": 7266890.0,
      "step": 3995
    },
    {
      "entropy": 5.893997573852539,
      "epoch": 3.4366136656639448,
      "grad_norm": 1.3125,
      "learning_rate": 0.00041725926288205945,
      "loss": 5.645,
      "mean_token_accuracy": 0.17508225589990617,
      "num_tokens": 7276114.0,
      "step": 4000
    },
    {
      "epoch": 3.4366136656639448,
      "eval_entropy": 5.724426681930955,
      "eval_loss": 6.096716403961182,
      "eval_mean_token_accuracy": 0.16408700819823657,
      "eval_num_tokens": 7276114.0,
      "eval_runtime": 2.82,
      "eval_samples_per_second": 1258.515,
      "eval_steps_per_second": 157.447,
      "step": 4000
    },
    {
      "entropy": 5.880351543426514,
      "epoch": 3.4409110442629998,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004170015154712658,
      "loss": 5.5963,
      "mean_token_accuracy": 0.17532578706741334,
      "num_tokens": 7284426.0,
      "step": 4005
    },
    {
      "entropy": 5.781671094894409,
      "epoch": 3.4452084228620543,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041674345798591993,
      "loss": 5.6023,
      "mean_token_accuracy": 0.17392815500497819,
      "num_tokens": 7294813.0,
      "step": 4010
    },
    {
      "entropy": 5.859708070755005,
      "epoch": 3.449505801461109,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004164850909895161,
      "loss": 5.5912,
      "mean_token_accuracy": 0.18120552897453307,
      "num_tokens": 7304655.0,
      "step": 4015
    },
    {
      "entropy": 5.838734722137451,
      "epoch": 3.4538031800601634,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004162264150462247,
      "loss": 5.577,
      "mean_token_accuracy": 0.1782179594039917,
      "num_tokens": 7313610.0,
      "step": 4020
    },
    {
      "entropy": 5.843617343902588,
      "epoch": 3.458100558659218,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041596743072089065,
      "loss": 5.592,
      "mean_token_accuracy": 0.179537858068943,
      "num_tokens": 7322243.0,
      "step": 4025
    },
    {
      "entropy": 5.893010950088501,
      "epoch": 3.4623979372582725,
      "grad_norm": 1.234375,
      "learning_rate": 0.000415708138579032,
      "loss": 5.5483,
      "mean_token_accuracy": 0.17429945319890977,
      "num_tokens": 7331040.0,
      "step": 4030
    },
    {
      "entropy": 5.770739793777466,
      "epoch": 3.466695315857327,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041544853918683923,
      "loss": 5.6356,
      "mean_token_accuracy": 0.1719396620988846,
      "num_tokens": 7340771.0,
      "step": 4035
    },
    {
      "entropy": 5.845178556442261,
      "epoch": 3.4709926944563816,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004151886331111737,
      "loss": 5.6849,
      "mean_token_accuracy": 0.17274893969297409,
      "num_tokens": 7349960.0,
      "step": 4040
    },
    {
      "entropy": 5.8230266094207765,
      "epoch": 3.475290073055436,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041492842091956646,
      "loss": 5.5052,
      "mean_token_accuracy": 0.18835709244012833,
      "num_tokens": 7357983.0,
      "step": 4045
    },
    {
      "entropy": 5.875933647155762,
      "epoch": 3.4795874516544907,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004146679031802167,
      "loss": 5.6221,
      "mean_token_accuracy": 0.175259929895401,
      "num_tokens": 7366814.0,
      "step": 4050
    },
    {
      "entropy": 5.818189239501953,
      "epoch": 3.4838848302535452,
      "grad_norm": 1.25,
      "learning_rate": 0.00041440708046199123,
      "loss": 5.4981,
      "mean_token_accuracy": 0.17988317757844924,
      "num_tokens": 7374773.0,
      "step": 4055
    },
    {
      "entropy": 5.747901582717896,
      "epoch": 3.4881822088525998,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004141459533344226,
      "loss": 5.5971,
      "mean_token_accuracy": 0.17479410022497177,
      "num_tokens": 7383937.0,
      "step": 4060
    },
    {
      "entropy": 5.8878881454467775,
      "epoch": 3.4924795874516543,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041388452236770795,
      "loss": 5.5739,
      "mean_token_accuracy": 0.1789700284600258,
      "num_tokens": 7392577.0,
      "step": 4065
    },
    {
      "entropy": 5.707971525192261,
      "epoch": 3.4967769660507093,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041362278813270823,
      "loss": 5.4402,
      "mean_token_accuracy": 0.19540609121322633,
      "num_tokens": 7401473.0,
      "step": 4070
    },
    {
      "entropy": 5.833003520965576,
      "epoch": 3.501074344649764,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041336075120094616,
      "loss": 5.6653,
      "mean_token_accuracy": 0.16431416869163512,
      "num_tokens": 7410831.0,
      "step": 4075
    },
    {
      "entropy": 5.923135662078858,
      "epoch": 3.5053717232488184,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041309841214460586,
      "loss": 5.6432,
      "mean_token_accuracy": 0.1758934736251831,
      "num_tokens": 7421563.0,
      "step": 4080
    },
    {
      "entropy": 5.765736246109009,
      "epoch": 3.509669101847873,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004128357715365309,
      "loss": 5.5831,
      "mean_token_accuracy": 0.181057670712471,
      "num_tokens": 7430174.0,
      "step": 4085
    },
    {
      "entropy": 5.803967905044556,
      "epoch": 3.5139664804469275,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041257282995022345,
      "loss": 5.5472,
      "mean_token_accuracy": 0.18591351211071014,
      "num_tokens": 7439034.0,
      "step": 4090
    },
    {
      "entropy": 5.806196165084839,
      "epoch": 3.518263859045982,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004123095879598426,
      "loss": 5.5849,
      "mean_token_accuracy": 0.17558215856552123,
      "num_tokens": 7447663.0,
      "step": 4095
    },
    {
      "entropy": 5.827493810653687,
      "epoch": 3.5225612376450366,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041204604614020397,
      "loss": 5.6462,
      "mean_token_accuracy": 0.16851385831832885,
      "num_tokens": 7456615.0,
      "step": 4100
    },
    {
      "entropy": 5.842707204818725,
      "epoch": 3.526858616244091,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004117822050667773,
      "loss": 5.6801,
      "mean_token_accuracy": 0.1722650334239006,
      "num_tokens": 7466203.0,
      "step": 4105
    },
    {
      "entropy": 5.858455467224121,
      "epoch": 3.5311559948431457,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041151806531568617,
      "loss": 5.6202,
      "mean_token_accuracy": 0.17578027695417403,
      "num_tokens": 7475411.0,
      "step": 4110
    },
    {
      "entropy": 5.765643739700318,
      "epoch": 3.5354533734422002,
      "grad_norm": 1.125,
      "learning_rate": 0.00041125362746370625,
      "loss": 5.648,
      "mean_token_accuracy": 0.17153314203023912,
      "num_tokens": 7484965.0,
      "step": 4115
    },
    {
      "entropy": 5.908149671554566,
      "epoch": 3.5397507520412548,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004109888920882639,
      "loss": 5.5495,
      "mean_token_accuracy": 0.18733757436275483,
      "num_tokens": 7494240.0,
      "step": 4120
    },
    {
      "entropy": 5.779342555999756,
      "epoch": 3.5440481306403093,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004107238597674356,
      "loss": 5.5768,
      "mean_token_accuracy": 0.17890745103359224,
      "num_tokens": 7503560.0,
      "step": 4125
    },
    {
      "entropy": 5.811849117279053,
      "epoch": 3.548345509239364,
      "grad_norm": 1.09375,
      "learning_rate": 0.000410458531079946,
      "loss": 5.5258,
      "mean_token_accuracy": 0.18783323764801024,
      "num_tokens": 7512650.0,
      "step": 4130
    },
    {
      "entropy": 5.852642059326172,
      "epoch": 3.5526428878384184,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004101929066051668,
      "loss": 5.6434,
      "mean_token_accuracy": 0.1778715267777443,
      "num_tokens": 7521864.0,
      "step": 4135
    },
    {
      "entropy": 5.7974433422088625,
      "epoch": 3.556940266437473,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004099269869231157,
      "loss": 5.5289,
      "mean_token_accuracy": 0.18009612262248992,
      "num_tokens": 7531013.0,
      "step": 4140
    },
    {
      "entropy": 5.7702248096466064,
      "epoch": 3.5612376450365275,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040966077261445495,
      "loss": 5.5485,
      "mean_token_accuracy": 0.17787420004606247,
      "num_tokens": 7539959.0,
      "step": 4145
    },
    {
      "entropy": 5.869132089614868,
      "epoch": 3.565535023635582,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004093942642604904,
      "loss": 5.5076,
      "mean_token_accuracy": 0.18241044282913207,
      "num_tokens": 7548354.0,
      "step": 4150
    },
    {
      "entropy": 5.838557243347168,
      "epoch": 3.5698324022346366,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040912746244316944,
      "loss": 5.649,
      "mean_token_accuracy": 0.17207367718219757,
      "num_tokens": 7558321.0,
      "step": 4155
    },
    {
      "entropy": 5.766114044189453,
      "epoch": 3.5741297808336916,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040886036774508095,
      "loss": 5.5331,
      "mean_token_accuracy": 0.18266483396291733,
      "num_tokens": 7567889.0,
      "step": 4160
    },
    {
      "entropy": 5.813411521911621,
      "epoch": 3.578427159432746,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004085929807494527,
      "loss": 5.5994,
      "mean_token_accuracy": 0.1729404091835022,
      "num_tokens": 7576752.0,
      "step": 4165
    },
    {
      "entropy": 5.765122556686402,
      "epoch": 3.5827245380318007,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004083253020401512,
      "loss": 5.4992,
      "mean_token_accuracy": 0.19227544218301773,
      "num_tokens": 7585413.0,
      "step": 4170
    },
    {
      "entropy": 5.7991783618927,
      "epoch": 3.5870219166308552,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004080573322016797,
      "loss": 5.4253,
      "mean_token_accuracy": 0.19008248448371887,
      "num_tokens": 7593966.0,
      "step": 4175
    },
    {
      "entropy": 5.780904483795166,
      "epoch": 3.59131929522991,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004077890718191773,
      "loss": 5.4585,
      "mean_token_accuracy": 0.18521400392055512,
      "num_tokens": 7602746.0,
      "step": 4180
    },
    {
      "entropy": 5.791021680831909,
      "epoch": 3.5956166738289643,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040752052147841733,
      "loss": 5.5294,
      "mean_token_accuracy": 0.1835671216249466,
      "num_tokens": 7611245.0,
      "step": 4185
    },
    {
      "entropy": 5.783264636993408,
      "epoch": 3.599914052428019,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004072516817658065,
      "loss": 5.5565,
      "mean_token_accuracy": 0.18244687765836715,
      "num_tokens": 7620234.0,
      "step": 4190
    },
    {
      "entropy": 5.8208447933197025,
      "epoch": 3.6042114310270734,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004069825532683831,
      "loss": 5.5664,
      "mean_token_accuracy": 0.1806373953819275,
      "num_tokens": 7629794.0,
      "step": 4195
    },
    {
      "entropy": 5.719178962707519,
      "epoch": 3.608508809626128,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040671313657381645,
      "loss": 5.5155,
      "mean_token_accuracy": 0.1842898353934288,
      "num_tokens": 7639497.0,
      "step": 4200
    },
    {
      "entropy": 5.724102449417114,
      "epoch": 3.6128061882251825,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040644343227040473,
      "loss": 5.4735,
      "mean_token_accuracy": 0.17806130796670913,
      "num_tokens": 7647647.0,
      "step": 4205
    },
    {
      "entropy": 5.798212146759033,
      "epoch": 3.617103566824237,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004061734409470745,
      "loss": 5.6376,
      "mean_token_accuracy": 0.18280768990516663,
      "num_tokens": 7657988.0,
      "step": 4210
    },
    {
      "entropy": 5.8245879173278805,
      "epoch": 3.621400945423292,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004059031631933788,
      "loss": 5.5732,
      "mean_token_accuracy": 0.17606823295354843,
      "num_tokens": 7667498.0,
      "step": 4215
    },
    {
      "entropy": 5.805159425735473,
      "epoch": 3.6256983240223466,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040563259959949615,
      "loss": 5.7038,
      "mean_token_accuracy": 0.16953735053539276,
      "num_tokens": 7677386.0,
      "step": 4220
    },
    {
      "entropy": 5.910855340957641,
      "epoch": 3.629995702621401,
      "grad_norm": 1.125,
      "learning_rate": 0.0004053617507562295,
      "loss": 5.5174,
      "mean_token_accuracy": 0.1791599601507187,
      "num_tokens": 7686643.0,
      "step": 4225
    },
    {
      "entropy": 5.82626519203186,
      "epoch": 3.6342930812204557,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040509061725500426,
      "loss": 5.5654,
      "mean_token_accuracy": 0.1779528334736824,
      "num_tokens": 7695089.0,
      "step": 4230
    },
    {
      "entropy": 5.729827260971069,
      "epoch": 3.6385904598195102,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004048191996878677,
      "loss": 5.5652,
      "mean_token_accuracy": 0.18009263128042222,
      "num_tokens": 7703854.0,
      "step": 4235
    },
    {
      "entropy": 5.75171422958374,
      "epoch": 3.642887838418565,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040454749864748734,
      "loss": 5.4995,
      "mean_token_accuracy": 0.18293689638376237,
      "num_tokens": 7712903.0,
      "step": 4240
    },
    {
      "entropy": 5.778556203842163,
      "epoch": 3.6471852170176193,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004042755147271496,
      "loss": 5.4487,
      "mean_token_accuracy": 0.18358390927314758,
      "num_tokens": 7721701.0,
      "step": 4245
    },
    {
      "entropy": 5.686666059494018,
      "epoch": 3.651482595616674,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004040032485207587,
      "loss": 5.5674,
      "mean_token_accuracy": 0.17704793065786362,
      "num_tokens": 7731318.0,
      "step": 4250
    },
    {
      "entropy": 5.859571504592895,
      "epoch": 3.6557799742157284,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004037307006228352,
      "loss": 5.4962,
      "mean_token_accuracy": 0.18351848870515824,
      "num_tokens": 7740413.0,
      "step": 4255
    },
    {
      "entropy": 5.75723934173584,
      "epoch": 3.660077352814783,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004034578716285147,
      "loss": 5.4695,
      "mean_token_accuracy": 0.18348928391933442,
      "num_tokens": 7749054.0,
      "step": 4260
    },
    {
      "entropy": 5.755145740509033,
      "epoch": 3.6643747314138375,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004031847621335467,
      "loss": 5.5006,
      "mean_token_accuracy": 0.18805221617221832,
      "num_tokens": 7757366.0,
      "step": 4265
    },
    {
      "entropy": 5.7585211277008055,
      "epoch": 3.668672110012892,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004029113727342933,
      "loss": 5.5321,
      "mean_token_accuracy": 0.17868287563323976,
      "num_tokens": 7766471.0,
      "step": 4270
    },
    {
      "entropy": 5.718609428405761,
      "epoch": 3.6729694886119466,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040263770402772746,
      "loss": 5.5233,
      "mean_token_accuracy": 0.17831087112426758,
      "num_tokens": 7775920.0,
      "step": 4275
    },
    {
      "entropy": 5.813801431655884,
      "epoch": 3.677266867211001,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004023637566114325,
      "loss": 5.575,
      "mean_token_accuracy": 0.1784959316253662,
      "num_tokens": 7784530.0,
      "step": 4280
    },
    {
      "entropy": 5.800271368026733,
      "epoch": 3.6815642458100557,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004020895310835999,
      "loss": 5.507,
      "mean_token_accuracy": 0.1820111408829689,
      "num_tokens": 7793656.0,
      "step": 4285
    },
    {
      "entropy": 5.762936735153199,
      "epoch": 3.6858616244091102,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040181502804302865,
      "loss": 5.5471,
      "mean_token_accuracy": 0.17266241908073426,
      "num_tokens": 7802185.0,
      "step": 4290
    },
    {
      "entropy": 5.742608070373535,
      "epoch": 3.690159003008165,
      "grad_norm": 1.25,
      "learning_rate": 0.00040154024808912377,
      "loss": 5.5357,
      "mean_token_accuracy": 0.1886899083852768,
      "num_tokens": 7810345.0,
      "step": 4295
    },
    {
      "entropy": 5.798238611221313,
      "epoch": 3.6944563816072193,
      "grad_norm": 1.25,
      "learning_rate": 0.0004012651918218947,
      "loss": 5.5674,
      "mean_token_accuracy": 0.17927315086126328,
      "num_tokens": 7818998.0,
      "step": 4300
    },
    {
      "entropy": 5.828871011734009,
      "epoch": 3.6987537602062743,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004009898598419544,
      "loss": 5.6923,
      "mean_token_accuracy": 0.165433169901371,
      "num_tokens": 7828638.0,
      "step": 4305
    },
    {
      "entropy": 5.873749923706055,
      "epoch": 3.703051138805329,
      "grad_norm": 1.25,
      "learning_rate": 0.000400714252750518,
      "loss": 5.6536,
      "mean_token_accuracy": 0.17079921066761017,
      "num_tokens": 7838812.0,
      "step": 4310
    },
    {
      "entropy": 5.888559722900391,
      "epoch": 3.7073485174043834,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004004383711494011,
      "loss": 5.5668,
      "mean_token_accuracy": 0.18253673613071442,
      "num_tokens": 7847458.0,
      "step": 4315
    },
    {
      "entropy": 5.783564519882202,
      "epoch": 3.711645896003438,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004001622156410189,
      "loss": 5.5958,
      "mean_token_accuracy": 0.1734521970152855,
      "num_tokens": 7856553.0,
      "step": 4320
    },
    {
      "entropy": 5.746991920471191,
      "epoch": 3.7159432746024925,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039988578682838467,
      "loss": 5.5496,
      "mean_token_accuracy": 0.18276905119419098,
      "num_tokens": 7864788.0,
      "step": 4325
    },
    {
      "entropy": 5.8101146697998045,
      "epoch": 3.720240653201547,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039960908531510843,
      "loss": 5.5251,
      "mean_token_accuracy": 0.1838279977440834,
      "num_tokens": 7873850.0,
      "step": 4330
    },
    {
      "entropy": 5.881118488311768,
      "epoch": 3.7245380318006016,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003993321117053956,
      "loss": 5.6441,
      "mean_token_accuracy": 0.16995635628700256,
      "num_tokens": 7882775.0,
      "step": 4335
    },
    {
      "entropy": 5.876806831359863,
      "epoch": 3.728835410399656,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039905486660404604,
      "loss": 5.5875,
      "mean_token_accuracy": 0.17562626004219056,
      "num_tokens": 7890570.0,
      "step": 4340
    },
    {
      "entropy": 5.780924463272095,
      "epoch": 3.7331327889987107,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039877735061645206,
      "loss": 5.56,
      "mean_token_accuracy": 0.18864739835262298,
      "num_tokens": 7900090.0,
      "step": 4345
    },
    {
      "entropy": 5.838262844085693,
      "epoch": 3.7374301675977653,
      "grad_norm": 1.25,
      "learning_rate": 0.0003984995643485977,
      "loss": 5.5805,
      "mean_token_accuracy": 0.17841584086418152,
      "num_tokens": 7908077.0,
      "step": 4350
    },
    {
      "entropy": 5.835147142410278,
      "epoch": 3.74172754619682,
      "grad_norm": 1.453125,
      "learning_rate": 0.00039822150840705716,
      "loss": 5.5803,
      "mean_token_accuracy": 0.1835550993680954,
      "num_tokens": 7916290.0,
      "step": 4355
    },
    {
      "entropy": 5.859473943710327,
      "epoch": 3.746024924795875,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039794318339899347,
      "loss": 5.6631,
      "mean_token_accuracy": 0.16900220811367034,
      "num_tokens": 7925835.0,
      "step": 4360
    },
    {
      "entropy": 5.784291505813599,
      "epoch": 3.7503223033949293,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039766458993215726,
      "loss": 5.6233,
      "mean_token_accuracy": 0.17886178493499755,
      "num_tokens": 7935076.0,
      "step": 4365
    },
    {
      "entropy": 5.801282930374145,
      "epoch": 3.754619681993984,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039738572861488527,
      "loss": 5.5319,
      "mean_token_accuracy": 0.17598362416028976,
      "num_tokens": 7943958.0,
      "step": 4370
    },
    {
      "entropy": 5.756379842758179,
      "epoch": 3.7589170605930384,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000397106600056099,
      "loss": 5.57,
      "mean_token_accuracy": 0.17566176503896713,
      "num_tokens": 7953189.0,
      "step": 4375
    },
    {
      "entropy": 5.765503168106079,
      "epoch": 3.763214439192093,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003968272048653039,
      "loss": 5.485,
      "mean_token_accuracy": 0.1919957995414734,
      "num_tokens": 7962927.0,
      "step": 4380
    },
    {
      "entropy": 5.67205696105957,
      "epoch": 3.7675118177911475,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003965475436525873,
      "loss": 5.4991,
      "mean_token_accuracy": 0.18613611906766891,
      "num_tokens": 7973087.0,
      "step": 4385
    },
    {
      "entropy": 5.736862230300903,
      "epoch": 3.771809196390202,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003962676170286174,
      "loss": 5.4622,
      "mean_token_accuracy": 0.18434722125530242,
      "num_tokens": 7982535.0,
      "step": 4390
    },
    {
      "entropy": 5.830058288574219,
      "epoch": 3.7761065749892566,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039598742560464223,
      "loss": 5.5533,
      "mean_token_accuracy": 0.1847641110420227,
      "num_tokens": 7990740.0,
      "step": 4395
    },
    {
      "entropy": 5.7972174167633055,
      "epoch": 3.780403953588311,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003957069699924877,
      "loss": 5.5605,
      "mean_token_accuracy": 0.18067938089370728,
      "num_tokens": 7999349.0,
      "step": 4400
    },
    {
      "entropy": 5.782556962966919,
      "epoch": 3.7847013321873657,
      "grad_norm": 1.1875,
      "learning_rate": 0.000395426250804557,
      "loss": 5.5483,
      "mean_token_accuracy": 0.18622617721557616,
      "num_tokens": 8007615.0,
      "step": 4405
    },
    {
      "entropy": 5.776251268386841,
      "epoch": 3.7889987107864203,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039514526865382847,
      "loss": 5.5413,
      "mean_token_accuracy": 0.1839849904179573,
      "num_tokens": 8017545.0,
      "step": 4410
    },
    {
      "entropy": 5.755090188980103,
      "epoch": 3.793296089385475,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003948640241538548,
      "loss": 5.4805,
      "mean_token_accuracy": 0.18133825063705444,
      "num_tokens": 8026381.0,
      "step": 4415
    },
    {
      "entropy": 5.772369623184204,
      "epoch": 3.7975934679845293,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003945825179187617,
      "loss": 5.5865,
      "mean_token_accuracy": 0.1726389318704605,
      "num_tokens": 8034745.0,
      "step": 4420
    },
    {
      "entropy": 5.796453189849854,
      "epoch": 3.801890846583584,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039430075056324604,
      "loss": 5.5172,
      "mean_token_accuracy": 0.188214536011219,
      "num_tokens": 8043995.0,
      "step": 4425
    },
    {
      "entropy": 5.82302565574646,
      "epoch": 3.8061882251826384,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039401872270257546,
      "loss": 5.6113,
      "mean_token_accuracy": 0.1751917839050293,
      "num_tokens": 8053059.0,
      "step": 4430
    },
    {
      "entropy": 5.781544780731201,
      "epoch": 3.810485603781693,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039373643495258567,
      "loss": 5.6454,
      "mean_token_accuracy": 0.18245631456375122,
      "num_tokens": 8062160.0,
      "step": 4435
    },
    {
      "entropy": 5.768618774414063,
      "epoch": 3.8147829823807475,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039345388792968056,
      "loss": 5.533,
      "mean_token_accuracy": 0.19349936544895172,
      "num_tokens": 8071260.0,
      "step": 4440
    },
    {
      "entropy": 5.830617380142212,
      "epoch": 3.819080360979802,
      "grad_norm": 1.5625,
      "learning_rate": 0.00039317108225082984,
      "loss": 5.6563,
      "mean_token_accuracy": 0.17359688729047776,
      "num_tokens": 8081540.0,
      "step": 4445
    },
    {
      "entropy": 5.788396501541138,
      "epoch": 3.8233777395788566,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039288801853356806,
      "loss": 5.6214,
      "mean_token_accuracy": 0.17893288731575013,
      "num_tokens": 8089785.0,
      "step": 4450
    },
    {
      "entropy": 5.8474733352661135,
      "epoch": 3.8276751181779116,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003926046973959932,
      "loss": 5.4445,
      "mean_token_accuracy": 0.18922778815031052,
      "num_tokens": 8098097.0,
      "step": 4455
    },
    {
      "entropy": 5.745357370376587,
      "epoch": 3.831972496776966,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003923211194567654,
      "loss": 5.6815,
      "mean_token_accuracy": 0.17651993036270142,
      "num_tokens": 8108693.0,
      "step": 4460
    },
    {
      "entropy": 5.785046243667603,
      "epoch": 3.8362698753760207,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039203728533510556,
      "loss": 5.5264,
      "mean_token_accuracy": 0.18122401535511018,
      "num_tokens": 8117181.0,
      "step": 4465
    },
    {
      "entropy": 5.742541551589966,
      "epoch": 3.8405672539750753,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000391753195650794,
      "loss": 5.5377,
      "mean_token_accuracy": 0.17964441180229188,
      "num_tokens": 8125398.0,
      "step": 4470
    },
    {
      "entropy": 5.804271411895752,
      "epoch": 3.84486463257413,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039146885102416895,
      "loss": 5.5527,
      "mean_token_accuracy": 0.18320963382720948,
      "num_tokens": 8135320.0,
      "step": 4475
    },
    {
      "entropy": 5.769350910186768,
      "epoch": 3.8491620111731844,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039118425207612553,
      "loss": 5.626,
      "mean_token_accuracy": 0.17700641453266144,
      "num_tokens": 8144320.0,
      "step": 4480
    },
    {
      "entropy": 5.683031034469605,
      "epoch": 3.853459389772239,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039089939942811396,
      "loss": 5.4879,
      "mean_token_accuracy": 0.18756311386823654,
      "num_tokens": 8153653.0,
      "step": 4485
    },
    {
      "entropy": 5.874190759658814,
      "epoch": 3.8577567683712934,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039061429370213863,
      "loss": 5.5523,
      "mean_token_accuracy": 0.18109647035598755,
      "num_tokens": 8162741.0,
      "step": 4490
    },
    {
      "entropy": 5.684871435165405,
      "epoch": 3.862054146970348,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039032893552075646,
      "loss": 5.4519,
      "mean_token_accuracy": 0.1927460551261902,
      "num_tokens": 8171078.0,
      "step": 4495
    },
    {
      "entropy": 5.679884243011474,
      "epoch": 3.8663515255694025,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003900433255070758,
      "loss": 5.5379,
      "mean_token_accuracy": 0.1865083694458008,
      "num_tokens": 8179968.0,
      "step": 4500
    },
    {
      "epoch": 3.8663515255694025,
      "eval_entropy": 5.751085638999939,
      "eval_loss": 6.0441975593566895,
      "eval_mean_token_accuracy": 0.16676933249576134,
      "eval_num_tokens": 8179968.0,
      "eval_runtime": 2.6275,
      "eval_samples_per_second": 1350.698,
      "eval_steps_per_second": 168.98,
      "step": 4500
    },
    {
      "entropy": 5.846665334701538,
      "epoch": 3.870648904168457,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038975746428475454,
      "loss": 5.4939,
      "mean_token_accuracy": 0.18137868642807006,
      "num_tokens": 8189261.0,
      "step": 4505
    },
    {
      "entropy": 5.816416263580322,
      "epoch": 3.874946282767512,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038947135247799955,
      "loss": 5.524,
      "mean_token_accuracy": 0.18105199337005615,
      "num_tokens": 8198302.0,
      "step": 4510
    },
    {
      "entropy": 5.7392942905426025,
      "epoch": 3.8792436613665666,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038918499071156443,
      "loss": 5.509,
      "mean_token_accuracy": 0.18849926739931105,
      "num_tokens": 8207098.0,
      "step": 4515
    },
    {
      "entropy": 5.859800863265991,
      "epoch": 3.883541039965621,
      "grad_norm": 1.28125,
      "learning_rate": 0.000388898379610749,
      "loss": 5.5683,
      "mean_token_accuracy": 0.17910422682762145,
      "num_tokens": 8216831.0,
      "step": 4520
    },
    {
      "entropy": 5.73723258972168,
      "epoch": 3.8878384185646757,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003886115198013973,
      "loss": 5.5688,
      "mean_token_accuracy": 0.18406275063753127,
      "num_tokens": 8225369.0,
      "step": 4525
    },
    {
      "entropy": 5.7593433380126955,
      "epoch": 3.8921357971637303,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003883244119098965,
      "loss": 5.6804,
      "mean_token_accuracy": 0.17068565785884857,
      "num_tokens": 8234440.0,
      "step": 4530
    },
    {
      "entropy": 5.809617424011231,
      "epoch": 3.896433175762785,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003880370565631754,
      "loss": 5.4782,
      "mean_token_accuracy": 0.18485231101512908,
      "num_tokens": 8243707.0,
      "step": 4535
    },
    {
      "entropy": 5.784097576141358,
      "epoch": 3.9007305543618394,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038774945438870337,
      "loss": 5.6482,
      "mean_token_accuracy": 0.17690181583166123,
      "num_tokens": 8254223.0,
      "step": 4540
    },
    {
      "entropy": 5.729510307312012,
      "epoch": 3.905027932960894,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038746160601448845,
      "loss": 5.4894,
      "mean_token_accuracy": 0.18488709330558778,
      "num_tokens": 8263105.0,
      "step": 4545
    },
    {
      "entropy": 5.777517700195313,
      "epoch": 3.9093253115599484,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003871735120690766,
      "loss": 5.5618,
      "mean_token_accuracy": 0.1846500426530838,
      "num_tokens": 8271478.0,
      "step": 4550
    },
    {
      "entropy": 5.824449872970581,
      "epoch": 3.913622690159003,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003868851731815497,
      "loss": 5.5989,
      "mean_token_accuracy": 0.17459371984004973,
      "num_tokens": 8280396.0,
      "step": 4555
    },
    {
      "entropy": 5.786659526824951,
      "epoch": 3.9179200687580575,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003865965899815247,
      "loss": 5.6006,
      "mean_token_accuracy": 0.17747600972652436,
      "num_tokens": 8290371.0,
      "step": 4560
    },
    {
      "entropy": 5.765139436721801,
      "epoch": 3.922217447357112,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003863077630991518,
      "loss": 5.4819,
      "mean_token_accuracy": 0.19038917124271393,
      "num_tokens": 8298976.0,
      "step": 4565
    },
    {
      "entropy": 5.781538963317871,
      "epoch": 3.9265148259561666,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003860186931651139,
      "loss": 5.5506,
      "mean_token_accuracy": 0.17990471571683883,
      "num_tokens": 8308752.0,
      "step": 4570
    },
    {
      "entropy": 5.796289443969727,
      "epoch": 3.930812204555221,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003857293808106238,
      "loss": 5.6226,
      "mean_token_accuracy": 0.17713063359260559,
      "num_tokens": 8317343.0,
      "step": 4575
    },
    {
      "entropy": 5.793017721176147,
      "epoch": 3.9351095831542757,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003854398266674241,
      "loss": 5.4421,
      "mean_token_accuracy": 0.18935199081897736,
      "num_tokens": 8326956.0,
      "step": 4580
    },
    {
      "entropy": 5.765391302108765,
      "epoch": 3.9394069617533303,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038515003136778544,
      "loss": 5.5775,
      "mean_token_accuracy": 0.1761196181178093,
      "num_tokens": 8335589.0,
      "step": 4585
    },
    {
      "entropy": 5.738382148742676,
      "epoch": 3.943704340352385,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038485999554450483,
      "loss": 5.5532,
      "mean_token_accuracy": 0.17819496840238572,
      "num_tokens": 8345517.0,
      "step": 4590
    },
    {
      "entropy": 5.65023365020752,
      "epoch": 3.9480017189514394,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038456971983090454,
      "loss": 5.4735,
      "mean_token_accuracy": 0.19519156962633133,
      "num_tokens": 8354702.0,
      "step": 4595
    },
    {
      "entropy": 5.762531280517578,
      "epoch": 3.9522990975504944,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003842792048608309,
      "loss": 5.5117,
      "mean_token_accuracy": 0.18563084304332733,
      "num_tokens": 8362940.0,
      "step": 4600
    },
    {
      "entropy": 5.837363004684448,
      "epoch": 3.956596476149549,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003839884512686523,
      "loss": 5.5702,
      "mean_token_accuracy": 0.18022384494543076,
      "num_tokens": 8372034.0,
      "step": 4605
    },
    {
      "entropy": 5.73058123588562,
      "epoch": 3.9608938547486034,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038369745968925846,
      "loss": 5.5893,
      "mean_token_accuracy": 0.17511263936758042,
      "num_tokens": 8381673.0,
      "step": 4610
    },
    {
      "entropy": 5.852107429504395,
      "epoch": 3.965191233347658,
      "grad_norm": 1.125,
      "learning_rate": 0.00038340623075805875,
      "loss": 5.5235,
      "mean_token_accuracy": 0.18120430856943132,
      "num_tokens": 8390804.0,
      "step": 4615
    },
    {
      "entropy": 5.858357334136963,
      "epoch": 3.9694886119467125,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038311476511098053,
      "loss": 5.5799,
      "mean_token_accuracy": 0.18346985429525375,
      "num_tokens": 8399644.0,
      "step": 4620
    },
    {
      "entropy": 5.723837566375733,
      "epoch": 3.973785990545767,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003828230633844685,
      "loss": 5.5773,
      "mean_token_accuracy": 0.1784892797470093,
      "num_tokens": 8409264.0,
      "step": 4625
    },
    {
      "entropy": 5.8094841003417965,
      "epoch": 3.9780833691448216,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038253112621548243,
      "loss": 5.5543,
      "mean_token_accuracy": 0.17779970914125443,
      "num_tokens": 8418383.0,
      "step": 4630
    },
    {
      "entropy": 5.802743577957154,
      "epoch": 3.982380747743876,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003822389542414966,
      "loss": 5.5427,
      "mean_token_accuracy": 0.17908314913511275,
      "num_tokens": 8427411.0,
      "step": 4635
    },
    {
      "entropy": 5.7805798053741455,
      "epoch": 3.9866781263429307,
      "grad_norm": 1.25,
      "learning_rate": 0.00038194654810049775,
      "loss": 5.5106,
      "mean_token_accuracy": 0.17903167456388475,
      "num_tokens": 8435537.0,
      "step": 4640
    },
    {
      "entropy": 5.763886308670044,
      "epoch": 3.9909755049419853,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000381653908430984,
      "loss": 5.5807,
      "mean_token_accuracy": 0.17593914568424224,
      "num_tokens": 8444400.0,
      "step": 4645
    },
    {
      "entropy": 5.824175500869751,
      "epoch": 3.99527288354104,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003813610358719634,
      "loss": 5.5658,
      "mean_token_accuracy": 0.18078377544879914,
      "num_tokens": 8453830.0,
      "step": 4650
    },
    {
      "entropy": 5.760702276229859,
      "epoch": 3.999570262140095,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038106793106295266,
      "loss": 5.5132,
      "mean_token_accuracy": 0.1935229018330574,
      "num_tokens": 8463033.0,
      "step": 4655
    },
    {
      "entropy": 5.71132246653239,
      "epoch": 4.003437902879243,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003807745946439754,
      "loss": 5.2938,
      "mean_token_accuracy": 0.20026440090603298,
      "num_tokens": 8470740.0,
      "step": 4660
    },
    {
      "entropy": 5.773604536056519,
      "epoch": 4.007735281478298,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003804810272555612,
      "loss": 5.2614,
      "mean_token_accuracy": 0.18773122727870942,
      "num_tokens": 8480480.0,
      "step": 4665
    },
    {
      "entropy": 5.669144868850708,
      "epoch": 4.012032660077352,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003801872295387439,
      "loss": 5.2037,
      "mean_token_accuracy": 0.2023132860660553,
      "num_tokens": 8489047.0,
      "step": 4670
    },
    {
      "entropy": 5.783167695999145,
      "epoch": 4.016330038676408,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003798932021350603,
      "loss": 5.2975,
      "mean_token_accuracy": 0.19550290554761887,
      "num_tokens": 8497763.0,
      "step": 4675
    },
    {
      "entropy": 5.785816287994384,
      "epoch": 4.020627417275462,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037959894568654864,
      "loss": 5.2799,
      "mean_token_accuracy": 0.1990286663174629,
      "num_tokens": 8506814.0,
      "step": 4680
    },
    {
      "entropy": 5.8259687423706055,
      "epoch": 4.024924795874517,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003793044608357474,
      "loss": 5.3884,
      "mean_token_accuracy": 0.18091341406106948,
      "num_tokens": 8516384.0,
      "step": 4685
    },
    {
      "entropy": 5.772263669967652,
      "epoch": 4.0292221744735714,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003790097482256939,
      "loss": 5.2203,
      "mean_token_accuracy": 0.19790298342704774,
      "num_tokens": 8524822.0,
      "step": 4690
    },
    {
      "entropy": 5.775947952270508,
      "epoch": 4.033519553072626,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003787148084999225,
      "loss": 5.2587,
      "mean_token_accuracy": 0.19575134813785552,
      "num_tokens": 8534129.0,
      "step": 4695
    },
    {
      "entropy": 5.706704425811767,
      "epoch": 4.0378169316716805,
      "grad_norm": 1.25,
      "learning_rate": 0.00037841964230246394,
      "loss": 5.3354,
      "mean_token_accuracy": 0.18651147931814194,
      "num_tokens": 8543235.0,
      "step": 4700
    },
    {
      "entropy": 5.685120868682861,
      "epoch": 4.042114310270735,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003781242502778429,
      "loss": 5.2203,
      "mean_token_accuracy": 0.20626447796821595,
      "num_tokens": 8551903.0,
      "step": 4705
    },
    {
      "entropy": 5.704382467269897,
      "epoch": 4.04641168886979,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037782863307107785,
      "loss": 5.3186,
      "mean_token_accuracy": 0.19430158734321595,
      "num_tokens": 8561173.0,
      "step": 4710
    },
    {
      "entropy": 5.760441493988037,
      "epoch": 4.050709067468844,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037753279132767833,
      "loss": 5.2118,
      "mean_token_accuracy": 0.20544237345457078,
      "num_tokens": 8569789.0,
      "step": 4715
    },
    {
      "entropy": 5.705115795135498,
      "epoch": 4.055006446067899,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00037723672569364453,
      "loss": 5.2225,
      "mean_token_accuracy": 0.2015332341194153,
      "num_tokens": 8577971.0,
      "step": 4720
    },
    {
      "entropy": 5.76477427482605,
      "epoch": 4.059303824666953,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037694043681546545,
      "loss": 5.2905,
      "mean_token_accuracy": 0.19610754549503326,
      "num_tokens": 8587299.0,
      "step": 4725
    },
    {
      "entropy": 5.664056777954102,
      "epoch": 4.063601203266008,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003766439253401177,
      "loss": 5.2594,
      "mean_token_accuracy": 0.19104208797216415,
      "num_tokens": 8595813.0,
      "step": 4730
    },
    {
      "entropy": 5.686329698562622,
      "epoch": 4.067898581865062,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037634719191506367,
      "loss": 5.2749,
      "mean_token_accuracy": 0.19637089520692824,
      "num_tokens": 8604552.0,
      "step": 4735
    },
    {
      "entropy": 5.690449476242065,
      "epoch": 4.072195960464117,
      "grad_norm": 1.359375,
      "learning_rate": 0.00037605023718825065,
      "loss": 5.2052,
      "mean_token_accuracy": 0.1984548658132553,
      "num_tokens": 8612701.0,
      "step": 4740
    },
    {
      "entropy": 5.743429183959961,
      "epoch": 4.0764933390631715,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000375753061808109,
      "loss": 5.271,
      "mean_token_accuracy": 0.19934578984975815,
      "num_tokens": 8622699.0,
      "step": 4745
    },
    {
      "entropy": 5.682605361938476,
      "epoch": 4.080790717662226,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037545566642355107,
      "loss": 5.2518,
      "mean_token_accuracy": 0.19045818001031875,
      "num_tokens": 8631821.0,
      "step": 4750
    },
    {
      "entropy": 5.70823187828064,
      "epoch": 4.0850880962612806,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003751580516839695,
      "loss": 5.2212,
      "mean_token_accuracy": 0.1951994478702545,
      "num_tokens": 8641814.0,
      "step": 4755
    },
    {
      "entropy": 5.777681016921997,
      "epoch": 4.089385474860335,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037486021823923574,
      "loss": 5.3044,
      "mean_token_accuracy": 0.1917086973786354,
      "num_tokens": 8649649.0,
      "step": 4760
    },
    {
      "entropy": 5.661882781982422,
      "epoch": 4.09368285345939,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037456216673969925,
      "loss": 5.2393,
      "mean_token_accuracy": 0.20246082097291945,
      "num_tokens": 8658216.0,
      "step": 4765
    },
    {
      "entropy": 5.748642110824585,
      "epoch": 4.097980232058444,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003742638978361851,
      "loss": 5.3067,
      "mean_token_accuracy": 0.19130209237337112,
      "num_tokens": 8667725.0,
      "step": 4770
    },
    {
      "entropy": 5.642697048187256,
      "epoch": 4.102277610657499,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037396541217999367,
      "loss": 5.1866,
      "mean_token_accuracy": 0.19986732602119445,
      "num_tokens": 8675739.0,
      "step": 4775
    },
    {
      "entropy": 5.649731636047363,
      "epoch": 4.106574989256553,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003736667104228981,
      "loss": 5.2405,
      "mean_token_accuracy": 0.1994873806834221,
      "num_tokens": 8685764.0,
      "step": 4780
    },
    {
      "entropy": 5.7156013488769535,
      "epoch": 4.110872367855608,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00037336779321714376,
      "loss": 5.2247,
      "mean_token_accuracy": 0.20834742188453675,
      "num_tokens": 8695476.0,
      "step": 4785
    },
    {
      "entropy": 5.641605615615845,
      "epoch": 4.115169746454662,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037306866121544633,
      "loss": 5.2637,
      "mean_token_accuracy": 0.19882164746522904,
      "num_tokens": 8705544.0,
      "step": 4790
    },
    {
      "entropy": 5.723520755767822,
      "epoch": 4.119467125053717,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003727693150709904,
      "loss": 5.3007,
      "mean_token_accuracy": 0.19622315913438798,
      "num_tokens": 8714883.0,
      "step": 4795
    },
    {
      "entropy": 5.781742286682129,
      "epoch": 4.1237645036527715,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00037246975543742843,
      "loss": 5.324,
      "mean_token_accuracy": 0.19419382363557816,
      "num_tokens": 8724589.0,
      "step": 4800
    },
    {
      "entropy": 5.631624698638916,
      "epoch": 4.128061882251826,
      "grad_norm": 1.203125,
      "learning_rate": 0.000372169982968879,
      "loss": 5.2078,
      "mean_token_accuracy": 0.20091456919908524,
      "num_tokens": 8733771.0,
      "step": 4805
    },
    {
      "entropy": 5.6701267719268795,
      "epoch": 4.132359260850881,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003718699983199252,
      "loss": 5.3011,
      "mean_token_accuracy": 0.1958358734846115,
      "num_tokens": 8742348.0,
      "step": 4810
    },
    {
      "entropy": 5.648713874816894,
      "epoch": 4.136656639449935,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003715698021456137,
      "loss": 5.227,
      "mean_token_accuracy": 0.20465355664491652,
      "num_tokens": 8751357.0,
      "step": 4815
    },
    {
      "entropy": 5.644321727752685,
      "epoch": 4.1409540180489905,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037126939510145294,
      "loss": 5.2809,
      "mean_token_accuracy": 0.1983148753643036,
      "num_tokens": 8760813.0,
      "step": 4820
    },
    {
      "entropy": 5.771704578399659,
      "epoch": 4.145251396648045,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003709687778434118,
      "loss": 5.3318,
      "mean_token_accuracy": 0.18711795806884765,
      "num_tokens": 8770228.0,
      "step": 4825
    },
    {
      "entropy": 5.693470478057861,
      "epoch": 4.1495487752471,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003706679510279183,
      "loss": 5.1638,
      "mean_token_accuracy": 0.20280520021915435,
      "num_tokens": 8779351.0,
      "step": 4830
    },
    {
      "entropy": 5.654620170593262,
      "epoch": 4.153846153846154,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003703669153118578,
      "loss": 5.3193,
      "mean_token_accuracy": 0.1901832863688469,
      "num_tokens": 8789116.0,
      "step": 4835
    },
    {
      "entropy": 5.647818756103516,
      "epoch": 4.158143532445209,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037006567135257216,
      "loss": 5.2977,
      "mean_token_accuracy": 0.19251067489385604,
      "num_tokens": 8797790.0,
      "step": 4840
    },
    {
      "entropy": 5.743458795547485,
      "epoch": 4.162440911044263,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036976421980785764,
      "loss": 5.338,
      "mean_token_accuracy": 0.1877649739384651,
      "num_tokens": 8808067.0,
      "step": 4845
    },
    {
      "entropy": 5.671063613891602,
      "epoch": 4.166738289643318,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003694625613359641,
      "loss": 5.2344,
      "mean_token_accuracy": 0.20174112319946289,
      "num_tokens": 8816587.0,
      "step": 4850
    },
    {
      "entropy": 5.723738718032837,
      "epoch": 4.171035668242372,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003691606965955929,
      "loss": 5.3039,
      "mean_token_accuracy": 0.19654205739498137,
      "num_tokens": 8826045.0,
      "step": 4855
    },
    {
      "entropy": 5.616712617874145,
      "epoch": 4.175333046841427,
      "grad_norm": 1.203125,
      "learning_rate": 0.000368858626245896,
      "loss": 5.2869,
      "mean_token_accuracy": 0.1938377410173416,
      "num_tokens": 8835427.0,
      "step": 4860
    },
    {
      "entropy": 5.656343650817871,
      "epoch": 4.1796304254404815,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003685563509464744,
      "loss": 5.2233,
      "mean_token_accuracy": 0.20234143882989883,
      "num_tokens": 8845167.0,
      "step": 4865
    },
    {
      "entropy": 5.763440036773682,
      "epoch": 4.183927804039536,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036825387135737647,
      "loss": 5.2474,
      "mean_token_accuracy": 0.1991298496723175,
      "num_tokens": 8853591.0,
      "step": 4870
    },
    {
      "entropy": 5.633469724655152,
      "epoch": 4.188225182638591,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036795118813909674,
      "loss": 5.3594,
      "mean_token_accuracy": 0.18792198300361634,
      "num_tokens": 8863647.0,
      "step": 4875
    },
    {
      "entropy": 5.766976833343506,
      "epoch": 4.192522561237645,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036764830195257437,
      "loss": 5.2759,
      "mean_token_accuracy": 0.19406790137290955,
      "num_tokens": 8872911.0,
      "step": 4880
    },
    {
      "entropy": 5.739557075500488,
      "epoch": 4.1968199398367,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003673452134591918,
      "loss": 5.3213,
      "mean_token_accuracy": 0.19075183123350142,
      "num_tokens": 8881001.0,
      "step": 4885
    },
    {
      "entropy": 5.643009233474731,
      "epoch": 4.201117318435754,
      "grad_norm": 1.296875,
      "learning_rate": 0.000367041923320773,
      "loss": 5.2021,
      "mean_token_accuracy": 0.20086978524923324,
      "num_tokens": 8890323.0,
      "step": 4890
    },
    {
      "entropy": 5.687097406387329,
      "epoch": 4.205414697034809,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036673843219958257,
      "loss": 5.266,
      "mean_token_accuracy": 0.2008528783917427,
      "num_tokens": 8900471.0,
      "step": 4895
    },
    {
      "entropy": 5.7050456523895265,
      "epoch": 4.209712075633863,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003664347407583238,
      "loss": 5.3126,
      "mean_token_accuracy": 0.19241509586572647,
      "num_tokens": 8909320.0,
      "step": 4900
    },
    {
      "entropy": 5.657163476943969,
      "epoch": 4.214009454232918,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003661308496601373,
      "loss": 5.2436,
      "mean_token_accuracy": 0.2002905011177063,
      "num_tokens": 8917453.0,
      "step": 4905
    },
    {
      "entropy": 5.702735042572021,
      "epoch": 4.218306832831972,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00036582675956859983,
      "loss": 5.2989,
      "mean_token_accuracy": 0.19501191824674607,
      "num_tokens": 8925737.0,
      "step": 4910
    },
    {
      "entropy": 5.641617774963379,
      "epoch": 4.222604211431027,
      "grad_norm": 1.25,
      "learning_rate": 0.00036552247114772263,
      "loss": 5.2384,
      "mean_token_accuracy": 0.1976437360048294,
      "num_tokens": 8935475.0,
      "step": 4915
    },
    {
      "entropy": 5.7294700145721436,
      "epoch": 4.2269015900300815,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036521798506194996,
      "loss": 5.2671,
      "mean_token_accuracy": 0.2042749524116516,
      "num_tokens": 8944683.0,
      "step": 4920
    },
    {
      "entropy": 5.683671855926514,
      "epoch": 4.231198968629136,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00036491330197615775,
      "loss": 5.3034,
      "mean_token_accuracy": 0.1903869852423668,
      "num_tokens": 8953837.0,
      "step": 4925
    },
    {
      "entropy": 5.680200624465942,
      "epoch": 4.235496347228191,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036460842255565197,
      "loss": 5.3401,
      "mean_token_accuracy": 0.19242161512374878,
      "num_tokens": 8964822.0,
      "step": 4930
    },
    {
      "entropy": 5.6954674243927,
      "epoch": 4.239793725827245,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003643033474661676,
      "loss": 5.3054,
      "mean_token_accuracy": 0.19037318974733353,
      "num_tokens": 8974363.0,
      "step": 4935
    },
    {
      "entropy": 5.71783938407898,
      "epoch": 4.2440911044263,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036399807737386657,
      "loss": 5.2287,
      "mean_token_accuracy": 0.19979746341705323,
      "num_tokens": 8983122.0,
      "step": 4940
    },
    {
      "entropy": 5.729825067520141,
      "epoch": 4.248388483025354,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003636926129453368,
      "loss": 5.3545,
      "mean_token_accuracy": 0.18944738805294037,
      "num_tokens": 8991618.0,
      "step": 4945
    },
    {
      "entropy": 5.697331809997559,
      "epoch": 4.252685861624409,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003633869548475904,
      "loss": 5.2405,
      "mean_token_accuracy": 0.19826923310756683,
      "num_tokens": 9000128.0,
      "step": 4950
    },
    {
      "entropy": 5.654035568237305,
      "epoch": 4.256983240223463,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003630811037480627,
      "loss": 5.2506,
      "mean_token_accuracy": 0.19913761019706727,
      "num_tokens": 9008951.0,
      "step": 4955
    },
    {
      "entropy": 5.707585716247559,
      "epoch": 4.261280618822518,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003627750603146101,
      "loss": 5.2868,
      "mean_token_accuracy": 0.19083305597305297,
      "num_tokens": 9018949.0,
      "step": 4960
    },
    {
      "entropy": 5.706443881988525,
      "epoch": 4.265577997421573,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003624688252155091,
      "loss": 5.2825,
      "mean_token_accuracy": 0.19386855512857437,
      "num_tokens": 9028910.0,
      "step": 4965
    },
    {
      "entropy": 5.7162127017974855,
      "epoch": 4.269875376020628,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003621623991194549,
      "loss": 5.3577,
      "mean_token_accuracy": 0.18539137542247772,
      "num_tokens": 9039012.0,
      "step": 4970
    },
    {
      "entropy": 5.752484273910523,
      "epoch": 4.274172754619682,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003618557826955594,
      "loss": 5.3187,
      "mean_token_accuracy": 0.19525409638881683,
      "num_tokens": 9048639.0,
      "step": 4975
    },
    {
      "entropy": 5.6776556968688965,
      "epoch": 4.278470133218737,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036154897661335063,
      "loss": 5.2847,
      "mean_token_accuracy": 0.19414151310920716,
      "num_tokens": 9057453.0,
      "step": 4980
    },
    {
      "entropy": 5.638094425201416,
      "epoch": 4.2827675118177915,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003612419815427702,
      "loss": 5.3033,
      "mean_token_accuracy": 0.1912328004837036,
      "num_tokens": 9066761.0,
      "step": 4985
    },
    {
      "entropy": 5.707727861404419,
      "epoch": 4.287064890416846,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003609347981541726,
      "loss": 5.3747,
      "mean_token_accuracy": 0.1863158091902733,
      "num_tokens": 9075535.0,
      "step": 4990
    },
    {
      "entropy": 5.740206003189087,
      "epoch": 4.291362269015901,
      "grad_norm": 1.359375,
      "learning_rate": 0.00036062742711832376,
      "loss": 5.2632,
      "mean_token_accuracy": 0.19991542398929596,
      "num_tokens": 9084559.0,
      "step": 4995
    },
    {
      "entropy": 5.719019842147827,
      "epoch": 4.295659647614955,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003603198691063991,
      "loss": 5.2521,
      "mean_token_accuracy": 0.19463850259780885,
      "num_tokens": 9093069.0,
      "step": 5000
    },
    {
      "epoch": 4.295659647614955,
      "eval_entropy": 5.532111000907314,
      "eval_loss": 6.042418956756592,
      "eval_mean_token_accuracy": 0.1693078381185596,
      "eval_num_tokens": 9093069.0,
      "eval_runtime": 2.6414,
      "eval_samples_per_second": 1343.627,
      "eval_steps_per_second": 168.095,
      "step": 5000
    },
    {
      "entropy": 5.629690361022949,
      "epoch": 4.29995702621401,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003600121247899824,
      "loss": 5.2463,
      "mean_token_accuracy": 0.19137472808361053,
      "num_tokens": 9101914.0,
      "step": 5005
    },
    {
      "entropy": 5.6700661182403564,
      "epoch": 4.304254404813064,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035970419484106404,
      "loss": 5.3313,
      "mean_token_accuracy": 0.19520494788885118,
      "num_tokens": 9110967.0,
      "step": 5010
    },
    {
      "entropy": 5.7633336067199705,
      "epoch": 4.308551783412119,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003593960799320402,
      "loss": 5.4063,
      "mean_token_accuracy": 0.18893446326255797,
      "num_tokens": 9119774.0,
      "step": 5015
    },
    {
      "entropy": 5.764187955856324,
      "epoch": 4.312849162011173,
      "grad_norm": 1.375,
      "learning_rate": 0.0003590877807357107,
      "loss": 5.3258,
      "mean_token_accuracy": 0.18940153568983079,
      "num_tokens": 9127738.0,
      "step": 5020
    },
    {
      "entropy": 5.6581501960754395,
      "epoch": 4.317146540610228,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003587792979252776,
      "loss": 5.2932,
      "mean_token_accuracy": 0.19798150062561035,
      "num_tokens": 9137060.0,
      "step": 5025
    },
    {
      "entropy": 5.65656476020813,
      "epoch": 4.321443919209282,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003584706321743442,
      "loss": 5.2103,
      "mean_token_accuracy": 0.1972648024559021,
      "num_tokens": 9145169.0,
      "step": 5030
    },
    {
      "entropy": 5.621655893325806,
      "epoch": 4.325741297808337,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000358161784156913,
      "loss": 5.2728,
      "mean_token_accuracy": 0.19494565725326538,
      "num_tokens": 9154092.0,
      "step": 5035
    },
    {
      "entropy": 5.71158275604248,
      "epoch": 4.3300386764073915,
      "grad_norm": 1.25,
      "learning_rate": 0.00035785275454738456,
      "loss": 5.2949,
      "mean_token_accuracy": 0.189323128759861,
      "num_tokens": 9162824.0,
      "step": 5040
    },
    {
      "entropy": 5.681181001663208,
      "epoch": 4.334336055006446,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00035754354402055635,
      "loss": 5.2177,
      "mean_token_accuracy": 0.20362782925367356,
      "num_tokens": 9170977.0,
      "step": 5045
    },
    {
      "entropy": 5.678464555740357,
      "epoch": 4.338633433605501,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003572341532516202,
      "loss": 5.2799,
      "mean_token_accuracy": 0.19674263447523116,
      "num_tokens": 9179539.0,
      "step": 5050
    },
    {
      "entropy": 5.605776309967041,
      "epoch": 4.342930812204555,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003569245829161622,
      "loss": 5.36,
      "mean_token_accuracy": 0.19239965826272964,
      "num_tokens": 9188861.0,
      "step": 5055
    },
    {
      "entropy": 5.736452865600586,
      "epoch": 4.34722819080361,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035661483369016004,
      "loss": 5.2872,
      "mean_token_accuracy": 0.19282012730836867,
      "num_tokens": 9197724.0,
      "step": 5060
    },
    {
      "entropy": 5.656119060516358,
      "epoch": 4.351525569402664,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003563049062499822,
      "loss": 5.3081,
      "mean_token_accuracy": 0.190804560482502,
      "num_tokens": 9206375.0,
      "step": 5065
    },
    {
      "entropy": 5.6067548274993895,
      "epoch": 4.355822948001719,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003559948012723865,
      "loss": 5.258,
      "mean_token_accuracy": 0.2009153798222542,
      "num_tokens": 9214675.0,
      "step": 5070
    },
    {
      "entropy": 5.710597896575928,
      "epoch": 4.360120326600773,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003556845194345181,
      "loss": 5.2908,
      "mean_token_accuracy": 0.1964338406920433,
      "num_tokens": 9224128.0,
      "step": 5075
    },
    {
      "entropy": 5.653571081161499,
      "epoch": 4.364417705199828,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003553740614139086,
      "loss": 5.2106,
      "mean_token_accuracy": 0.2021344095468521,
      "num_tokens": 9232568.0,
      "step": 5080
    },
    {
      "entropy": 5.635593557357788,
      "epoch": 4.368715083798882,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003550634278884742,
      "loss": 5.309,
      "mean_token_accuracy": 0.19702151864767076,
      "num_tokens": 9241809.0,
      "step": 5085
    },
    {
      "entropy": 5.708782768249511,
      "epoch": 4.373012462397937,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035475261953651433,
      "loss": 5.3112,
      "mean_token_accuracy": 0.19213190525770188,
      "num_tokens": 9250845.0,
      "step": 5090
    },
    {
      "entropy": 5.642183685302735,
      "epoch": 4.3773098409969915,
      "grad_norm": 1.34375,
      "learning_rate": 0.00035444163703671026,
      "loss": 5.2741,
      "mean_token_accuracy": 0.19802940934896468,
      "num_tokens": 9259465.0,
      "step": 5095
    },
    {
      "entropy": 5.636918735504151,
      "epoch": 4.381607219596046,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035413048106812357,
      "loss": 5.2407,
      "mean_token_accuracy": 0.20345366895198821,
      "num_tokens": 9267853.0,
      "step": 5100
    },
    {
      "entropy": 5.7112630844116214,
      "epoch": 4.385904598195101,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035381915231019425,
      "loss": 5.4249,
      "mean_token_accuracy": 0.182015061378479,
      "num_tokens": 9276664.0,
      "step": 5105
    },
    {
      "entropy": 5.71500678062439,
      "epoch": 4.390201976794156,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003535076514427401,
      "loss": 5.2468,
      "mean_token_accuracy": 0.19312726557254792,
      "num_tokens": 9285482.0,
      "step": 5110
    },
    {
      "entropy": 5.7605279922485355,
      "epoch": 4.39449935539321,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035319597914595436,
      "loss": 5.3595,
      "mean_token_accuracy": 0.18737665116786956,
      "num_tokens": 9293936.0,
      "step": 5115
    },
    {
      "entropy": 5.7501624584197994,
      "epoch": 4.398796733992265,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003528841361004049,
      "loss": 5.3825,
      "mean_token_accuracy": 0.1810387596487999,
      "num_tokens": 9303998.0,
      "step": 5120
    },
    {
      "entropy": 5.671031045913696,
      "epoch": 4.40309411259132,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003525721229870323,
      "loss": 5.3048,
      "mean_token_accuracy": 0.19352999329566956,
      "num_tokens": 9313117.0,
      "step": 5125
    },
    {
      "entropy": 5.727086925506592,
      "epoch": 4.407391491190374,
      "grad_norm": 1.25,
      "learning_rate": 0.00035225994048714823,
      "loss": 5.31,
      "mean_token_accuracy": 0.19567365795373917,
      "num_tokens": 9321446.0,
      "step": 5130
    },
    {
      "entropy": 5.656890153884888,
      "epoch": 4.411688869789429,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003519475892824348,
      "loss": 5.3033,
      "mean_token_accuracy": 0.19630219787359238,
      "num_tokens": 9330752.0,
      "step": 5135
    },
    {
      "entropy": 5.673178958892822,
      "epoch": 4.415986248388483,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003516350700549419,
      "loss": 5.3361,
      "mean_token_accuracy": 0.19190378189086915,
      "num_tokens": 9339322.0,
      "step": 5140
    },
    {
      "entropy": 5.711392068862915,
      "epoch": 4.420283626987538,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035132238348708697,
      "loss": 5.3541,
      "mean_token_accuracy": 0.19540640711784363,
      "num_tokens": 9349024.0,
      "step": 5145
    },
    {
      "entropy": 5.827709197998047,
      "epoch": 4.424581005586592,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035100953026165224,
      "loss": 5.4639,
      "mean_token_accuracy": 0.18463700115680695,
      "num_tokens": 9358833.0,
      "step": 5150
    },
    {
      "entropy": 5.763640642166138,
      "epoch": 4.428878384185647,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003506965110617841,
      "loss": 5.301,
      "mean_token_accuracy": 0.1951870560646057,
      "num_tokens": 9368276.0,
      "step": 5155
    },
    {
      "entropy": 5.69295482635498,
      "epoch": 4.4331757627847015,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003503833265709915,
      "loss": 5.3687,
      "mean_token_accuracy": 0.1910207003355026,
      "num_tokens": 9378501.0,
      "step": 5160
    },
    {
      "entropy": 5.746016502380371,
      "epoch": 4.437473141383756,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035006997747314404,
      "loss": 5.3505,
      "mean_token_accuracy": 0.18749817013740538,
      "num_tokens": 9387789.0,
      "step": 5165
    },
    {
      "entropy": 5.745327663421631,
      "epoch": 4.441770519982811,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034975646445247106,
      "loss": 5.4011,
      "mean_token_accuracy": 0.1885218068957329,
      "num_tokens": 9397041.0,
      "step": 5170
    },
    {
      "entropy": 5.672356414794922,
      "epoch": 4.446067898581865,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003494427881935596,
      "loss": 5.3182,
      "mean_token_accuracy": 0.1993885487318039,
      "num_tokens": 9405393.0,
      "step": 5175
    },
    {
      "entropy": 5.638329553604126,
      "epoch": 4.45036527718092,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034912894938135325,
      "loss": 5.2796,
      "mean_token_accuracy": 0.19539430737495422,
      "num_tokens": 9415127.0,
      "step": 5180
    },
    {
      "entropy": 5.73799467086792,
      "epoch": 4.454662655779974,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003488149487011506,
      "loss": 5.4087,
      "mean_token_accuracy": 0.1871132954955101,
      "num_tokens": 9424416.0,
      "step": 5185
    },
    {
      "entropy": 5.790239953994751,
      "epoch": 4.458960034379029,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034850078683860346,
      "loss": 5.3661,
      "mean_token_accuracy": 0.18556878715753555,
      "num_tokens": 9434523.0,
      "step": 5190
    },
    {
      "entropy": 5.677935028076172,
      "epoch": 4.463257412978083,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003481864644797159,
      "loss": 5.3572,
      "mean_token_accuracy": 0.19903489351272582,
      "num_tokens": 9443605.0,
      "step": 5195
    },
    {
      "entropy": 5.666917705535889,
      "epoch": 4.467554791577138,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003478719823108424,
      "loss": 5.3579,
      "mean_token_accuracy": 0.18392576277256012,
      "num_tokens": 9453268.0,
      "step": 5200
    },
    {
      "entropy": 5.678340339660645,
      "epoch": 4.471852170176192,
      "grad_norm": 1.25,
      "learning_rate": 0.00034755734101868613,
      "loss": 5.2247,
      "mean_token_accuracy": 0.20239908695220948,
      "num_tokens": 9461578.0,
      "step": 5205
    },
    {
      "entropy": 5.618629360198975,
      "epoch": 4.476149548775247,
      "grad_norm": 1.25,
      "learning_rate": 0.00034724254129029795,
      "loss": 5.2637,
      "mean_token_accuracy": 0.19489650130271913,
      "num_tokens": 9470722.0,
      "step": 5210
    },
    {
      "entropy": 5.706383275985718,
      "epoch": 4.4804469273743015,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003469275838130748,
      "loss": 5.3685,
      "mean_token_accuracy": 0.1928001955151558,
      "num_tokens": 9479695.0,
      "step": 5215
    },
    {
      "entropy": 5.737493562698364,
      "epoch": 4.484744305973356,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003466124692747577,
      "loss": 5.2998,
      "mean_token_accuracy": 0.19324341863393785,
      "num_tokens": 9488444.0,
      "step": 5220
    },
    {
      "entropy": 5.632571029663086,
      "epoch": 4.489041684572411,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034629719836343106,
      "loss": 5.242,
      "mean_token_accuracy": 0.19845967143774032,
      "num_tokens": 9497413.0,
      "step": 5225
    },
    {
      "entropy": 5.66089506149292,
      "epoch": 4.493339063171465,
      "grad_norm": 1.375,
      "learning_rate": 0.0003459817717675203,
      "loss": 5.2919,
      "mean_token_accuracy": 0.2009187415242195,
      "num_tokens": 9506135.0,
      "step": 5230
    },
    {
      "entropy": 5.670729255676269,
      "epoch": 4.49763644177052,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003456661901757913,
      "loss": 5.3686,
      "mean_token_accuracy": 0.18641350269317628,
      "num_tokens": 9516918.0,
      "step": 5235
    },
    {
      "entropy": 5.776004791259766,
      "epoch": 4.501933820369574,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034535045427734796,
      "loss": 5.3167,
      "mean_token_accuracy": 0.1980054035782814,
      "num_tokens": 9526052.0,
      "step": 5240
    },
    {
      "entropy": 5.66345739364624,
      "epoch": 4.506231198968629,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003450345647616313,
      "loss": 5.3816,
      "mean_token_accuracy": 0.19250868111848832,
      "num_tokens": 9535200.0,
      "step": 5245
    },
    {
      "entropy": 5.668608617782593,
      "epoch": 4.510528577567683,
      "grad_norm": 1.25,
      "learning_rate": 0.0003447185223184177,
      "loss": 5.3293,
      "mean_token_accuracy": 0.19699818789958953,
      "num_tokens": 9544786.0,
      "step": 5250
    },
    {
      "entropy": 5.676352453231812,
      "epoch": 4.514825956166739,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034440232763781765,
      "loss": 5.28,
      "mean_token_accuracy": 0.19427439868450164,
      "num_tokens": 9553694.0,
      "step": 5255
    },
    {
      "entropy": 5.633441591262818,
      "epoch": 4.519123334765792,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000344085981410274,
      "loss": 5.334,
      "mean_token_accuracy": 0.19646416902542113,
      "num_tokens": 9563332.0,
      "step": 5260
    },
    {
      "entropy": 5.595858097076416,
      "epoch": 4.523420713364848,
      "grad_norm": 1.125,
      "learning_rate": 0.00034376948432656036,
      "loss": 5.2664,
      "mean_token_accuracy": 0.19862610548734666,
      "num_tokens": 9572367.0,
      "step": 5265
    },
    {
      "entropy": 5.765823745727539,
      "epoch": 4.527718091963902,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003434528370777798,
      "loss": 5.3465,
      "mean_token_accuracy": 0.1927720606327057,
      "num_tokens": 9582535.0,
      "step": 5270
    },
    {
      "entropy": 5.707382583618164,
      "epoch": 4.532015470562957,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034313604035536344,
      "loss": 5.3156,
      "mean_token_accuracy": 0.20104162245988846,
      "num_tokens": 9590688.0,
      "step": 5275
    },
    {
      "entropy": 5.619611024856567,
      "epoch": 4.5363128491620115,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003428190948510687,
      "loss": 5.3373,
      "mean_token_accuracy": 0.19468231946229936,
      "num_tokens": 9599209.0,
      "step": 5280
    },
    {
      "entropy": 5.691408920288086,
      "epoch": 4.540610227761066,
      "grad_norm": 1.25,
      "learning_rate": 0.0003425020012569778,
      "loss": 5.3576,
      "mean_token_accuracy": 0.19136266857385636,
      "num_tokens": 9608575.0,
      "step": 5285
    },
    {
      "entropy": 5.769779300689697,
      "epoch": 4.544907606360121,
      "grad_norm": 1.25,
      "learning_rate": 0.00034218476026549665,
      "loss": 5.3335,
      "mean_token_accuracy": 0.19288893789052963,
      "num_tokens": 9617312.0,
      "step": 5290
    },
    {
      "entropy": 5.652000284194946,
      "epoch": 4.549204984959175,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003418673725693524,
      "loss": 5.3063,
      "mean_token_accuracy": 0.20449619591236115,
      "num_tokens": 9626398.0,
      "step": 5295
    },
    {
      "entropy": 5.624038887023926,
      "epoch": 4.55350236355823,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003415498388615932,
      "loss": 5.2875,
      "mean_token_accuracy": 0.18730463981628417,
      "num_tokens": 9635470.0,
      "step": 5300
    },
    {
      "entropy": 5.683203220367432,
      "epoch": 4.557799742157284,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003412321598355857,
      "loss": 5.2563,
      "mean_token_accuracy": 0.19899836033582688,
      "num_tokens": 9644728.0,
      "step": 5305
    },
    {
      "entropy": 5.685236597061158,
      "epoch": 4.562097120756339,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003409143361850139,
      "loss": 5.3145,
      "mean_token_accuracy": 0.19440280348062516,
      "num_tokens": 9654129.0,
      "step": 5310
    },
    {
      "entropy": 5.6666731357574465,
      "epoch": 4.566394499355393,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003405963686038775,
      "loss": 5.3716,
      "mean_token_accuracy": 0.19374995827674865,
      "num_tokens": 9662648.0,
      "step": 5315
    },
    {
      "entropy": 5.708667945861817,
      "epoch": 4.570691877954448,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003402782577864908,
      "loss": 5.3585,
      "mean_token_accuracy": 0.19610581994056703,
      "num_tokens": 9672082.0,
      "step": 5320
    },
    {
      "entropy": 5.7350396633148195,
      "epoch": 4.574989256553502,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033996000442748056,
      "loss": 5.277,
      "mean_token_accuracy": 0.20057252049446106,
      "num_tokens": 9681422.0,
      "step": 5325
    },
    {
      "entropy": 5.682338762283325,
      "epoch": 4.579286635152557,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033964160922178495,
      "loss": 5.315,
      "mean_token_accuracy": 0.19318680465221405,
      "num_tokens": 9690675.0,
      "step": 5330
    },
    {
      "entropy": 5.673515510559082,
      "epoch": 4.5835840137516115,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003393230728646518,
      "loss": 5.2972,
      "mean_token_accuracy": 0.1987778827548027,
      "num_tokens": 9700200.0,
      "step": 5335
    },
    {
      "entropy": 5.625961208343506,
      "epoch": 4.587881392350666,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033900439605163724,
      "loss": 5.3084,
      "mean_token_accuracy": 0.1911470979452133,
      "num_tokens": 9709533.0,
      "step": 5340
    },
    {
      "entropy": 5.648856353759766,
      "epoch": 4.592178770949721,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033868557947860407,
      "loss": 5.3347,
      "mean_token_accuracy": 0.19477206617593765,
      "num_tokens": 9719250.0,
      "step": 5345
    },
    {
      "entropy": 5.701521396636963,
      "epoch": 4.596476149548775,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033836662384172014,
      "loss": 5.2558,
      "mean_token_accuracy": 0.19975402802228928,
      "num_tokens": 9727837.0,
      "step": 5350
    },
    {
      "entropy": 5.682444763183594,
      "epoch": 4.60077352814783,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003380475298374573,
      "loss": 5.3526,
      "mean_token_accuracy": 0.1950669288635254,
      "num_tokens": 9737125.0,
      "step": 5355
    },
    {
      "entropy": 5.70787353515625,
      "epoch": 4.605070906746884,
      "grad_norm": 1.203125,
      "learning_rate": 0.000337728298162589,
      "loss": 5.3584,
      "mean_token_accuracy": 0.19272541999816895,
      "num_tokens": 9746309.0,
      "step": 5360
    },
    {
      "entropy": 5.69848222732544,
      "epoch": 4.609368285345939,
      "grad_norm": 1.375,
      "learning_rate": 0.00033740892951418993,
      "loss": 5.2625,
      "mean_token_accuracy": 0.2030378833413124,
      "num_tokens": 9755633.0,
      "step": 5365
    },
    {
      "entropy": 5.690783548355102,
      "epoch": 4.613665663944993,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003370894245896333,
      "loss": 5.2966,
      "mean_token_accuracy": 0.18761005401611328,
      "num_tokens": 9765179.0,
      "step": 5370
    },
    {
      "entropy": 5.742674732208252,
      "epoch": 4.617963042544048,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033676978408659047,
      "loss": 5.2908,
      "mean_token_accuracy": 0.19366790801286698,
      "num_tokens": 9774085.0,
      "step": 5375
    },
    {
      "entropy": 5.689094543457031,
      "epoch": 4.622260421143102,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003364500087030283,
      "loss": 5.4185,
      "mean_token_accuracy": 0.18378324955701827,
      "num_tokens": 9784650.0,
      "step": 5380
    },
    {
      "entropy": 5.704228019714355,
      "epoch": 4.626557799742157,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033613009913720845,
      "loss": 5.2785,
      "mean_token_accuracy": 0.19022756963968276,
      "num_tokens": 9793947.0,
      "step": 5385
    },
    {
      "entropy": 5.6313502311706545,
      "epoch": 4.6308551783412115,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033581005608768563,
      "loss": 5.2805,
      "mean_token_accuracy": 0.19847767651081086,
      "num_tokens": 9803593.0,
      "step": 5390
    },
    {
      "entropy": 5.654166793823242,
      "epoch": 4.635152556940266,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003354898802533058,
      "loss": 5.3129,
      "mean_token_accuracy": 0.19294002801179885,
      "num_tokens": 9812295.0,
      "step": 5395
    },
    {
      "entropy": 5.680404758453369,
      "epoch": 4.6394499355393215,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003351695723332051,
      "loss": 5.3136,
      "mean_token_accuracy": 0.19971238523721696,
      "num_tokens": 9820586.0,
      "step": 5400
    },
    {
      "entropy": 5.644966077804566,
      "epoch": 4.643747314138375,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033484913302680807,
      "loss": 5.2433,
      "mean_token_accuracy": 0.19560457319021224,
      "num_tokens": 9829080.0,
      "step": 5405
    },
    {
      "entropy": 5.65889983177185,
      "epoch": 4.648044692737431,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033452856303382595,
      "loss": 5.2668,
      "mean_token_accuracy": 0.19484354853630065,
      "num_tokens": 9838421.0,
      "step": 5410
    },
    {
      "entropy": 5.667877340316773,
      "epoch": 4.652342071336484,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003342078630542555,
      "loss": 5.2651,
      "mean_token_accuracy": 0.2015419289469719,
      "num_tokens": 9847151.0,
      "step": 5415
    },
    {
      "entropy": 5.657855892181397,
      "epoch": 4.65663944993554,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033388703378837737,
      "loss": 5.3137,
      "mean_token_accuracy": 0.1955198287963867,
      "num_tokens": 9856803.0,
      "step": 5420
    },
    {
      "entropy": 5.682938909530639,
      "epoch": 4.660936828534594,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003335660759367544,
      "loss": 5.2166,
      "mean_token_accuracy": 0.2060480907559395,
      "num_tokens": 9865617.0,
      "step": 5425
    },
    {
      "entropy": 5.670103454589844,
      "epoch": 4.665234207133649,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033324499020023025,
      "loss": 5.2667,
      "mean_token_accuracy": 0.20502229779958725,
      "num_tokens": 9875454.0,
      "step": 5430
    },
    {
      "entropy": 5.636929273605347,
      "epoch": 4.669531585732703,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003329237772799277,
      "loss": 5.265,
      "mean_token_accuracy": 0.19763256311416627,
      "num_tokens": 9884770.0,
      "step": 5435
    },
    {
      "entropy": 5.658205842971801,
      "epoch": 4.673828964331758,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003326024378772477,
      "loss": 5.2854,
      "mean_token_accuracy": 0.19928305000066757,
      "num_tokens": 9893594.0,
      "step": 5440
    },
    {
      "entropy": 5.7388335227966305,
      "epoch": 4.678126342930812,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003322809726938667,
      "loss": 5.3862,
      "mean_token_accuracy": 0.18453315496444703,
      "num_tokens": 9902260.0,
      "step": 5445
    },
    {
      "entropy": 5.664768075942993,
      "epoch": 4.682423721529867,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033195938243173645,
      "loss": 5.2968,
      "mean_token_accuracy": 0.19606153070926666,
      "num_tokens": 9911020.0,
      "step": 5450
    },
    {
      "entropy": 5.693754386901856,
      "epoch": 4.6867211001289215,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003316376677930814,
      "loss": 5.3161,
      "mean_token_accuracy": 0.19240303486585617,
      "num_tokens": 9918696.0,
      "step": 5455
    },
    {
      "entropy": 5.63634877204895,
      "epoch": 4.691018478727976,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003313158294803977,
      "loss": 5.3379,
      "mean_token_accuracy": 0.1874315157532692,
      "num_tokens": 9927638.0,
      "step": 5460
    },
    {
      "entropy": 5.682800245285034,
      "epoch": 4.695315857327031,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033099386819645176,
      "loss": 5.3087,
      "mean_token_accuracy": 0.19127850085496903,
      "num_tokens": 9936969.0,
      "step": 5465
    },
    {
      "entropy": 5.690246868133545,
      "epoch": 4.699613235926085,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003306717846442782,
      "loss": 5.2284,
      "mean_token_accuracy": 0.19865972101688384,
      "num_tokens": 9945229.0,
      "step": 5470
    },
    {
      "entropy": 5.643537521362305,
      "epoch": 4.70391061452514,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003303495795271788,
      "loss": 5.2189,
      "mean_token_accuracy": 0.19834866374731064,
      "num_tokens": 9953759.0,
      "step": 5475
    },
    {
      "entropy": 5.6468414783477785,
      "epoch": 4.708207993124194,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033002725354872075,
      "loss": 5.3217,
      "mean_token_accuracy": 0.19023809731006622,
      "num_tokens": 9962771.0,
      "step": 5480
    },
    {
      "entropy": 5.664292621612549,
      "epoch": 4.712505371723249,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032970480741273514,
      "loss": 5.3401,
      "mean_token_accuracy": 0.18536012172698973,
      "num_tokens": 9972481.0,
      "step": 5485
    },
    {
      "entropy": 5.740135145187378,
      "epoch": 4.716802750322303,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003293822418233155,
      "loss": 5.2809,
      "mean_token_accuracy": 0.19446541666984557,
      "num_tokens": 9980773.0,
      "step": 5490
    },
    {
      "entropy": 5.707640600204468,
      "epoch": 4.721100128921358,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003290595574848161,
      "loss": 5.3688,
      "mean_token_accuracy": 0.18625104576349258,
      "num_tokens": 9989830.0,
      "step": 5495
    },
    {
      "entropy": 5.626060867309571,
      "epoch": 4.725397507520412,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003287367551018505,
      "loss": 5.3014,
      "mean_token_accuracy": 0.19618325531482697,
      "num_tokens": 9999234.0,
      "step": 5500
    },
    {
      "epoch": 4.725397507520412,
      "eval_entropy": 5.536909558751562,
      "eval_loss": 6.004858493804932,
      "eval_mean_token_accuracy": 0.172105307111869,
      "eval_num_tokens": 9999234.0,
      "eval_runtime": 2.8193,
      "eval_samples_per_second": 1258.836,
      "eval_steps_per_second": 157.488,
      "step": 5500
    },
    {
      "entropy": 5.655096483230591,
      "epoch": 4.729694886119467,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003284138353792903,
      "loss": 5.3493,
      "mean_token_accuracy": 0.19083576053380966,
      "num_tokens": 10008671.0,
      "step": 5505
    },
    {
      "entropy": 5.643358278274536,
      "epoch": 4.7339922647185215,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003280907990222628,
      "loss": 5.3055,
      "mean_token_accuracy": 0.19403896629810333,
      "num_tokens": 10017170.0,
      "step": 5510
    },
    {
      "entropy": 5.668510293960571,
      "epoch": 4.738289643317576,
      "grad_norm": 1.421875,
      "learning_rate": 0.00032776764673615055,
      "loss": 5.3433,
      "mean_token_accuracy": 0.1903414621949196,
      "num_tokens": 10025712.0,
      "step": 5515
    },
    {
      "entropy": 5.678393459320068,
      "epoch": 4.742587021916631,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003274443792265888,
      "loss": 5.3274,
      "mean_token_accuracy": 0.20014615207910538,
      "num_tokens": 10035297.0,
      "step": 5520
    },
    {
      "entropy": 5.683428382873535,
      "epoch": 4.746884400515685,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032712099719946474,
      "loss": 5.3227,
      "mean_token_accuracy": 0.19473165273666382,
      "num_tokens": 10043903.0,
      "step": 5525
    },
    {
      "entropy": 5.66631851196289,
      "epoch": 4.75118177911474,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032679750136091533,
      "loss": 5.34,
      "mean_token_accuracy": 0.18970831036567687,
      "num_tokens": 10053035.0,
      "step": 5530
    },
    {
      "entropy": 5.6000752449035645,
      "epoch": 4.755479157713794,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003264738924173262,
      "loss": 5.2806,
      "mean_token_accuracy": 0.19872922599315643,
      "num_tokens": 10061911.0,
      "step": 5535
    },
    {
      "entropy": 5.660582160949707,
      "epoch": 4.759776536312849,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032615017107533,
      "loss": 5.2926,
      "mean_token_accuracy": 0.19948810786008836,
      "num_tokens": 10070738.0,
      "step": 5540
    },
    {
      "entropy": 5.656670951843262,
      "epoch": 4.764073914911903,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003258263380418047,
      "loss": 5.3201,
      "mean_token_accuracy": 0.19468791484832765,
      "num_tokens": 10080638.0,
      "step": 5545
    },
    {
      "entropy": 5.746007871627808,
      "epoch": 4.768371293510958,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032550239402387226,
      "loss": 5.3474,
      "mean_token_accuracy": 0.18813855797052384,
      "num_tokens": 10089429.0,
      "step": 5550
    },
    {
      "entropy": 5.6471367359161375,
      "epoch": 4.772668672110013,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032517833972889695,
      "loss": 5.2309,
      "mean_token_accuracy": 0.20105767697095872,
      "num_tokens": 10098109.0,
      "step": 5555
    },
    {
      "entropy": 5.724657249450684,
      "epoch": 4.776966050709067,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032485417586448375,
      "loss": 5.3279,
      "mean_token_accuracy": 0.1960192173719406,
      "num_tokens": 10106808.0,
      "step": 5560
    },
    {
      "entropy": 5.6491917133331295,
      "epoch": 4.781263429308122,
      "grad_norm": 1.359375,
      "learning_rate": 0.000324529903138477,
      "loss": 5.3308,
      "mean_token_accuracy": 0.18897253274917603,
      "num_tokens": 10116372.0,
      "step": 5565
    },
    {
      "entropy": 5.6238702774047855,
      "epoch": 4.785560807907177,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003242055222589587,
      "loss": 5.2484,
      "mean_token_accuracy": 0.20149259716272355,
      "num_tokens": 10125256.0,
      "step": 5570
    },
    {
      "entropy": 5.701531219482422,
      "epoch": 4.7898581865062315,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000323881033934247,
      "loss": 5.3695,
      "mean_token_accuracy": 0.18730445057153702,
      "num_tokens": 10134784.0,
      "step": 5575
    },
    {
      "entropy": 5.772956705093383,
      "epoch": 4.794155565105286,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032355643887289486,
      "loss": 5.317,
      "mean_token_accuracy": 0.19378293454647064,
      "num_tokens": 10144324.0,
      "step": 5580
    },
    {
      "entropy": 5.693816709518432,
      "epoch": 4.798452943704341,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003232317377836881,
      "loss": 5.3529,
      "mean_token_accuracy": 0.19022511094808578,
      "num_tokens": 10152866.0,
      "step": 5585
    },
    {
      "entropy": 5.667680740356445,
      "epoch": 4.802750322303395,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000322906931375644,
      "loss": 5.2661,
      "mean_token_accuracy": 0.200899201631546,
      "num_tokens": 10162457.0,
      "step": 5590
    },
    {
      "entropy": 5.6878539562225345,
      "epoch": 4.80704770090245,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032258202035801,
      "loss": 5.3411,
      "mean_token_accuracy": 0.1909174606204033,
      "num_tokens": 10171604.0,
      "step": 5595
    },
    {
      "entropy": 5.704989957809448,
      "epoch": 4.811345079501504,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000322257005440262,
      "loss": 5.3048,
      "mean_token_accuracy": 0.1973528489470482,
      "num_tokens": 10180762.0,
      "step": 5600
    },
    {
      "entropy": 5.565345191955567,
      "epoch": 4.815642458100559,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003219318873321025,
      "loss": 5.224,
      "mean_token_accuracy": 0.2138982519507408,
      "num_tokens": 10189122.0,
      "step": 5605
    },
    {
      "entropy": 5.670344829559326,
      "epoch": 4.819939836699613,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032160666674345954,
      "loss": 5.3303,
      "mean_token_accuracy": 0.18996544927358627,
      "num_tokens": 10197280.0,
      "step": 5610
    },
    {
      "entropy": 5.702713298797607,
      "epoch": 4.824237215298668,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032128134438448504,
      "loss": 5.366,
      "mean_token_accuracy": 0.188306000828743,
      "num_tokens": 10207507.0,
      "step": 5615
    },
    {
      "entropy": 5.705723094940185,
      "epoch": 4.828534593897722,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032095592096555284,
      "loss": 5.3414,
      "mean_token_accuracy": 0.1951340302824974,
      "num_tokens": 10217584.0,
      "step": 5620
    },
    {
      "entropy": 5.6782811164855955,
      "epoch": 4.832831972496777,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003206303971972577,
      "loss": 5.2859,
      "mean_token_accuracy": 0.19992830753326415,
      "num_tokens": 10226388.0,
      "step": 5625
    },
    {
      "entropy": 5.639026689529419,
      "epoch": 4.8371293510958315,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003203047737904134,
      "loss": 5.3168,
      "mean_token_accuracy": 0.1947314038872719,
      "num_tokens": 10235333.0,
      "step": 5630
    },
    {
      "entropy": 5.6958168029785154,
      "epoch": 4.841426729694886,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031997905145605135,
      "loss": 5.3463,
      "mean_token_accuracy": 0.19245222210884094,
      "num_tokens": 10243985.0,
      "step": 5635
    },
    {
      "entropy": 5.724058294296265,
      "epoch": 4.845724108293941,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031965323090541874,
      "loss": 5.3449,
      "mean_token_accuracy": 0.18725754618644713,
      "num_tokens": 10252968.0,
      "step": 5640
    },
    {
      "entropy": 5.669808340072632,
      "epoch": 4.850021486892995,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003193273128499777,
      "loss": 5.218,
      "mean_token_accuracy": 0.2013757660984993,
      "num_tokens": 10261890.0,
      "step": 5645
    },
    {
      "entropy": 5.65012378692627,
      "epoch": 4.85431886549205,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031900129800140287,
      "loss": 5.3252,
      "mean_token_accuracy": 0.19297685027122496,
      "num_tokens": 10271363.0,
      "step": 5650
    },
    {
      "entropy": 5.692674970626831,
      "epoch": 4.858616244091104,
      "grad_norm": 1.421875,
      "learning_rate": 0.00031867518707158027,
      "loss": 5.3592,
      "mean_token_accuracy": 0.1890120878815651,
      "num_tokens": 10280608.0,
      "step": 5655
    },
    {
      "entropy": 5.625173950195313,
      "epoch": 4.862913622690159,
      "grad_norm": 1.3125,
      "learning_rate": 0.000318348980772606,
      "loss": 5.294,
      "mean_token_accuracy": 0.19891018122434617,
      "num_tokens": 10289972.0,
      "step": 5660
    },
    {
      "entropy": 5.7193211078643795,
      "epoch": 4.867211001289213,
      "grad_norm": 1.34375,
      "learning_rate": 0.00031802267981678414,
      "loss": 5.3438,
      "mean_token_accuracy": 0.19350936710834504,
      "num_tokens": 10298740.0,
      "step": 5665
    },
    {
      "entropy": 5.718724060058594,
      "epoch": 4.871508379888268,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031769628491662563,
      "loss": 5.3073,
      "mean_token_accuracy": 0.19792592972517015,
      "num_tokens": 10307706.0,
      "step": 5670
    },
    {
      "entropy": 5.681690311431884,
      "epoch": 4.8758057584873224,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031736979678484634,
      "loss": 5.3291,
      "mean_token_accuracy": 0.203986956179142,
      "num_tokens": 10317549.0,
      "step": 5675
    },
    {
      "entropy": 5.6633750915527346,
      "epoch": 4.880103137086377,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031704321613436597,
      "loss": 5.382,
      "mean_token_accuracy": 0.18851124197244645,
      "num_tokens": 10327681.0,
      "step": 5680
    },
    {
      "entropy": 5.6151251792907715,
      "epoch": 4.8844005156854315,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003167165436783061,
      "loss": 5.316,
      "mean_token_accuracy": 0.1952875792980194,
      "num_tokens": 10336261.0,
      "step": 5685
    },
    {
      "entropy": 5.604975843429566,
      "epoch": 4.888697894284486,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031638978012998875,
      "loss": 5.23,
      "mean_token_accuracy": 0.207156503200531,
      "num_tokens": 10344770.0,
      "step": 5690
    },
    {
      "entropy": 5.653034973144531,
      "epoch": 4.892995272883541,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000316062926202935,
      "loss": 5.3735,
      "mean_token_accuracy": 0.1833770230412483,
      "num_tokens": 10354246.0,
      "step": 5695
    },
    {
      "entropy": 5.6863391399383545,
      "epoch": 4.897292651482596,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003157359826108632,
      "loss": 5.2943,
      "mean_token_accuracy": 0.19547336995601655,
      "num_tokens": 10362693.0,
      "step": 5700
    },
    {
      "entropy": 5.703089332580566,
      "epoch": 4.90159003008165,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031540895006768727,
      "loss": 5.3144,
      "mean_token_accuracy": 0.19449802339076996,
      "num_tokens": 10371639.0,
      "step": 5705
    },
    {
      "entropy": 5.689019203186035,
      "epoch": 4.905887408680705,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003150818292875158,
      "loss": 5.3158,
      "mean_token_accuracy": 0.19616936147212982,
      "num_tokens": 10381237.0,
      "step": 5710
    },
    {
      "entropy": 5.727988290786743,
      "epoch": 4.91018478727976,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003147546209846497,
      "loss": 5.2949,
      "mean_token_accuracy": 0.19279846847057341,
      "num_tokens": 10389932.0,
      "step": 5715
    },
    {
      "entropy": 5.58194055557251,
      "epoch": 4.914482165878814,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003144273258735812,
      "loss": 5.1907,
      "mean_token_accuracy": 0.19747881293296815,
      "num_tokens": 10398938.0,
      "step": 5720
    },
    {
      "entropy": 5.594201898574829,
      "epoch": 4.918779544477869,
      "grad_norm": 1.546875,
      "learning_rate": 0.0003140999446689919,
      "loss": 5.2976,
      "mean_token_accuracy": 0.19196115136146547,
      "num_tokens": 10407980.0,
      "step": 5725
    },
    {
      "entropy": 5.617248249053955,
      "epoch": 4.923076923076923,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003137724780857516,
      "loss": 5.3693,
      "mean_token_accuracy": 0.19130895286798477,
      "num_tokens": 10416990.0,
      "step": 5730
    },
    {
      "entropy": 5.730205249786377,
      "epoch": 4.927374301675978,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031344492683891634,
      "loss": 5.3783,
      "mean_token_accuracy": 0.1947888031601906,
      "num_tokens": 10426573.0,
      "step": 5735
    },
    {
      "entropy": 5.733776712417603,
      "epoch": 4.931671680275032,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003131172916437272,
      "loss": 5.3511,
      "mean_token_accuracy": 0.1913149729371071,
      "num_tokens": 10435162.0,
      "step": 5740
    },
    {
      "entropy": 5.701927185058594,
      "epoch": 4.935969058874087,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031278957321560845,
      "loss": 5.3312,
      "mean_token_accuracy": 0.19573891162872314,
      "num_tokens": 10444374.0,
      "step": 5745
    },
    {
      "entropy": 5.76324429512024,
      "epoch": 4.9402664374731415,
      "grad_norm": 1.421875,
      "learning_rate": 0.00031246177227016615,
      "loss": 5.3773,
      "mean_token_accuracy": 0.17885937839746474,
      "num_tokens": 10452679.0,
      "step": 5750
    },
    {
      "entropy": 5.63156042098999,
      "epoch": 4.944563816072196,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031213388952318653,
      "loss": 5.2934,
      "mean_token_accuracy": 0.19874859005212783,
      "num_tokens": 10461801.0,
      "step": 5755
    },
    {
      "entropy": 5.674669218063355,
      "epoch": 4.948861194671251,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003118059256906345,
      "loss": 5.2974,
      "mean_token_accuracy": 0.19443876594305037,
      "num_tokens": 10471176.0,
      "step": 5760
    },
    {
      "entropy": 5.684893035888672,
      "epoch": 4.953158573270305,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031147788148865204,
      "loss": 5.3822,
      "mean_token_accuracy": 0.18452401906251908,
      "num_tokens": 10480403.0,
      "step": 5765
    },
    {
      "entropy": 5.669341945648194,
      "epoch": 4.95745595186936,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003111497576335564,
      "loss": 5.3012,
      "mean_token_accuracy": 0.19741881638765335,
      "num_tokens": 10489574.0,
      "step": 5770
    },
    {
      "entropy": 5.667093515396118,
      "epoch": 4.961753330468414,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003108215548418391,
      "loss": 5.3118,
      "mean_token_accuracy": 0.19258766919374465,
      "num_tokens": 10499631.0,
      "step": 5775
    },
    {
      "entropy": 5.690812587738037,
      "epoch": 4.966050709067469,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0003104932738301637,
      "loss": 5.2898,
      "mean_token_accuracy": 0.19252416491508484,
      "num_tokens": 10508128.0,
      "step": 5780
    },
    {
      "entropy": 5.671875381469727,
      "epoch": 4.970348087666523,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031016491531536477,
      "loss": 5.2701,
      "mean_token_accuracy": 0.19783350378274916,
      "num_tokens": 10517544.0,
      "step": 5785
    },
    {
      "entropy": 5.607844114303589,
      "epoch": 4.974645466265578,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003098364800144462,
      "loss": 5.3297,
      "mean_token_accuracy": 0.19353507310152054,
      "num_tokens": 10526244.0,
      "step": 5790
    },
    {
      "entropy": 5.707543277740479,
      "epoch": 4.9789428448646325,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003095079686445792,
      "loss": 5.4062,
      "mean_token_accuracy": 0.18872842341661453,
      "num_tokens": 10535887.0,
      "step": 5795
    },
    {
      "entropy": 5.685035848617554,
      "epoch": 4.983240223463687,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030917938192310146,
      "loss": 5.2509,
      "mean_token_accuracy": 0.19753166735172273,
      "num_tokens": 10544420.0,
      "step": 5800
    },
    {
      "entropy": 5.708291721343994,
      "epoch": 4.9875376020627415,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030885072056751494,
      "loss": 5.3407,
      "mean_token_accuracy": 0.1906682774424553,
      "num_tokens": 10553114.0,
      "step": 5805
    },
    {
      "entropy": 5.620190620422363,
      "epoch": 4.991834980661796,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030852198529548476,
      "loss": 5.3714,
      "mean_token_accuracy": 0.1967923805117607,
      "num_tokens": 10562272.0,
      "step": 5810
    },
    {
      "entropy": 5.662737321853638,
      "epoch": 4.996132359260851,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003081931768248373,
      "loss": 5.2923,
      "mean_token_accuracy": 0.19946238845586778,
      "num_tokens": 10571757.0,
      "step": 5815
    },
    {
      "entropy": 5.642363495296902,
      "epoch": 5.0,
      "grad_norm": 1.921875,
      "learning_rate": 0.0003078642958735588,
      "loss": 5.2621,
      "mean_token_accuracy": 0.19936485588550568,
      "num_tokens": 10579660.0,
      "step": 5820
    },
    {
      "entropy": 5.710732412338257,
      "epoch": 5.0042973785990545,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030753534315979393,
      "loss": 5.1079,
      "mean_token_accuracy": 0.2072200894355774,
      "num_tokens": 10589139.0,
      "step": 5825
    },
    {
      "entropy": 5.639051294326782,
      "epoch": 5.008594757198109,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003072063194018438,
      "loss": 4.9454,
      "mean_token_accuracy": 0.21830174028873445,
      "num_tokens": 10597915.0,
      "step": 5830
    },
    {
      "entropy": 5.6694016456604,
      "epoch": 5.012892135797164,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003068772253181648,
      "loss": 5.128,
      "mean_token_accuracy": 0.20124182999134063,
      "num_tokens": 10606491.0,
      "step": 5835
    },
    {
      "entropy": 5.678703927993775,
      "epoch": 5.017189514396218,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003065480616273671,
      "loss": 5.1135,
      "mean_token_accuracy": 0.19897813349962234,
      "num_tokens": 10615852.0,
      "step": 5840
    },
    {
      "entropy": 5.785790872573853,
      "epoch": 5.021486892995273,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003062188290482123,
      "loss": 5.0883,
      "mean_token_accuracy": 0.20427374988794328,
      "num_tokens": 10625442.0,
      "step": 5845
    },
    {
      "entropy": 5.643905305862427,
      "epoch": 5.025784271594327,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030588952829961304,
      "loss": 5.0243,
      "mean_token_accuracy": 0.21368957608938216,
      "num_tokens": 10634972.0,
      "step": 5850
    },
    {
      "entropy": 5.6278886795043945,
      "epoch": 5.030081650193382,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003055601601006303,
      "loss": 5.0022,
      "mean_token_accuracy": 0.21471837013959885,
      "num_tokens": 10644487.0,
      "step": 5855
    },
    {
      "entropy": 5.665062952041626,
      "epoch": 5.034379028792436,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003052307251704728,
      "loss": 5.0248,
      "mean_token_accuracy": 0.2257661610841751,
      "num_tokens": 10654144.0,
      "step": 5860
    },
    {
      "entropy": 5.648764514923096,
      "epoch": 5.038676407391491,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003049012242284946,
      "loss": 5.0945,
      "mean_token_accuracy": 0.20216402262449265,
      "num_tokens": 10663023.0,
      "step": 5865
    },
    {
      "entropy": 5.6526196002960205,
      "epoch": 5.0429737859905455,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003045716579941941,
      "loss": 5.1521,
      "mean_token_accuracy": 0.20288993865251542,
      "num_tokens": 10672001.0,
      "step": 5870
    },
    {
      "entropy": 5.617119789123535,
      "epoch": 5.0472711645896,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030424202718721215,
      "loss": 5.0979,
      "mean_token_accuracy": 0.20371967405080796,
      "num_tokens": 10682654.0,
      "step": 5875
    },
    {
      "entropy": 5.689823913574219,
      "epoch": 5.051568543188655,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030391233252733085,
      "loss": 5.1023,
      "mean_token_accuracy": 0.2105075567960739,
      "num_tokens": 10691429.0,
      "step": 5880
    },
    {
      "entropy": 5.642111682891846,
      "epoch": 5.055865921787709,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030358257473447144,
      "loss": 5.0462,
      "mean_token_accuracy": 0.2136613130569458,
      "num_tokens": 10701130.0,
      "step": 5885
    },
    {
      "entropy": 5.595071744918823,
      "epoch": 5.060163300386764,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030325275452869316,
      "loss": 5.03,
      "mean_token_accuracy": 0.2161984533071518,
      "num_tokens": 10709779.0,
      "step": 5890
    },
    {
      "entropy": 5.577562379837036,
      "epoch": 5.064460678985819,
      "grad_norm": 1.328125,
      "learning_rate": 0.00030292287263019153,
      "loss": 5.0806,
      "mean_token_accuracy": 0.21391009986400605,
      "num_tokens": 10718795.0,
      "step": 5895
    },
    {
      "entropy": 5.491227769851685,
      "epoch": 5.068758057584874,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030259292975929675,
      "loss": 4.9789,
      "mean_token_accuracy": 0.2186121702194214,
      "num_tokens": 10728202.0,
      "step": 5900
    },
    {
      "entropy": 5.643309593200684,
      "epoch": 5.073055436183928,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003022629266364723,
      "loss": 5.032,
      "mean_token_accuracy": 0.20635049790143967,
      "num_tokens": 10737050.0,
      "step": 5905
    },
    {
      "entropy": 5.660158157348633,
      "epoch": 5.077352814782983,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00030193286398231276,
      "loss": 5.0543,
      "mean_token_accuracy": 0.21502950489521028,
      "num_tokens": 10745261.0,
      "step": 5910
    },
    {
      "entropy": 5.63793535232544,
      "epoch": 5.081650193382037,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030160274251754337,
      "loss": 5.1036,
      "mean_token_accuracy": 0.2068781077861786,
      "num_tokens": 10755008.0,
      "step": 5915
    },
    {
      "entropy": 5.608860921859741,
      "epoch": 5.085947571981092,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030127256296301724,
      "loss": 5.12,
      "mean_token_accuracy": 0.20165755599737167,
      "num_tokens": 10763951.0,
      "step": 5920
    },
    {
      "entropy": 5.635695457458496,
      "epoch": 5.090244950580146,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003009423260397148,
      "loss": 5.0187,
      "mean_token_accuracy": 0.21295688897371293,
      "num_tokens": 10772770.0,
      "step": 5925
    },
    {
      "entropy": 5.691758060455323,
      "epoch": 5.094542329179201,
      "grad_norm": 1.25,
      "learning_rate": 0.00030061203246874125,
      "loss": 5.1346,
      "mean_token_accuracy": 0.1994927316904068,
      "num_tokens": 10781827.0,
      "step": 5930
    },
    {
      "entropy": 5.695685720443725,
      "epoch": 5.0988397077782555,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030028168297132593,
      "loss": 5.1807,
      "mean_token_accuracy": 0.20654743313789367,
      "num_tokens": 10792321.0,
      "step": 5935
    },
    {
      "entropy": 5.614190006256104,
      "epoch": 5.10313708637731,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002999512782688199,
      "loss": 5.143,
      "mean_token_accuracy": 0.21135859340429305,
      "num_tokens": 10801689.0,
      "step": 5940
    },
    {
      "entropy": 5.693757820129394,
      "epoch": 5.1074344649763646,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002996208190826951,
      "loss": 5.0707,
      "mean_token_accuracy": 0.20253875404596328,
      "num_tokens": 10810513.0,
      "step": 5945
    },
    {
      "entropy": 5.634591341018677,
      "epoch": 5.111731843575419,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029929030613454227,
      "loss": 5.048,
      "mean_token_accuracy": 0.2108563408255577,
      "num_tokens": 10819581.0,
      "step": 5950
    },
    {
      "entropy": 5.562780046463013,
      "epoch": 5.116029222174474,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002989597401460697,
      "loss": 5.0209,
      "mean_token_accuracy": 0.21204763799905776,
      "num_tokens": 10828139.0,
      "step": 5955
    },
    {
      "entropy": 5.624406099319458,
      "epoch": 5.120326600773528,
      "grad_norm": 1.421875,
      "learning_rate": 0.00029862912183910105,
      "loss": 5.0411,
      "mean_token_accuracy": 0.2158510982990265,
      "num_tokens": 10836256.0,
      "step": 5960
    },
    {
      "entropy": 5.674741792678833,
      "epoch": 5.124623979372583,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029829845193557496,
      "loss": 5.0766,
      "mean_token_accuracy": 0.21190978735685348,
      "num_tokens": 10846255.0,
      "step": 5965
    },
    {
      "entropy": 5.598087358474731,
      "epoch": 5.128921357971637,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002979677311575421,
      "loss": 5.0698,
      "mean_token_accuracy": 0.21416724920272828,
      "num_tokens": 10855546.0,
      "step": 5970
    },
    {
      "entropy": 5.650513219833374,
      "epoch": 5.133218736570692,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002976369602271646,
      "loss": 5.1456,
      "mean_token_accuracy": 0.20612256675958635,
      "num_tokens": 10864417.0,
      "step": 5975
    },
    {
      "entropy": 5.6583473682403564,
      "epoch": 5.137516115169746,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002973061398667138,
      "loss": 5.115,
      "mean_token_accuracy": 0.20908812284469605,
      "num_tokens": 10874527.0,
      "step": 5980
    },
    {
      "entropy": 5.622729682922364,
      "epoch": 5.141813493768801,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029697527079856916,
      "loss": 5.1189,
      "mean_token_accuracy": 0.20565245449543,
      "num_tokens": 10883486.0,
      "step": 5985
    },
    {
      "entropy": 5.600504064559937,
      "epoch": 5.1461108723678555,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029664435374521665,
      "loss": 5.0135,
      "mean_token_accuracy": 0.21093275398015976,
      "num_tokens": 10891972.0,
      "step": 5990
    },
    {
      "entropy": 5.654717493057251,
      "epoch": 5.15040825096691,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029631338942924664,
      "loss": 5.0279,
      "mean_token_accuracy": 0.2153494417667389,
      "num_tokens": 10901350.0,
      "step": 5995
    },
    {
      "entropy": 5.589114761352539,
      "epoch": 5.154705629565965,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002959823785733531,
      "loss": 5.0504,
      "mean_token_accuracy": 0.2096717193722725,
      "num_tokens": 10910114.0,
      "step": 6000
    },
    {
      "epoch": 5.154705629565965,
      "eval_entropy": 5.469111401755531,
      "eval_loss": 5.996286392211914,
      "eval_mean_token_accuracy": 0.17323775945214537,
      "eval_num_tokens": 10910114.0,
      "eval_runtime": 2.6288,
      "eval_samples_per_second": 1350.052,
      "eval_steps_per_second": 168.899,
      "step": 6000
    },
    {
      "entropy": 5.605376148223877,
      "epoch": 5.159003008165019,
      "grad_norm": 1.375,
      "learning_rate": 0.0002956513219003312,
      "loss": 5.1129,
      "mean_token_accuracy": 0.21038171350955964,
      "num_tokens": 10919781.0,
      "step": 6005
    },
    {
      "entropy": 5.624401426315307,
      "epoch": 5.163300386764074,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029532022013307666,
      "loss": 5.1214,
      "mean_token_accuracy": 0.20623801350593568,
      "num_tokens": 10929561.0,
      "step": 6010
    },
    {
      "entropy": 5.664988040924072,
      "epoch": 5.167597765363128,
      "grad_norm": 1.25,
      "learning_rate": 0.00029498907399458325,
      "loss": 5.0589,
      "mean_token_accuracy": 0.20685831010341643,
      "num_tokens": 10939123.0,
      "step": 6015
    },
    {
      "entropy": 5.645172739028931,
      "epoch": 5.171895143962183,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002946578842079418,
      "loss": 5.1203,
      "mean_token_accuracy": 0.2070673644542694,
      "num_tokens": 10947990.0,
      "step": 6020
    },
    {
      "entropy": 5.6086883544921875,
      "epoch": 5.176192522561237,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002943266514963384,
      "loss": 5.0293,
      "mean_token_accuracy": 0.22365083992481233,
      "num_tokens": 10956569.0,
      "step": 6025
    },
    {
      "entropy": 5.657001495361328,
      "epoch": 5.180489901160292,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002939953765830529,
      "loss": 5.0799,
      "mean_token_accuracy": 0.20403582453727723,
      "num_tokens": 10965466.0,
      "step": 6030
    },
    {
      "entropy": 5.581847620010376,
      "epoch": 5.184787279759346,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029366406019145735,
      "loss": 5.0863,
      "mean_token_accuracy": 0.20574845522642135,
      "num_tokens": 10975051.0,
      "step": 6035
    },
    {
      "entropy": 5.5563623905181885,
      "epoch": 5.189084658358402,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002933327030450143,
      "loss": 4.9932,
      "mean_token_accuracy": 0.21749001741409302,
      "num_tokens": 10983940.0,
      "step": 6040
    },
    {
      "entropy": 5.619913864135742,
      "epoch": 5.193382036957456,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029300130586727545,
      "loss": 5.103,
      "mean_token_accuracy": 0.20355535596609114,
      "num_tokens": 10994029.0,
      "step": 6045
    },
    {
      "entropy": 5.72647876739502,
      "epoch": 5.197679415556511,
      "grad_norm": 1.25,
      "learning_rate": 0.00029266986938187943,
      "loss": 5.1648,
      "mean_token_accuracy": 0.19960147142410278,
      "num_tokens": 11003616.0,
      "step": 6050
    },
    {
      "entropy": 5.5896858215332035,
      "epoch": 5.2019767941555655,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002923383943125514,
      "loss": 5.0434,
      "mean_token_accuracy": 0.21000080108642577,
      "num_tokens": 11012068.0,
      "step": 6055
    },
    {
      "entropy": 5.617713069915771,
      "epoch": 5.20627417275462,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002920068813831002,
      "loss": 5.0995,
      "mean_token_accuracy": 0.20867936313152313,
      "num_tokens": 11020510.0,
      "step": 6060
    },
    {
      "entropy": 5.594635248184204,
      "epoch": 5.210571551353675,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002916753313174178,
      "loss": 5.1098,
      "mean_token_accuracy": 0.20451925098896026,
      "num_tokens": 11029804.0,
      "step": 6065
    },
    {
      "entropy": 5.6301109313964846,
      "epoch": 5.214868929952729,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002913437448394768,
      "loss": 5.0885,
      "mean_token_accuracy": 0.21020381152629852,
      "num_tokens": 11038586.0,
      "step": 6070
    },
    {
      "entropy": 5.697144985198975,
      "epoch": 5.219166308551784,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029101212267332955,
      "loss": 5.1655,
      "mean_token_accuracy": 0.2009891614317894,
      "num_tokens": 11048240.0,
      "step": 6075
    },
    {
      "entropy": 5.621762275695801,
      "epoch": 5.223463687150838,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029068046554310637,
      "loss": 5.0361,
      "mean_token_accuracy": 0.20836093723773957,
      "num_tokens": 11056703.0,
      "step": 6080
    },
    {
      "entropy": 5.626343250274658,
      "epoch": 5.227761065749893,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002903487741730139,
      "loss": 5.0991,
      "mean_token_accuracy": 0.20425437688827514,
      "num_tokens": 11066246.0,
      "step": 6085
    },
    {
      "entropy": 5.631587648391724,
      "epoch": 5.232058444348947,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029001704928733354,
      "loss": 5.1357,
      "mean_token_accuracy": 0.20830837190151213,
      "num_tokens": 11075277.0,
      "step": 6090
    },
    {
      "entropy": 5.553812313079834,
      "epoch": 5.236355822948002,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002896852916104198,
      "loss": 5.0265,
      "mean_token_accuracy": 0.21738798767328263,
      "num_tokens": 11083759.0,
      "step": 6095
    },
    {
      "entropy": 5.645947551727295,
      "epoch": 5.240653201547056,
      "grad_norm": 1.25,
      "learning_rate": 0.0002893535018666988,
      "loss": 5.0112,
      "mean_token_accuracy": 0.21021998971700667,
      "num_tokens": 11091960.0,
      "step": 6100
    },
    {
      "entropy": 5.562863445281982,
      "epoch": 5.244950580146111,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00028902168078066674,
      "loss": 5.1232,
      "mean_token_accuracy": 0.20529252141714097,
      "num_tokens": 11101134.0,
      "step": 6105
    },
    {
      "entropy": 5.582944440841675,
      "epoch": 5.2492479587451655,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002886898290768883,
      "loss": 5.0625,
      "mean_token_accuracy": 0.20981228798627855,
      "num_tokens": 11110282.0,
      "step": 6110
    },
    {
      "entropy": 5.656411790847779,
      "epoch": 5.25354533734422,
      "grad_norm": 1.328125,
      "learning_rate": 0.000288357947479995,
      "loss": 5.1507,
      "mean_token_accuracy": 0.20492044240236282,
      "num_tokens": 11119591.0,
      "step": 6115
    },
    {
      "entropy": 5.66609902381897,
      "epoch": 5.257842715943275,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028802603671468347,
      "loss": 5.1272,
      "mean_token_accuracy": 0.2056983694434166,
      "num_tokens": 11129164.0,
      "step": 6120
    },
    {
      "entropy": 5.6240333080291744,
      "epoch": 5.262140094542329,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028769409750571413,
      "loss": 5.1071,
      "mean_token_accuracy": 0.20120790004730224,
      "num_tokens": 11137973.0,
      "step": 6125
    },
    {
      "entropy": 5.602197074890137,
      "epoch": 5.266437473141384,
      "grad_norm": 1.25,
      "learning_rate": 0.00028736213057790975,
      "loss": 5.1708,
      "mean_token_accuracy": 0.20217742919921874,
      "num_tokens": 11147285.0,
      "step": 6130
    },
    {
      "entropy": 5.662440013885498,
      "epoch": 5.270734851740438,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002870301366561533,
      "loss": 5.1532,
      "mean_token_accuracy": 0.20672872066497802,
      "num_tokens": 11155303.0,
      "step": 6135
    },
    {
      "entropy": 5.607999324798584,
      "epoch": 5.275032230339493,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002866981164653867,
      "loss": 4.9986,
      "mean_token_accuracy": 0.22090290039777755,
      "num_tokens": 11163553.0,
      "step": 6140
    },
    {
      "entropy": 5.549283981323242,
      "epoch": 5.279329608938547,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002863660707306095,
      "loss": 4.9519,
      "mean_token_accuracy": 0.22529712319374084,
      "num_tokens": 11171865.0,
      "step": 6145
    },
    {
      "entropy": 5.641676235198974,
      "epoch": 5.283626987537602,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028603400017687675,
      "loss": 5.1494,
      "mean_token_accuracy": 0.20834967344999314,
      "num_tokens": 11181137.0,
      "step": 6150
    },
    {
      "entropy": 5.490835523605346,
      "epoch": 5.287924366136656,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028570190552929794,
      "loss": 5.0123,
      "mean_token_accuracy": 0.22571585774421693,
      "num_tokens": 11190174.0,
      "step": 6155
    },
    {
      "entropy": 5.540759801864624,
      "epoch": 5.292221744735711,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000285369787513035,
      "loss": 5.034,
      "mean_token_accuracy": 0.21682270914316176,
      "num_tokens": 11197964.0,
      "step": 6160
    },
    {
      "entropy": 5.686545419692993,
      "epoch": 5.2965191233347655,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028503764685330077,
      "loss": 5.1294,
      "mean_token_accuracy": 0.21260504126548768,
      "num_tokens": 11207974.0,
      "step": 6165
    },
    {
      "entropy": 5.609825706481933,
      "epoch": 5.30081650193382,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028470548427535794,
      "loss": 5.0721,
      "mean_token_accuracy": 0.20835259407758713,
      "num_tokens": 11216430.0,
      "step": 6170
    },
    {
      "entropy": 5.601636028289795,
      "epoch": 5.305113880532875,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028437330050451654,
      "loss": 5.1755,
      "mean_token_accuracy": 0.19842823445796967,
      "num_tokens": 11226189.0,
      "step": 6175
    },
    {
      "entropy": 5.606355905532837,
      "epoch": 5.309411259131929,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002840410962661334,
      "loss": 5.1293,
      "mean_token_accuracy": 0.20548384189605712,
      "num_tokens": 11234691.0,
      "step": 6180
    },
    {
      "entropy": 5.55111517906189,
      "epoch": 5.313708637730985,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002837088722856098,
      "loss": 4.9634,
      "mean_token_accuracy": 0.22443660348653793,
      "num_tokens": 11243852.0,
      "step": 6185
    },
    {
      "entropy": 5.641619396209717,
      "epoch": 5.318006016330038,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028337662928838996,
      "loss": 5.1347,
      "mean_token_accuracy": 0.2034019872546196,
      "num_tokens": 11253416.0,
      "step": 6190
    },
    {
      "entropy": 5.578536081314087,
      "epoch": 5.322303394929094,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028304436799995986,
      "loss": 5.035,
      "mean_token_accuracy": 0.216870254278183,
      "num_tokens": 11262869.0,
      "step": 6195
    },
    {
      "entropy": 5.589238405227661,
      "epoch": 5.326600773528148,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028271208914584534,
      "loss": 5.1615,
      "mean_token_accuracy": 0.19855283051729203,
      "num_tokens": 11272386.0,
      "step": 6200
    },
    {
      "entropy": 5.64242992401123,
      "epoch": 5.330898152127203,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028237979345161065,
      "loss": 5.0082,
      "mean_token_accuracy": 0.21714682281017303,
      "num_tokens": 11281590.0,
      "step": 6205
    },
    {
      "entropy": 5.549373817443848,
      "epoch": 5.335195530726257,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002820474816428568,
      "loss": 5.0685,
      "mean_token_accuracy": 0.21339880228042601,
      "num_tokens": 11290873.0,
      "step": 6210
    },
    {
      "entropy": 5.585452938079834,
      "epoch": 5.339492909325312,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002817151544452198,
      "loss": 5.032,
      "mean_token_accuracy": 0.21075835078954697,
      "num_tokens": 11299064.0,
      "step": 6215
    },
    {
      "entropy": 5.603043460845948,
      "epoch": 5.343790287924366,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028138281258436947,
      "loss": 5.0912,
      "mean_token_accuracy": 0.20531507432460785,
      "num_tokens": 11307390.0,
      "step": 6220
    },
    {
      "entropy": 5.55312442779541,
      "epoch": 5.348087666523421,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002810504567860078,
      "loss": 5.0162,
      "mean_token_accuracy": 0.2138412669301033,
      "num_tokens": 11315606.0,
      "step": 6225
    },
    {
      "entropy": 5.65919771194458,
      "epoch": 5.3523850451224755,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002807180877758667,
      "loss": 5.1612,
      "mean_token_accuracy": 0.1957780212163925,
      "num_tokens": 11323821.0,
      "step": 6230
    },
    {
      "entropy": 5.606141757965088,
      "epoch": 5.35668242372153,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00028038570627970754,
      "loss": 5.094,
      "mean_token_accuracy": 0.21104342192411424,
      "num_tokens": 11331850.0,
      "step": 6235
    },
    {
      "entropy": 5.623849773406983,
      "epoch": 5.360979802320585,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002800533130233184,
      "loss": 5.0614,
      "mean_token_accuracy": 0.2049120232462883,
      "num_tokens": 11340125.0,
      "step": 6240
    },
    {
      "entropy": 5.654445934295654,
      "epoch": 5.365277180919639,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002797209087325135,
      "loss": 5.1203,
      "mean_token_accuracy": 0.20036470592021943,
      "num_tokens": 11349184.0,
      "step": 6245
    },
    {
      "entropy": 5.633037328720093,
      "epoch": 5.369574559518694,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027938849413313083,
      "loss": 5.1503,
      "mean_token_accuracy": 0.19971506297588348,
      "num_tokens": 11357536.0,
      "step": 6250
    },
    {
      "entropy": 5.584692192077637,
      "epoch": 5.373871938117748,
      "grad_norm": 1.296875,
      "learning_rate": 0.000279056069951031,
      "loss": 5.1252,
      "mean_token_accuracy": 0.21012808233499528,
      "num_tokens": 11367242.0,
      "step": 6255
    },
    {
      "entropy": 5.676827383041382,
      "epoch": 5.378169316716803,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027872363691209564,
      "loss": 5.0882,
      "mean_token_accuracy": 0.2116141527891159,
      "num_tokens": 11374932.0,
      "step": 6260
    },
    {
      "entropy": 5.633297729492187,
      "epoch": 5.382466695315857,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002783911957422256,
      "loss": 5.0915,
      "mean_token_accuracy": 0.20485981553792953,
      "num_tokens": 11383575.0,
      "step": 6265
    },
    {
      "entropy": 5.63403286933899,
      "epoch": 5.386764073914912,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002780587471673394,
      "loss": 5.1127,
      "mean_token_accuracy": 0.20935218185186386,
      "num_tokens": 11392285.0,
      "step": 6270
    },
    {
      "entropy": 5.624064111709595,
      "epoch": 5.391061452513966,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027772629191337206,
      "loss": 5.0973,
      "mean_token_accuracy": 0.21106354892253876,
      "num_tokens": 11401054.0,
      "step": 6275
    },
    {
      "entropy": 5.597875642776489,
      "epoch": 5.395358831113021,
      "grad_norm": 1.375,
      "learning_rate": 0.00027739383070627283,
      "loss": 5.1075,
      "mean_token_accuracy": 0.2075465053319931,
      "num_tokens": 11410529.0,
      "step": 6280
    },
    {
      "entropy": 5.58003830909729,
      "epoch": 5.3996562097120755,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002770613642720041,
      "loss": 5.0606,
      "mean_token_accuracy": 0.212888865172863,
      "num_tokens": 11419961.0,
      "step": 6285
    },
    {
      "entropy": 5.677826547622681,
      "epoch": 5.40395358831113,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027672889333653984,
      "loss": 5.2052,
      "mean_token_accuracy": 0.19134974032640456,
      "num_tokens": 11429529.0,
      "step": 6290
    },
    {
      "entropy": 5.621359586715698,
      "epoch": 5.408250966910185,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002763964186258635,
      "loss": 5.0667,
      "mean_token_accuracy": 0.21625042110681533,
      "num_tokens": 11438254.0,
      "step": 6295
    },
    {
      "entropy": 5.631159687042237,
      "epoch": 5.412548345509239,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002760639408659671,
      "loss": 5.1248,
      "mean_token_accuracy": 0.20336071997880936,
      "num_tokens": 11447587.0,
      "step": 6300
    },
    {
      "entropy": 5.612961101531982,
      "epoch": 5.416845724108294,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002757314607828489,
      "loss": 5.0746,
      "mean_token_accuracy": 0.21611074209213257,
      "num_tokens": 11455493.0,
      "step": 6305
    },
    {
      "entropy": 5.569397640228272,
      "epoch": 5.421143102707348,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027539897910251293,
      "loss": 5.049,
      "mean_token_accuracy": 0.2157267540693283,
      "num_tokens": 11464143.0,
      "step": 6310
    },
    {
      "entropy": 5.5827923774719235,
      "epoch": 5.425440481306403,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00027506649655096595,
      "loss": 5.0211,
      "mean_token_accuracy": 0.2126669242978096,
      "num_tokens": 11471813.0,
      "step": 6315
    },
    {
      "entropy": 5.6086499214172365,
      "epoch": 5.429737859905457,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002747340138542171,
      "loss": 5.0616,
      "mean_token_accuracy": 0.20510358214378357,
      "num_tokens": 11481374.0,
      "step": 6320
    },
    {
      "entropy": 5.729965400695801,
      "epoch": 5.434035238504512,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002744015317382757,
      "loss": 5.2225,
      "mean_token_accuracy": 0.19964760094881057,
      "num_tokens": 11490575.0,
      "step": 6325
    },
    {
      "entropy": 5.632559537887573,
      "epoch": 5.438332617103566,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002740690509291498,
      "loss": 5.1102,
      "mean_token_accuracy": 0.20633225739002228,
      "num_tokens": 11499898.0,
      "step": 6330
    },
    {
      "entropy": 5.634370899200439,
      "epoch": 5.442629995702621,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002737365721528445,
      "loss": 5.0888,
      "mean_token_accuracy": 0.20238687992095947,
      "num_tokens": 11508544.0,
      "step": 6335
    },
    {
      "entropy": 5.600213670730591,
      "epoch": 5.446927374301676,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002734040961353607,
      "loss": 5.0949,
      "mean_token_accuracy": 0.214790241420269,
      "num_tokens": 11519239.0,
      "step": 6340
    },
    {
      "entropy": 5.605601978302002,
      "epoch": 5.451224752900731,
      "grad_norm": 1.3125,
      "learning_rate": 0.000273071623602693,
      "loss": 5.0464,
      "mean_token_accuracy": 0.21412968635559082,
      "num_tokens": 11529014.0,
      "step": 6345
    },
    {
      "entropy": 5.604939794540405,
      "epoch": 5.4555221314997855,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027273915528082865,
      "loss": 5.0639,
      "mean_token_accuracy": 0.2144625097513199,
      "num_tokens": 11538367.0,
      "step": 6350
    },
    {
      "entropy": 5.6622786045074465,
      "epoch": 5.45981951009884,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002724066918957455,
      "loss": 5.2255,
      "mean_token_accuracy": 0.19533251971006393,
      "num_tokens": 11548166.0,
      "step": 6355
    },
    {
      "entropy": 5.531717109680176,
      "epoch": 5.464116888697895,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002720742341734107,
      "loss": 5.0778,
      "mean_token_accuracy": 0.21518887877464293,
      "num_tokens": 11557187.0,
      "step": 6360
    },
    {
      "entropy": 5.683267259597779,
      "epoch": 5.468414267296949,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027174178283977904,
      "loss": 5.1283,
      "mean_token_accuracy": 0.20605890303850175,
      "num_tokens": 11566181.0,
      "step": 6365
    },
    {
      "entropy": 5.59360704421997,
      "epoch": 5.472711645896004,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027140933862079136,
      "loss": 5.0827,
      "mean_token_accuracy": 0.21250177323818206,
      "num_tokens": 11576157.0,
      "step": 6370
    },
    {
      "entropy": 5.550892305374146,
      "epoch": 5.477009024495058,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000271076902242373,
      "loss": 5.0335,
      "mean_token_accuracy": 0.2080536112189293,
      "num_tokens": 11585325.0,
      "step": 6375
    },
    {
      "entropy": 5.569453763961792,
      "epoch": 5.481306403094113,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000270744474430432,
      "loss": 5.0228,
      "mean_token_accuracy": 0.21742244958877563,
      "num_tokens": 11594623.0,
      "step": 6380
    },
    {
      "entropy": 5.697995948791504,
      "epoch": 5.485603781693167,
      "grad_norm": 1.328125,
      "learning_rate": 0.000270412055910858,
      "loss": 5.225,
      "mean_token_accuracy": 0.20334868282079696,
      "num_tokens": 11604370.0,
      "step": 6385
    },
    {
      "entropy": 5.550177526473999,
      "epoch": 5.489901160292222,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002700796474095201,
      "loss": 5.0204,
      "mean_token_accuracy": 0.22556949704885482,
      "num_tokens": 11613779.0,
      "step": 6390
    },
    {
      "entropy": 5.60788049697876,
      "epoch": 5.494198538891276,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002697472496522656,
      "loss": 5.0986,
      "mean_token_accuracy": 0.20862444788217543,
      "num_tokens": 11623037.0,
      "step": 6395
    },
    {
      "entropy": 5.662399482727051,
      "epoch": 5.498495917490331,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002694148633649184,
      "loss": 5.1355,
      "mean_token_accuracy": 0.20515082776546478,
      "num_tokens": 11631640.0,
      "step": 6400
    },
    {
      "entropy": 5.621501350402832,
      "epoch": 5.5027932960893855,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002690824892732772,
      "loss": 5.0968,
      "mean_token_accuracy": 0.2078380689024925,
      "num_tokens": 11640500.0,
      "step": 6405
    },
    {
      "entropy": 5.643038368225097,
      "epoch": 5.50709067468844,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002687501281031142,
      "loss": 5.1564,
      "mean_token_accuracy": 0.2032139077782631,
      "num_tokens": 11649173.0,
      "step": 6410
    },
    {
      "entropy": 5.470699501037598,
      "epoch": 5.511388053287495,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002684177805801734,
      "loss": 4.9775,
      "mean_token_accuracy": 0.22281333953142166,
      "num_tokens": 11658808.0,
      "step": 6415
    },
    {
      "entropy": 5.60187783241272,
      "epoch": 5.515685431886549,
      "grad_norm": 1.25,
      "learning_rate": 0.00026808544743016886,
      "loss": 5.0991,
      "mean_token_accuracy": 0.20230972170829772,
      "num_tokens": 11667600.0,
      "step": 6420
    },
    {
      "entropy": 5.585718154907227,
      "epoch": 5.519982810485604,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002677531293787835,
      "loss": 5.093,
      "mean_token_accuracy": 0.20506487935781478,
      "num_tokens": 11675597.0,
      "step": 6425
    },
    {
      "entropy": 5.477433729171753,
      "epoch": 5.524280189084658,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000267420827151667,
      "loss": 5.0229,
      "mean_token_accuracy": 0.21978774070739746,
      "num_tokens": 11684549.0,
      "step": 6430
    },
    {
      "entropy": 5.574510002136231,
      "epoch": 5.528577567683713,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002670885414744347,
      "loss": 5.1091,
      "mean_token_accuracy": 0.20890313535928726,
      "num_tokens": 11693043.0,
      "step": 6435
    },
    {
      "entropy": 5.688928318023682,
      "epoch": 5.532874946282767,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002667562730726655,
      "loss": 5.2027,
      "mean_token_accuracy": 0.20405805706977845,
      "num_tokens": 11702982.0,
      "step": 6440
    },
    {
      "entropy": 5.6774157047271725,
      "epoch": 5.537172324881822,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026642402267190095,
      "loss": 5.1964,
      "mean_token_accuracy": 0.2006876528263092,
      "num_tokens": 11711994.0,
      "step": 6445
    },
    {
      "entropy": 5.645321416854858,
      "epoch": 5.541469703480876,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026609179099764313,
      "loss": 5.1305,
      "mean_token_accuracy": 0.20326993465423585,
      "num_tokens": 11722165.0,
      "step": 6450
    },
    {
      "entropy": 5.614281892776489,
      "epoch": 5.545767082079931,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026575957877535323,
      "loss": 5.1042,
      "mean_token_accuracy": 0.20908745974302292,
      "num_tokens": 11731265.0,
      "step": 6455
    },
    {
      "entropy": 5.652119207382202,
      "epoch": 5.5500644606789855,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026542738673044985,
      "loss": 5.131,
      "mean_token_accuracy": 0.206353822350502,
      "num_tokens": 11741779.0,
      "step": 6460
    },
    {
      "entropy": 5.588283205032349,
      "epoch": 5.55436183927804,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002650952155883077,
      "loss": 5.0848,
      "mean_token_accuracy": 0.213355553150177,
      "num_tokens": 11749976.0,
      "step": 6465
    },
    {
      "entropy": 5.574606037139892,
      "epoch": 5.558659217877095,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002647630660742559,
      "loss": 5.1052,
      "mean_token_accuracy": 0.2052710771560669,
      "num_tokens": 11759781.0,
      "step": 6470
    },
    {
      "entropy": 5.604959440231323,
      "epoch": 5.56295659647615,
      "grad_norm": 1.203125,
      "learning_rate": 0.000264430938913576,
      "loss": 5.0664,
      "mean_token_accuracy": 0.209891377389431,
      "num_tokens": 11769544.0,
      "step": 6475
    },
    {
      "entropy": 5.611039066314698,
      "epoch": 5.567253975075204,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026409883483150123,
      "loss": 5.0606,
      "mean_token_accuracy": 0.2052990049123764,
      "num_tokens": 11778831.0,
      "step": 6480
    },
    {
      "entropy": 5.6141054153442385,
      "epoch": 5.571551353674259,
      "grad_norm": 1.3125,
      "learning_rate": 0.000263766754553214,
      "loss": 5.1548,
      "mean_token_accuracy": 0.20090280175209047,
      "num_tokens": 11788813.0,
      "step": 6485
    },
    {
      "entropy": 5.645803022384643,
      "epoch": 5.575848732273313,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002634346988038448,
      "loss": 5.0806,
      "mean_token_accuracy": 0.20870089083909987,
      "num_tokens": 11797335.0,
      "step": 6490
    },
    {
      "entropy": 5.602471971511841,
      "epoch": 5.580146110872368,
      "grad_norm": 1.375,
      "learning_rate": 0.00026310266830847093,
      "loss": 5.1018,
      "mean_token_accuracy": 0.207021065056324,
      "num_tokens": 11806741.0,
      "step": 6495
    },
    {
      "entropy": 5.6450357913970945,
      "epoch": 5.584443489471423,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026277066379211406,
      "loss": 5.1245,
      "mean_token_accuracy": 0.20850823819637299,
      "num_tokens": 11815551.0,
      "step": 6500
    },
    {
      "epoch": 5.584443489471423,
      "eval_entropy": 5.473724893621497,
      "eval_loss": 5.9721598625183105,
      "eval_mean_token_accuracy": 0.1755129244044289,
      "eval_num_tokens": 11815551.0,
      "eval_runtime": 2.6252,
      "eval_samples_per_second": 1351.903,
      "eval_steps_per_second": 169.131,
      "step": 6500
    },
    {
      "entropy": 5.639531564712525,
      "epoch": 5.588740868070477,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002624386859797396,
      "loss": 5.1651,
      "mean_token_accuracy": 0.20570480823516846,
      "num_tokens": 11824483.0,
      "step": 6505
    },
    {
      "entropy": 5.533856678009033,
      "epoch": 5.593038246669532,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026210673559625406,
      "loss": 4.9479,
      "mean_token_accuracy": 0.22729039788246155,
      "num_tokens": 11832383.0,
      "step": 6510
    },
    {
      "entropy": 5.6554076194763185,
      "epoch": 5.597335625268586,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002617748133665047,
      "loss": 5.1921,
      "mean_token_accuracy": 0.2002184957265854,
      "num_tokens": 11841430.0,
      "step": 6515
    },
    {
      "entropy": 5.5909423828125,
      "epoch": 5.601633003867641,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002614429200152768,
      "loss": 5.1573,
      "mean_token_accuracy": 0.2021533489227295,
      "num_tokens": 11850863.0,
      "step": 6520
    },
    {
      "entropy": 5.570891809463501,
      "epoch": 5.6059303824666955,
      "grad_norm": 1.375,
      "learning_rate": 0.000261111056267293,
      "loss": 5.0111,
      "mean_token_accuracy": 0.21224440038204193,
      "num_tokens": 11859392.0,
      "step": 6525
    },
    {
      "entropy": 5.6367401599884035,
      "epoch": 5.61022776106575,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026077922284721084,
      "loss": 5.0562,
      "mean_token_accuracy": 0.2160949394106865,
      "num_tokens": 11868762.0,
      "step": 6530
    },
    {
      "entropy": 5.565837287902832,
      "epoch": 5.614525139664805,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00026044742047962206,
      "loss": 5.0423,
      "mean_token_accuracy": 0.22076621800661086,
      "num_tokens": 11876722.0,
      "step": 6535
    },
    {
      "entropy": 5.580392551422119,
      "epoch": 5.618822518263859,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026011564988905023,
      "loss": 5.1742,
      "mean_token_accuracy": 0.20562173277139664,
      "num_tokens": 11885614.0,
      "step": 6540
    },
    {
      "entropy": 5.672798585891724,
      "epoch": 5.623119896862914,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002597839117999499,
      "loss": 5.1727,
      "mean_token_accuracy": 0.20341680496931075,
      "num_tokens": 11894702.0,
      "step": 6545
    },
    {
      "entropy": 5.595587491989136,
      "epoch": 5.627417275461968,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002594522069367044,
      "loss": 5.0774,
      "mean_token_accuracy": 0.2159542426466942,
      "num_tokens": 11902829.0,
      "step": 6550
    },
    {
      "entropy": 5.5586662769317625,
      "epoch": 5.631714654061023,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002591205360236245,
      "loss": 5.0955,
      "mean_token_accuracy": 0.20684662759304046,
      "num_tokens": 11912377.0,
      "step": 6555
    },
    {
      "entropy": 5.598919820785523,
      "epoch": 5.636012032660077,
      "grad_norm": 1.390625,
      "learning_rate": 0.000258788899784947,
      "loss": 5.1072,
      "mean_token_accuracy": 0.20253228396177292,
      "num_tokens": 11920563.0,
      "step": 6560
    },
    {
      "entropy": 5.634590768814087,
      "epoch": 5.640309411259132,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025845729894483283,
      "loss": 5.1358,
      "mean_token_accuracy": 0.20433918684720992,
      "num_tokens": 11930190.0,
      "step": 6565
    },
    {
      "entropy": 5.626949787139893,
      "epoch": 5.644606789858186,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002581257342273657,
      "loss": 5.196,
      "mean_token_accuracy": 0.20145849138498306,
      "num_tokens": 11939840.0,
      "step": 6570
    },
    {
      "entropy": 5.622612953186035,
      "epoch": 5.648904168457241,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002577942063565504,
      "loss": 5.1356,
      "mean_token_accuracy": 0.20517174452543258,
      "num_tokens": 11948260.0,
      "step": 6575
    },
    {
      "entropy": 5.6295702934265135,
      "epoch": 5.6532015470562955,
      "grad_norm": 1.5,
      "learning_rate": 0.0002574627160563114,
      "loss": 5.175,
      "mean_token_accuracy": 0.2095462918281555,
      "num_tokens": 11956776.0,
      "step": 6580
    },
    {
      "entropy": 5.696208381652832,
      "epoch": 5.65749892565535,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002571312640504909,
      "loss": 5.1983,
      "mean_token_accuracy": 0.20611528307199478,
      "num_tokens": 11966375.0,
      "step": 6585
    },
    {
      "entropy": 5.7025713443756105,
      "epoch": 5.661796304254405,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002567998510628476,
      "loss": 5.2235,
      "mean_token_accuracy": 0.19549818336963654,
      "num_tokens": 11975835.0,
      "step": 6590
    },
    {
      "entropy": 5.625443983078003,
      "epoch": 5.666093682853459,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025646847781705506,
      "loss": 5.0885,
      "mean_token_accuracy": 0.20967599004507065,
      "num_tokens": 11984672.0,
      "step": 6595
    },
    {
      "entropy": 5.639520692825317,
      "epoch": 5.670391061452514,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002561371450367,
      "loss": 5.1098,
      "mean_token_accuracy": 0.20212200433015823,
      "num_tokens": 11993954.0,
      "step": 6600
    },
    {
      "entropy": 5.600101089477539,
      "epoch": 5.674688440051568,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025580585344528076,
      "loss": 5.1686,
      "mean_token_accuracy": 0.20414370447397232,
      "num_tokens": 12002523.0,
      "step": 6605
    },
    {
      "entropy": 5.640940093994141,
      "epoch": 5.678985818650623,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002554746037662058,
      "loss": 5.1931,
      "mean_token_accuracy": 0.1946851432323456,
      "num_tokens": 12011638.0,
      "step": 6610
    },
    {
      "entropy": 5.6768091201782225,
      "epoch": 5.683283197249677,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002551433967227919,
      "loss": 5.1369,
      "mean_token_accuracy": 0.20348850190639495,
      "num_tokens": 12021319.0,
      "step": 6615
    },
    {
      "entropy": 5.618343257904053,
      "epoch": 5.687580575848732,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000254812233038263,
      "loss": 5.0957,
      "mean_token_accuracy": 0.2116297960281372,
      "num_tokens": 12030255.0,
      "step": 6620
    },
    {
      "entropy": 5.616571569442749,
      "epoch": 5.691877954447786,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025448111343574813,
      "loss": 5.0926,
      "mean_token_accuracy": 0.20955381393432618,
      "num_tokens": 12038884.0,
      "step": 6625
    },
    {
      "entropy": 5.605348443984985,
      "epoch": 5.696175333046842,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002541500386382802,
      "loss": 5.0751,
      "mean_token_accuracy": 0.21031966358423232,
      "num_tokens": 12047477.0,
      "step": 6630
    },
    {
      "entropy": 5.588452768325806,
      "epoch": 5.7004727116458955,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025381900936879433,
      "loss": 5.1595,
      "mean_token_accuracy": 0.2018212527036667,
      "num_tokens": 12056902.0,
      "step": 6635
    },
    {
      "entropy": 5.649710464477539,
      "epoch": 5.704770090244951,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002534880263501259,
      "loss": 5.1021,
      "mean_token_accuracy": 0.20528373420238494,
      "num_tokens": 12065721.0,
      "step": 6640
    },
    {
      "entropy": 5.5990218162536625,
      "epoch": 5.7090674688440055,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002531570903050097,
      "loss": 5.0812,
      "mean_token_accuracy": 0.21414047330617905,
      "num_tokens": 12074870.0,
      "step": 6645
    },
    {
      "entropy": 5.645281839370727,
      "epoch": 5.71336484744306,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002528262019560776,
      "loss": 5.116,
      "mean_token_accuracy": 0.20314151644706727,
      "num_tokens": 12084557.0,
      "step": 6650
    },
    {
      "entropy": 5.5747274398803714,
      "epoch": 5.717662226042115,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002524953620258579,
      "loss": 4.9728,
      "mean_token_accuracy": 0.22318569868803023,
      "num_tokens": 12093074.0,
      "step": 6655
    },
    {
      "entropy": 5.56484808921814,
      "epoch": 5.721959604641169,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002521645712367724,
      "loss": 5.008,
      "mean_token_accuracy": 0.21946241110563278,
      "num_tokens": 12102785.0,
      "step": 6660
    },
    {
      "entropy": 5.537666368484497,
      "epoch": 5.726256983240224,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025183383031113606,
      "loss": 5.0451,
      "mean_token_accuracy": 0.21798370480537416,
      "num_tokens": 12112535.0,
      "step": 6665
    },
    {
      "entropy": 5.489774513244629,
      "epoch": 5.730554361839278,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025150313997115476,
      "loss": 5.0234,
      "mean_token_accuracy": 0.2155287116765976,
      "num_tokens": 12121604.0,
      "step": 6670
    },
    {
      "entropy": 5.610029220581055,
      "epoch": 5.734851740438333,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002511725009389244,
      "loss": 5.1741,
      "mean_token_accuracy": 0.20120700746774672,
      "num_tokens": 12131276.0,
      "step": 6675
    },
    {
      "entropy": 5.630321931838989,
      "epoch": 5.739149119037387,
      "grad_norm": 1.421875,
      "learning_rate": 0.000250841913936428,
      "loss": 5.1416,
      "mean_token_accuracy": 0.21131885945796966,
      "num_tokens": 12140180.0,
      "step": 6680
    },
    {
      "entropy": 5.603497123718261,
      "epoch": 5.743446497636442,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002505113796855357,
      "loss": 5.1261,
      "mean_token_accuracy": 0.20314648896455764,
      "num_tokens": 12149635.0,
      "step": 6685
    },
    {
      "entropy": 5.620114374160766,
      "epoch": 5.747743876235496,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025018089890800225,
      "loss": 5.083,
      "mean_token_accuracy": 0.2130756914615631,
      "num_tokens": 12157565.0,
      "step": 6690
    },
    {
      "entropy": 5.624129390716552,
      "epoch": 5.752041254834551,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024985047232546544,
      "loss": 5.1682,
      "mean_token_accuracy": 0.20432327687740326,
      "num_tokens": 12166647.0,
      "step": 6695
    },
    {
      "entropy": 5.5786402225494385,
      "epoch": 5.7563386334336055,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024952010065944485,
      "loss": 5.1285,
      "mean_token_accuracy": 0.20651274919509888,
      "num_tokens": 12175554.0,
      "step": 6700
    },
    {
      "entropy": 5.654118490219116,
      "epoch": 5.76063601203266,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002491897846313402,
      "loss": 5.1228,
      "mean_token_accuracy": 0.21216176599264144,
      "num_tokens": 12184756.0,
      "step": 6705
    },
    {
      "entropy": 5.6816307544708256,
      "epoch": 5.764933390631715,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002488595249624297,
      "loss": 5.1973,
      "mean_token_accuracy": 0.200140979886055,
      "num_tokens": 12194724.0,
      "step": 6710
    },
    {
      "entropy": 5.633313941955566,
      "epoch": 5.769230769230769,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024852932237386837,
      "loss": 5.1561,
      "mean_token_accuracy": 0.20503623485565187,
      "num_tokens": 12203804.0,
      "step": 6715
    },
    {
      "entropy": 5.61313648223877,
      "epoch": 5.773528147829824,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024819917758668673,
      "loss": 5.1063,
      "mean_token_accuracy": 0.20522952675819398,
      "num_tokens": 12212868.0,
      "step": 6720
    },
    {
      "entropy": 5.638158321380615,
      "epoch": 5.777825526428878,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024786909132178906,
      "loss": 5.1704,
      "mean_token_accuracy": 0.20416699945926667,
      "num_tokens": 12221650.0,
      "step": 6725
    },
    {
      "entropy": 5.608384847640991,
      "epoch": 5.782122905027933,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024753906429995194,
      "loss": 5.0425,
      "mean_token_accuracy": 0.2169565737247467,
      "num_tokens": 12231541.0,
      "step": 6730
    },
    {
      "entropy": 5.665906286239624,
      "epoch": 5.786420283626987,
      "grad_norm": 1.375,
      "learning_rate": 0.0002472090972418222,
      "loss": 5.1741,
      "mean_token_accuracy": 0.1999053344130516,
      "num_tokens": 12240899.0,
      "step": 6735
    },
    {
      "entropy": 5.632330083847046,
      "epoch": 5.790717662226042,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002468791908679163,
      "loss": 5.1218,
      "mean_token_accuracy": 0.20034747570753098,
      "num_tokens": 12250352.0,
      "step": 6740
    },
    {
      "entropy": 5.659326839447021,
      "epoch": 5.795015040825096,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002465493458986175,
      "loss": 5.202,
      "mean_token_accuracy": 0.19808356314897538,
      "num_tokens": 12259975.0,
      "step": 6745
    },
    {
      "entropy": 5.516726970672607,
      "epoch": 5.799312419424151,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024621956305417587,
      "loss": 5.0374,
      "mean_token_accuracy": 0.21402962505817413,
      "num_tokens": 12269203.0,
      "step": 6750
    },
    {
      "entropy": 5.620855379104614,
      "epoch": 5.8036097980232055,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000245889843054705,
      "loss": 5.1273,
      "mean_token_accuracy": 0.2102059692144394,
      "num_tokens": 12279481.0,
      "step": 6755
    },
    {
      "entropy": 5.576697921752929,
      "epoch": 5.80790717662226,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00024556018662018163,
      "loss": 5.1278,
      "mean_token_accuracy": 0.2093784898519516,
      "num_tokens": 12288848.0,
      "step": 6760
    },
    {
      "entropy": 5.672864055633545,
      "epoch": 5.812204555221315,
      "grad_norm": 1.25,
      "learning_rate": 0.00024523059447044377,
      "loss": 5.1127,
      "mean_token_accuracy": 0.2025643453001976,
      "num_tokens": 12297346.0,
      "step": 6765
    },
    {
      "entropy": 5.6612489223480225,
      "epoch": 5.816501933820369,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002449010673251887,
      "loss": 5.108,
      "mean_token_accuracy": 0.199971479177475,
      "num_tokens": 12306233.0,
      "step": 6770
    },
    {
      "entropy": 5.666787195205688,
      "epoch": 5.820799312419425,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002445716059039723,
      "loss": 5.2152,
      "mean_token_accuracy": 0.19629299491643906,
      "num_tokens": 12315609.0,
      "step": 6775
    },
    {
      "entropy": 5.650843477249145,
      "epoch": 5.825096691018478,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00024424221092620644,
      "loss": 5.1698,
      "mean_token_accuracy": 0.20234289318323134,
      "num_tokens": 12323915.0,
      "step": 6780
    },
    {
      "entropy": 5.559540605545044,
      "epoch": 5.829394069617534,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024391288311115822,
      "loss": 5.0729,
      "mean_token_accuracy": 0.20640904158353807,
      "num_tokens": 12334077.0,
      "step": 6785
    },
    {
      "entropy": 5.607386589050293,
      "epoch": 5.833691448216588,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002435836231779478,
      "loss": 5.1449,
      "mean_token_accuracy": 0.20625257939100267,
      "num_tokens": 12342411.0,
      "step": 6790
    },
    {
      "entropy": 5.560425472259522,
      "epoch": 5.837988826815643,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024325443184554724,
      "loss": 5.0201,
      "mean_token_accuracy": 0.22260660976171492,
      "num_tokens": 12351308.0,
      "step": 6795
    },
    {
      "entropy": 5.589815759658814,
      "epoch": 5.842286205414697,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024292530983277904,
      "loss": 5.1819,
      "mean_token_accuracy": 0.20337367951869964,
      "num_tokens": 12359673.0,
      "step": 6800
    },
    {
      "entropy": 5.601134204864502,
      "epoch": 5.846583584013752,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024259625785831408,
      "loss": 5.0346,
      "mean_token_accuracy": 0.21620783656835557,
      "num_tokens": 12367876.0,
      "step": 6805
    },
    {
      "entropy": 5.65500841140747,
      "epoch": 5.850880962612806,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024226727664067023,
      "loss": 5.173,
      "mean_token_accuracy": 0.20276736170053483,
      "num_tokens": 12377040.0,
      "step": 6810
    },
    {
      "entropy": 5.708501243591309,
      "epoch": 5.855178341211861,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024193836689821109,
      "loss": 5.2501,
      "mean_token_accuracy": 0.19747238904237746,
      "num_tokens": 12387622.0,
      "step": 6815
    },
    {
      "entropy": 5.525884103775025,
      "epoch": 5.8594757198109155,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002416095293491439,
      "loss": 5.0212,
      "mean_token_accuracy": 0.21559033989906312,
      "num_tokens": 12396447.0,
      "step": 6820
    },
    {
      "entropy": 5.571948862075805,
      "epoch": 5.86377309840997,
      "grad_norm": 1.375,
      "learning_rate": 0.0002412807647115186,
      "loss": 5.0373,
      "mean_token_accuracy": 0.22129399627447127,
      "num_tokens": 12405887.0,
      "step": 6825
    },
    {
      "entropy": 5.589397811889649,
      "epoch": 5.868070477009025,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024095207370322574,
      "loss": 5.0781,
      "mean_token_accuracy": 0.20624638497829437,
      "num_tokens": 12414543.0,
      "step": 6830
    },
    {
      "entropy": 5.583137893676758,
      "epoch": 5.872367855608079,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024062345704199507,
      "loss": 5.0716,
      "mean_token_accuracy": 0.2093906432390213,
      "num_tokens": 12423370.0,
      "step": 6835
    },
    {
      "entropy": 5.667108011245728,
      "epoch": 5.876665234207134,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00024029491544539405,
      "loss": 5.1929,
      "mean_token_accuracy": 0.19691274762153627,
      "num_tokens": 12433980.0,
      "step": 6840
    },
    {
      "entropy": 5.582393407821655,
      "epoch": 5.880962612806188,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023996644963082616,
      "loss": 5.1202,
      "mean_token_accuracy": 0.21780195087194443,
      "num_tokens": 12443300.0,
      "step": 6845
    },
    {
      "entropy": 5.59122257232666,
      "epoch": 5.885259991405243,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023963806031552948,
      "loss": 5.1719,
      "mean_token_accuracy": 0.20466887801885605,
      "num_tokens": 12452462.0,
      "step": 6850
    },
    {
      "entropy": 5.524906063079834,
      "epoch": 5.889557370004297,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023930974821657504,
      "loss": 5.08,
      "mean_token_accuracy": 0.21200208961963654,
      "num_tokens": 12461605.0,
      "step": 6855
    },
    {
      "entropy": 5.729503440856933,
      "epoch": 5.893854748603352,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023898151405086533,
      "loss": 5.1552,
      "mean_token_accuracy": 0.20130899399518967,
      "num_tokens": 12470905.0,
      "step": 6860
    },
    {
      "entropy": 5.6675981998443605,
      "epoch": 5.8981521272024064,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023865335853513232,
      "loss": 5.149,
      "mean_token_accuracy": 0.20523007810115815,
      "num_tokens": 12478913.0,
      "step": 6865
    },
    {
      "entropy": 5.608606386184692,
      "epoch": 5.902449505801461,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023832528238593677,
      "loss": 5.235,
      "mean_token_accuracy": 0.20041323751211165,
      "num_tokens": 12487561.0,
      "step": 6870
    },
    {
      "entropy": 5.584070920944214,
      "epoch": 5.9067468844005155,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023799728631966556,
      "loss": 5.0991,
      "mean_token_accuracy": 0.21308113932609557,
      "num_tokens": 12496781.0,
      "step": 6875
    },
    {
      "entropy": 5.667148685455322,
      "epoch": 5.91104426299957,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002376693710525313,
      "loss": 5.1877,
      "mean_token_accuracy": 0.19720988422632219,
      "num_tokens": 12505716.0,
      "step": 6880
    },
    {
      "entropy": 5.661831998825074,
      "epoch": 5.915341641598625,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023734153730056967,
      "loss": 5.1433,
      "mean_token_accuracy": 0.20124264061450958,
      "num_tokens": 12515594.0,
      "step": 6885
    },
    {
      "entropy": 5.6076099395751955,
      "epoch": 5.919639020197679,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023701378577963873,
      "loss": 5.0053,
      "mean_token_accuracy": 0.22576265037059784,
      "num_tokens": 12523439.0,
      "step": 6890
    },
    {
      "entropy": 5.5970909118652346,
      "epoch": 5.923936398796734,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002366861172054166,
      "loss": 5.0709,
      "mean_token_accuracy": 0.21236156225204467,
      "num_tokens": 12532100.0,
      "step": 6895
    },
    {
      "entropy": 5.567718172073365,
      "epoch": 5.928233777395788,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023635853229340054,
      "loss": 5.1159,
      "mean_token_accuracy": 0.20989579558372498,
      "num_tokens": 12539689.0,
      "step": 6900
    },
    {
      "entropy": 5.562050867080688,
      "epoch": 5.932531155994843,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023603103175890512,
      "loss": 5.1356,
      "mean_token_accuracy": 0.19828849881887436,
      "num_tokens": 12548486.0,
      "step": 6905
    },
    {
      "entropy": 5.607208585739135,
      "epoch": 5.936828534593897,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023570361631706062,
      "loss": 5.012,
      "mean_token_accuracy": 0.22305711805820466,
      "num_tokens": 12557423.0,
      "step": 6910
    },
    {
      "entropy": 5.62832088470459,
      "epoch": 5.941125913192952,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023537628668281142,
      "loss": 5.1811,
      "mean_token_accuracy": 0.20315117239952088,
      "num_tokens": 12566086.0,
      "step": 6915
    },
    {
      "entropy": 5.553533887863159,
      "epoch": 5.945423291792007,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023504904357091468,
      "loss": 5.0695,
      "mean_token_accuracy": 0.21100456416606903,
      "num_tokens": 12575827.0,
      "step": 6920
    },
    {
      "entropy": 5.609203481674195,
      "epoch": 5.949720670391061,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002347218876959384,
      "loss": 5.0641,
      "mean_token_accuracy": 0.21586052030324937,
      "num_tokens": 12585044.0,
      "step": 6925
    },
    {
      "entropy": 5.611778116226196,
      "epoch": 5.954018048990116,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002343948197722604,
      "loss": 5.0984,
      "mean_token_accuracy": 0.20927244275808335,
      "num_tokens": 12594677.0,
      "step": 6930
    },
    {
      "entropy": 5.562850570678711,
      "epoch": 5.958315427589171,
      "grad_norm": 1.375,
      "learning_rate": 0.00023406784051406638,
      "loss": 5.1121,
      "mean_token_accuracy": 0.20633335709571837,
      "num_tokens": 12604829.0,
      "step": 6935
    },
    {
      "entropy": 5.720876455307007,
      "epoch": 5.9626128061882255,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023374095063534816,
      "loss": 5.1027,
      "mean_token_accuracy": 0.21204436123371123,
      "num_tokens": 12613869.0,
      "step": 6940
    },
    {
      "entropy": 5.624848699569702,
      "epoch": 5.96691018478728,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023341415084990276,
      "loss": 5.0909,
      "mean_token_accuracy": 0.21083870828151702,
      "num_tokens": 12623248.0,
      "step": 6945
    },
    {
      "entropy": 5.465065288543701,
      "epoch": 5.971207563386335,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023308744187132996,
      "loss": 5.0138,
      "mean_token_accuracy": 0.22293835133314133,
      "num_tokens": 12631973.0,
      "step": 6950
    },
    {
      "entropy": 5.612492275238037,
      "epoch": 5.975504941985389,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023276082441303197,
      "loss": 5.1493,
      "mean_token_accuracy": 0.20858085602521897,
      "num_tokens": 12641435.0,
      "step": 6955
    },
    {
      "entropy": 5.652571535110473,
      "epoch": 5.979802320584444,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023243429918821056,
      "loss": 5.1152,
      "mean_token_accuracy": 0.20603489577770234,
      "num_tokens": 12651077.0,
      "step": 6960
    },
    {
      "entropy": 5.682791709899902,
      "epoch": 5.984099699183498,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023210786690986646,
      "loss": 5.2036,
      "mean_token_accuracy": 0.20165102630853654,
      "num_tokens": 12659929.0,
      "step": 6965
    },
    {
      "entropy": 5.562752199172974,
      "epoch": 5.988397077782553,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023178152829079712,
      "loss": 5.0435,
      "mean_token_accuracy": 0.2154797360301018,
      "num_tokens": 12670725.0,
      "step": 6970
    },
    {
      "entropy": 5.573461532592773,
      "epoch": 5.992694456381607,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023145528404359562,
      "loss": 5.1089,
      "mean_token_accuracy": 0.21357564628124237,
      "num_tokens": 12680820.0,
      "step": 6975
    },
    {
      "entropy": 5.659629678726196,
      "epoch": 5.996991834980662,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002311291348806492,
      "loss": 5.1874,
      "mean_token_accuracy": 0.1981000930070877,
      "num_tokens": 12689785.0,
      "step": 6980
    },
    {
      "entropy": 5.607626703050402,
      "epoch": 6.000859475719811,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002308030815141372,
      "loss": 5.0913,
      "mean_token_accuracy": 0.2045852823389901,
      "num_tokens": 12697221.0,
      "step": 6985
    },
    {
      "entropy": 5.568561220169068,
      "epoch": 6.005156854318866,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023047712465602976,
      "loss": 4.8902,
      "mean_token_accuracy": 0.2256504088640213,
      "num_tokens": 12707127.0,
      "step": 6990
    },
    {
      "entropy": 5.546060848236084,
      "epoch": 6.00945423291792,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023015126501808641,
      "loss": 4.9056,
      "mean_token_accuracy": 0.21919240355491637,
      "num_tokens": 12715364.0,
      "step": 6995
    },
    {
      "entropy": 5.612027645111084,
      "epoch": 6.013751611516975,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022982550331185437,
      "loss": 4.9182,
      "mean_token_accuracy": 0.22386979162693024,
      "num_tokens": 12724914.0,
      "step": 7000
    },
    {
      "epoch": 6.013751611516975,
      "eval_entropy": 5.456920374621142,
      "eval_loss": 5.958405494689941,
      "eval_mean_token_accuracy": 0.1763649323770592,
      "eval_num_tokens": 12724914.0,
      "eval_runtime": 2.828,
      "eval_samples_per_second": 1254.942,
      "eval_steps_per_second": 157.0,
      "step": 7000
    },
    {
      "entropy": 5.537625646591186,
      "epoch": 6.0180489901160295,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00022949984024866704,
      "loss": 4.8997,
      "mean_token_accuracy": 0.22314970791339875,
      "num_tokens": 12735193.0,
      "step": 7005
    },
    {
      "entropy": 5.6589820861816404,
      "epoch": 6.022346368715084,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002291742765396424,
      "loss": 4.9659,
      "mean_token_accuracy": 0.22305929362773896,
      "num_tokens": 12743945.0,
      "step": 7010
    },
    {
      "entropy": 5.658946895599366,
      "epoch": 6.0266437473141385,
      "grad_norm": 1.21875,
      "learning_rate": 0.00022884881289568133,
      "loss": 4.9701,
      "mean_token_accuracy": 0.21271953880786895,
      "num_tokens": 12753130.0,
      "step": 7015
    },
    {
      "entropy": 5.583167600631714,
      "epoch": 6.030941125913193,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002285234500274665,
      "loss": 4.9421,
      "mean_token_accuracy": 0.21728518307209016,
      "num_tokens": 12762108.0,
      "step": 7020
    },
    {
      "entropy": 5.665998697280884,
      "epoch": 6.035238504512248,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022819818864546016,
      "loss": 5.0104,
      "mean_token_accuracy": 0.21445933729410172,
      "num_tokens": 12772102.0,
      "step": 7025
    },
    {
      "entropy": 5.615815830230713,
      "epoch": 6.039535883111302,
      "grad_norm": 1.375,
      "learning_rate": 0.00022787302945990345,
      "loss": 4.9489,
      "mean_token_accuracy": 0.22382267862558364,
      "num_tokens": 12781225.0,
      "step": 7030
    },
    {
      "entropy": 5.592608308792114,
      "epoch": 6.043833261710357,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022754797318081383,
      "loss": 4.9183,
      "mean_token_accuracy": 0.21551852226257323,
      "num_tokens": 12789896.0,
      "step": 7035
    },
    {
      "entropy": 5.574016523361206,
      "epoch": 6.048130640309411,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022722302051798442,
      "loss": 4.8569,
      "mean_token_accuracy": 0.22924461960792542,
      "num_tokens": 12798596.0,
      "step": 7040
    },
    {
      "entropy": 5.528965425491333,
      "epoch": 6.052428018908466,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002268981721809819,
      "loss": 4.8766,
      "mean_token_accuracy": 0.23032936155796052,
      "num_tokens": 12807285.0,
      "step": 7045
    },
    {
      "entropy": 5.607598876953125,
      "epoch": 6.05672539750752,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002265734288791451,
      "loss": 4.9296,
      "mean_token_accuracy": 0.22038827538490297,
      "num_tokens": 12816668.0,
      "step": 7050
    },
    {
      "entropy": 5.66830005645752,
      "epoch": 6.061022776106575,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022624879132158377,
      "loss": 5.0283,
      "mean_token_accuracy": 0.21151769757270814,
      "num_tokens": 12825943.0,
      "step": 7055
    },
    {
      "entropy": 5.568456220626831,
      "epoch": 6.0653201547056295,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022592426021717654,
      "loss": 4.865,
      "mean_token_accuracy": 0.23040080517530442,
      "num_tokens": 12835693.0,
      "step": 7060
    },
    {
      "entropy": 5.552309465408325,
      "epoch": 6.069617533304684,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002255998362745696,
      "loss": 4.7413,
      "mean_token_accuracy": 0.24069842845201492,
      "num_tokens": 12844201.0,
      "step": 7065
    },
    {
      "entropy": 5.523254013061523,
      "epoch": 6.073914911903739,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022527552020217513,
      "loss": 4.8911,
      "mean_token_accuracy": 0.22590842396020888,
      "num_tokens": 12853220.0,
      "step": 7070
    },
    {
      "entropy": 5.576641654968261,
      "epoch": 6.078212290502793,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002249513127081697,
      "loss": 4.9783,
      "mean_token_accuracy": 0.2157495841383934,
      "num_tokens": 12862486.0,
      "step": 7075
    },
    {
      "entropy": 5.683221006393433,
      "epoch": 6.082509669101848,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022462721450049316,
      "loss": 4.9834,
      "mean_token_accuracy": 0.2175670623779297,
      "num_tokens": 12871717.0,
      "step": 7080
    },
    {
      "entropy": 5.598150157928467,
      "epoch": 6.086807047700902,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002243032262868464,
      "loss": 4.9312,
      "mean_token_accuracy": 0.2116045191884041,
      "num_tokens": 12881278.0,
      "step": 7085
    },
    {
      "entropy": 5.588177824020386,
      "epoch": 6.091104426299957,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022397934877469,
      "loss": 4.9688,
      "mean_token_accuracy": 0.21471360325813293,
      "num_tokens": 12890720.0,
      "step": 7090
    },
    {
      "entropy": 5.630496501922607,
      "epoch": 6.095401804899011,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002236555826712432,
      "loss": 4.9757,
      "mean_token_accuracy": 0.21892329156398774,
      "num_tokens": 12900428.0,
      "step": 7095
    },
    {
      "entropy": 5.633159637451172,
      "epoch": 6.099699183498066,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022333192868348152,
      "loss": 4.9778,
      "mean_token_accuracy": 0.21538899689912797,
      "num_tokens": 12910177.0,
      "step": 7100
    },
    {
      "entropy": 5.60188102722168,
      "epoch": 6.10399656209712,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022300838751813606,
      "loss": 4.9958,
      "mean_token_accuracy": 0.2148565888404846,
      "num_tokens": 12920734.0,
      "step": 7105
    },
    {
      "entropy": 5.598736047744751,
      "epoch": 6.108293940696175,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022268495988169145,
      "loss": 4.8858,
      "mean_token_accuracy": 0.2221657946705818,
      "num_tokens": 12929585.0,
      "step": 7110
    },
    {
      "entropy": 5.492549180984497,
      "epoch": 6.1125913192952295,
      "grad_norm": 1.375,
      "learning_rate": 0.00022236164648038433,
      "loss": 4.8845,
      "mean_token_accuracy": 0.2196088671684265,
      "num_tokens": 12938933.0,
      "step": 7115
    },
    {
      "entropy": 5.523064231872558,
      "epoch": 6.116888697894285,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002220384480202019,
      "loss": 4.8437,
      "mean_token_accuracy": 0.2295871540904045,
      "num_tokens": 12947461.0,
      "step": 7120
    },
    {
      "entropy": 5.550280618667602,
      "epoch": 6.1211860764933395,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022171536520688046,
      "loss": 4.9075,
      "mean_token_accuracy": 0.22688549011945724,
      "num_tokens": 12956507.0,
      "step": 7125
    },
    {
      "entropy": 5.646392297744751,
      "epoch": 6.125483455092394,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022139239874590362,
      "loss": 5.0502,
      "mean_token_accuracy": 0.21544915437698364,
      "num_tokens": 12965740.0,
      "step": 7130
    },
    {
      "entropy": 5.6248448371887205,
      "epoch": 6.1297808336914485,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002210695493425013,
      "loss": 4.955,
      "mean_token_accuracy": 0.21726755797863007,
      "num_tokens": 12975057.0,
      "step": 7135
    },
    {
      "entropy": 5.519511795043945,
      "epoch": 6.134078212290503,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022074681770164735,
      "loss": 4.8571,
      "mean_token_accuracy": 0.22992920875549316,
      "num_tokens": 12984087.0,
      "step": 7140
    },
    {
      "entropy": 5.49357647895813,
      "epoch": 6.138375590889558,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022042420452805868,
      "loss": 4.874,
      "mean_token_accuracy": 0.23499309569597243,
      "num_tokens": 12992793.0,
      "step": 7145
    },
    {
      "entropy": 5.62636170387268,
      "epoch": 6.142672969488612,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022010171052619365,
      "loss": 4.9585,
      "mean_token_accuracy": 0.22705546170473098,
      "num_tokens": 13000769.0,
      "step": 7150
    },
    {
      "entropy": 5.635001277923584,
      "epoch": 6.146970348087667,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021977933640025,
      "loss": 4.9577,
      "mean_token_accuracy": 0.2211778700351715,
      "num_tokens": 13010677.0,
      "step": 7155
    },
    {
      "entropy": 5.5190966606140135,
      "epoch": 6.151267726686721,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021945708285416434,
      "loss": 4.8064,
      "mean_token_accuracy": 0.23988419026136398,
      "num_tokens": 13019791.0,
      "step": 7160
    },
    {
      "entropy": 5.549937391281128,
      "epoch": 6.155565105285776,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002191349505916093,
      "loss": 4.965,
      "mean_token_accuracy": 0.22050906866788864,
      "num_tokens": 13029223.0,
      "step": 7165
    },
    {
      "entropy": 5.578757095336914,
      "epoch": 6.15986248388483,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021881294031599318,
      "loss": 4.9295,
      "mean_token_accuracy": 0.21950550824403764,
      "num_tokens": 13038716.0,
      "step": 7170
    },
    {
      "entropy": 5.569738054275513,
      "epoch": 6.164159862483885,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002184910527304576,
      "loss": 4.9593,
      "mean_token_accuracy": 0.2233647584915161,
      "num_tokens": 13047915.0,
      "step": 7175
    },
    {
      "entropy": 5.532005834579468,
      "epoch": 6.1684572410829395,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021816928853787636,
      "loss": 4.906,
      "mean_token_accuracy": 0.22901789844036102,
      "num_tokens": 13056613.0,
      "step": 7180
    },
    {
      "entropy": 5.574031066894531,
      "epoch": 6.172754619681994,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021784764844085398,
      "loss": 4.9671,
      "mean_token_accuracy": 0.22078002840280533,
      "num_tokens": 13066658.0,
      "step": 7185
    },
    {
      "entropy": 5.698163461685181,
      "epoch": 6.177051998281049,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002175261331417238,
      "loss": 4.9714,
      "mean_token_accuracy": 0.21732018291950225,
      "num_tokens": 13074798.0,
      "step": 7190
    },
    {
      "entropy": 5.5674072265625,
      "epoch": 6.181349376880103,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021720474334254675,
      "loss": 4.8932,
      "mean_token_accuracy": 0.22398940920829774,
      "num_tokens": 13084173.0,
      "step": 7195
    },
    {
      "entropy": 5.566124963760376,
      "epoch": 6.185646755479158,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021688347974510962,
      "loss": 4.8925,
      "mean_token_accuracy": 0.2297670528292656,
      "num_tokens": 13093096.0,
      "step": 7200
    },
    {
      "entropy": 5.543036317825317,
      "epoch": 6.189944134078212,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021656234305092377,
      "loss": 4.9285,
      "mean_token_accuracy": 0.22698211967945098,
      "num_tokens": 13101191.0,
      "step": 7205
    },
    {
      "entropy": 5.583077335357666,
      "epoch": 6.194241512677267,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002162413339612234,
      "loss": 4.9453,
      "mean_token_accuracy": 0.22232597768306733,
      "num_tokens": 13109829.0,
      "step": 7210
    },
    {
      "entropy": 5.6000134468078615,
      "epoch": 6.198538891276321,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021592045317696406,
      "loss": 4.9089,
      "mean_token_accuracy": 0.22355552911758422,
      "num_tokens": 13119314.0,
      "step": 7215
    },
    {
      "entropy": 5.556233024597168,
      "epoch": 6.202836269875376,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021559970139882102,
      "loss": 4.8781,
      "mean_token_accuracy": 0.22706296145915986,
      "num_tokens": 13128113.0,
      "step": 7220
    },
    {
      "entropy": 5.621548748016357,
      "epoch": 6.20713364847443,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002152790793271881,
      "loss": 5.0134,
      "mean_token_accuracy": 0.21199150681495665,
      "num_tokens": 13136892.0,
      "step": 7225
    },
    {
      "entropy": 5.582968902587891,
      "epoch": 6.211431027073485,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021495858766217558,
      "loss": 4.9612,
      "mean_token_accuracy": 0.2196844384074211,
      "num_tokens": 13146960.0,
      "step": 7230
    },
    {
      "entropy": 5.56456995010376,
      "epoch": 6.2157284056725395,
      "grad_norm": 1.375,
      "learning_rate": 0.00021463822710360932,
      "loss": 4.8536,
      "mean_token_accuracy": 0.2300641804933548,
      "num_tokens": 13156147.0,
      "step": 7235
    },
    {
      "entropy": 5.591127204895019,
      "epoch": 6.220025784271594,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021431799835102867,
      "loss": 4.9456,
      "mean_token_accuracy": 0.20903619527816772,
      "num_tokens": 13164588.0,
      "step": 7240
    },
    {
      "entropy": 5.530172872543335,
      "epoch": 6.224323162870649,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021399790210368524,
      "loss": 4.8726,
      "mean_token_accuracy": 0.22854987680912017,
      "num_tokens": 13174361.0,
      "step": 7245
    },
    {
      "entropy": 5.604793977737427,
      "epoch": 6.228620541469703,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021367793906054133,
      "loss": 5.0833,
      "mean_token_accuracy": 0.21208511143922806,
      "num_tokens": 13185266.0,
      "step": 7250
    },
    {
      "entropy": 5.5820169925689695,
      "epoch": 6.232917920068758,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021335810992026823,
      "loss": 4.9497,
      "mean_token_accuracy": 0.23028518855571747,
      "num_tokens": 13194227.0,
      "step": 7255
    },
    {
      "entropy": 5.637402534484863,
      "epoch": 6.237215298667812,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021303841538124497,
      "loss": 4.9865,
      "mean_token_accuracy": 0.21631672829389573,
      "num_tokens": 13202569.0,
      "step": 7260
    },
    {
      "entropy": 5.564388465881348,
      "epoch": 6.241512677266867,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021271885614155685,
      "loss": 4.9241,
      "mean_token_accuracy": 0.22911797165870668,
      "num_tokens": 13212201.0,
      "step": 7265
    },
    {
      "entropy": 5.508797883987427,
      "epoch": 6.245810055865922,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002123994328989932,
      "loss": 4.8316,
      "mean_token_accuracy": 0.22354399114847184,
      "num_tokens": 13220802.0,
      "step": 7270
    },
    {
      "entropy": 5.637790727615356,
      "epoch": 6.250107434464977,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021208014635104688,
      "loss": 5.0074,
      "mean_token_accuracy": 0.21484729200601577,
      "num_tokens": 13229519.0,
      "step": 7275
    },
    {
      "entropy": 5.6006574630737305,
      "epoch": 6.254404813064031,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021176099719491209,
      "loss": 4.9537,
      "mean_token_accuracy": 0.22258995175361634,
      "num_tokens": 13238865.0,
      "step": 7280
    },
    {
      "entropy": 5.47272834777832,
      "epoch": 6.258702191663086,
      "grad_norm": 1.59375,
      "learning_rate": 0.00021144198612748312,
      "loss": 4.8686,
      "mean_token_accuracy": 0.22965049594640732,
      "num_tokens": 13247259.0,
      "step": 7285
    },
    {
      "entropy": 5.556898069381714,
      "epoch": 6.26299957026214,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021112311384535243,
      "loss": 4.9644,
      "mean_token_accuracy": 0.22257940024137496,
      "num_tokens": 13256692.0,
      "step": 7290
    },
    {
      "entropy": 5.616722822189331,
      "epoch": 6.267296948861195,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021080438104480976,
      "loss": 4.9196,
      "mean_token_accuracy": 0.2185341089963913,
      "num_tokens": 13266109.0,
      "step": 7295
    },
    {
      "entropy": 5.6341911315917965,
      "epoch": 6.2715943274602495,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021048578842184019,
      "loss": 5.0294,
      "mean_token_accuracy": 0.20308708250522614,
      "num_tokens": 13275484.0,
      "step": 7300
    },
    {
      "entropy": 5.539116525650025,
      "epoch": 6.275891706059304,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021016733667212245,
      "loss": 4.9123,
      "mean_token_accuracy": 0.2227211132645607,
      "num_tokens": 13284755.0,
      "step": 7305
    },
    {
      "entropy": 5.551125621795654,
      "epoch": 6.280189084658359,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00020984902649102806,
      "loss": 4.8959,
      "mean_token_accuracy": 0.2237284243106842,
      "num_tokens": 13294386.0,
      "step": 7310
    },
    {
      "entropy": 5.577311325073242,
      "epoch": 6.284486463257413,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020953085857361924,
      "loss": 4.8964,
      "mean_token_accuracy": 0.2272123247385025,
      "num_tokens": 13303926.0,
      "step": 7315
    },
    {
      "entropy": 5.612677955627442,
      "epoch": 6.288783841856468,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020921283361464754,
      "loss": 4.9999,
      "mean_token_accuracy": 0.21563175171613694,
      "num_tokens": 13312727.0,
      "step": 7320
    },
    {
      "entropy": 5.5714458465576175,
      "epoch": 6.293081220455522,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020889495230855232,
      "loss": 4.9743,
      "mean_token_accuracy": 0.22164907157421113,
      "num_tokens": 13321706.0,
      "step": 7325
    },
    {
      "entropy": 5.514965009689331,
      "epoch": 6.297378599054577,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020857721534945923,
      "loss": 4.8834,
      "mean_token_accuracy": 0.22938343286514282,
      "num_tokens": 13330436.0,
      "step": 7330
    },
    {
      "entropy": 5.5984455108642575,
      "epoch": 6.301675977653631,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002082596234311789,
      "loss": 4.9995,
      "mean_token_accuracy": 0.2088770717382431,
      "num_tokens": 13339334.0,
      "step": 7335
    },
    {
      "entropy": 5.641850757598877,
      "epoch": 6.305973356252686,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002079421772472051,
      "loss": 5.0469,
      "mean_token_accuracy": 0.2071719855070114,
      "num_tokens": 13348969.0,
      "step": 7340
    },
    {
      "entropy": 5.607429504394531,
      "epoch": 6.31027073485174,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002076248774907134,
      "loss": 5.0031,
      "mean_token_accuracy": 0.21442738473415374,
      "num_tokens": 13358467.0,
      "step": 7345
    },
    {
      "entropy": 5.5636971950531,
      "epoch": 6.314568113450795,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020730772485455962,
      "loss": 4.8171,
      "mean_token_accuracy": 0.2342108443379402,
      "num_tokens": 13366413.0,
      "step": 7350
    },
    {
      "entropy": 5.592044115066528,
      "epoch": 6.3188654920498495,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002069907200312785,
      "loss": 5.0061,
      "mean_token_accuracy": 0.21233259737491608,
      "num_tokens": 13376620.0,
      "step": 7355
    },
    {
      "entropy": 5.596766614913941,
      "epoch": 6.323162870648904,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020667386371308162,
      "loss": 4.9785,
      "mean_token_accuracy": 0.2176347553730011,
      "num_tokens": 13385492.0,
      "step": 7360
    },
    {
      "entropy": 5.542520666122437,
      "epoch": 6.327460249247959,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020635715659185673,
      "loss": 4.924,
      "mean_token_accuracy": 0.22444240748882294,
      "num_tokens": 13395562.0,
      "step": 7365
    },
    {
      "entropy": 5.480905342102051,
      "epoch": 6.331757627847013,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020604059935916551,
      "loss": 4.8673,
      "mean_token_accuracy": 0.22702675610780715,
      "num_tokens": 13403357.0,
      "step": 7370
    },
    {
      "entropy": 5.5856396675109865,
      "epoch": 6.336055006446068,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020572419270624255,
      "loss": 4.9639,
      "mean_token_accuracy": 0.2227653369307518,
      "num_tokens": 13412527.0,
      "step": 7375
    },
    {
      "entropy": 5.632455062866211,
      "epoch": 6.340352385045122,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020540793732399339,
      "loss": 5.0336,
      "mean_token_accuracy": 0.211682990193367,
      "num_tokens": 13422455.0,
      "step": 7380
    },
    {
      "entropy": 5.698804330825806,
      "epoch": 6.344649763644177,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020509183390299325,
      "loss": 5.13,
      "mean_token_accuracy": 0.20494184494018555,
      "num_tokens": 13431677.0,
      "step": 7385
    },
    {
      "entropy": 5.520529699325562,
      "epoch": 6.348947142243231,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020477588313348594,
      "loss": 4.8386,
      "mean_token_accuracy": 0.2345274955034256,
      "num_tokens": 13440522.0,
      "step": 7390
    },
    {
      "entropy": 5.608442640304565,
      "epoch": 6.353244520842286,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020446008570538154,
      "loss": 4.9418,
      "mean_token_accuracy": 0.21886708736419677,
      "num_tokens": 13450021.0,
      "step": 7395
    },
    {
      "entropy": 5.553961277008057,
      "epoch": 6.35754189944134,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002041444423082554,
      "loss": 4.9047,
      "mean_token_accuracy": 0.2260735273361206,
      "num_tokens": 13458115.0,
      "step": 7400
    },
    {
      "entropy": 5.614691972732544,
      "epoch": 6.361839278040395,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020382895363134652,
      "loss": 4.9663,
      "mean_token_accuracy": 0.20537222772836686,
      "num_tokens": 13466798.0,
      "step": 7405
    },
    {
      "entropy": 5.528500699996949,
      "epoch": 6.3661366566394495,
      "grad_norm": 1.375,
      "learning_rate": 0.00020351362036355602,
      "loss": 4.9769,
      "mean_token_accuracy": 0.216089104115963,
      "num_tokens": 13476096.0,
      "step": 7410
    },
    {
      "entropy": 5.5570347785949705,
      "epoch": 6.370434035238505,
      "grad_norm": 1.375,
      "learning_rate": 0.0002031984431934459,
      "loss": 4.9291,
      "mean_token_accuracy": 0.22416253834962846,
      "num_tokens": 13484601.0,
      "step": 7415
    },
    {
      "entropy": 5.584917068481445,
      "epoch": 6.3747314138375595,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020288342280923695,
      "loss": 4.9017,
      "mean_token_accuracy": 0.2288140043616295,
      "num_tokens": 13493994.0,
      "step": 7420
    },
    {
      "entropy": 5.578990650177002,
      "epoch": 6.379028792436614,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020256855989880785,
      "loss": 4.9449,
      "mean_token_accuracy": 0.2124490797519684,
      "num_tokens": 13502890.0,
      "step": 7425
    },
    {
      "entropy": 5.514740657806397,
      "epoch": 6.383326171035669,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020225385514969336,
      "loss": 4.916,
      "mean_token_accuracy": 0.22774983197450638,
      "num_tokens": 13512980.0,
      "step": 7430
    },
    {
      "entropy": 5.584985589981079,
      "epoch": 6.387623549634723,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020193930924908277,
      "loss": 4.9992,
      "mean_token_accuracy": 0.21512799113988876,
      "num_tokens": 13521558.0,
      "step": 7435
    },
    {
      "entropy": 5.522138500213623,
      "epoch": 6.391920928233778,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020162492288381867,
      "loss": 4.8986,
      "mean_token_accuracy": 0.22678508907556533,
      "num_tokens": 13531506.0,
      "step": 7440
    },
    {
      "entropy": 5.514893913269043,
      "epoch": 6.396218306832832,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002013106967403953,
      "loss": 4.8725,
      "mean_token_accuracy": 0.23114074915647506,
      "num_tokens": 13540559.0,
      "step": 7445
    },
    {
      "entropy": 5.5967857360839846,
      "epoch": 6.400515685431887,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002009966315049569,
      "loss": 4.9512,
      "mean_token_accuracy": 0.2251010686159134,
      "num_tokens": 13550654.0,
      "step": 7450
    },
    {
      "entropy": 5.5890641689300535,
      "epoch": 6.404813064030941,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002006827278632964,
      "loss": 4.9807,
      "mean_token_accuracy": 0.2174374431371689,
      "num_tokens": 13560708.0,
      "step": 7455
    },
    {
      "entropy": 5.6201427459716795,
      "epoch": 6.409110442629996,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020036898650085377,
      "loss": 4.8741,
      "mean_token_accuracy": 0.22463560551404954,
      "num_tokens": 13569330.0,
      "step": 7460
    },
    {
      "entropy": 5.655684852600098,
      "epoch": 6.41340782122905,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020005540810271493,
      "loss": 5.0472,
      "mean_token_accuracy": 0.20860252827405928,
      "num_tokens": 13577500.0,
      "step": 7465
    },
    {
      "entropy": 5.537591648101807,
      "epoch": 6.417705199828105,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019974199335360976,
      "loss": 4.9529,
      "mean_token_accuracy": 0.21955338269472122,
      "num_tokens": 13586087.0,
      "step": 7470
    },
    {
      "entropy": 5.531689262390136,
      "epoch": 6.4220025784271595,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019942874293791068,
      "loss": 4.9095,
      "mean_token_accuracy": 0.21984111964702607,
      "num_tokens": 13595346.0,
      "step": 7475
    },
    {
      "entropy": 5.643185997009278,
      "epoch": 6.426299957026214,
      "grad_norm": 1.375,
      "learning_rate": 0.00019911565753963145,
      "loss": 5.051,
      "mean_token_accuracy": 0.20972351133823394,
      "num_tokens": 13604755.0,
      "step": 7480
    },
    {
      "entropy": 5.579855823516846,
      "epoch": 6.430597335625269,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001988027378424254,
      "loss": 4.9532,
      "mean_token_accuracy": 0.2187460482120514,
      "num_tokens": 13613860.0,
      "step": 7485
    },
    {
      "entropy": 5.483318185806274,
      "epoch": 6.434894714224323,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019848998452958429,
      "loss": 4.8477,
      "mean_token_accuracy": 0.23034254014492034,
      "num_tokens": 13622574.0,
      "step": 7490
    },
    {
      "entropy": 5.577697849273681,
      "epoch": 6.439192092823378,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019817739828403602,
      "loss": 4.9847,
      "mean_token_accuracy": 0.21383119523525237,
      "num_tokens": 13632366.0,
      "step": 7495
    },
    {
      "entropy": 5.65091462135315,
      "epoch": 6.443489471422432,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019786497978834422,
      "loss": 4.8778,
      "mean_token_accuracy": 0.23622201085090638,
      "num_tokens": 13640682.0,
      "step": 7500
    },
    {
      "epoch": 6.443489471422432,
      "eval_entropy": 5.446689291580303,
      "eval_loss": 5.9549760818481445,
      "eval_mean_token_accuracy": 0.17768076151139564,
      "eval_num_tokens": 13640682.0,
      "eval_runtime": 2.6462,
      "eval_samples_per_second": 1341.174,
      "eval_steps_per_second": 167.788,
      "step": 7500
    },
    {
      "entropy": 5.566683006286621,
      "epoch": 6.447786850021487,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019755272972470602,
      "loss": 4.948,
      "mean_token_accuracy": 0.2239464581012726,
      "num_tokens": 13649675.0,
      "step": 7505
    },
    {
      "entropy": 5.557274532318115,
      "epoch": 6.452084228620541,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019724064877495057,
      "loss": 4.9374,
      "mean_token_accuracy": 0.21715119183063508,
      "num_tokens": 13658260.0,
      "step": 7510
    },
    {
      "entropy": 5.540681171417236,
      "epoch": 6.456381607219596,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019692873762053808,
      "loss": 4.8747,
      "mean_token_accuracy": 0.22564099729061127,
      "num_tokens": 13666571.0,
      "step": 7515
    },
    {
      "entropy": 5.610409069061279,
      "epoch": 6.46067898581865,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00019661699694255785,
      "loss": 4.9655,
      "mean_token_accuracy": 0.22639194875955582,
      "num_tokens": 13675707.0,
      "step": 7520
    },
    {
      "entropy": 5.64504108428955,
      "epoch": 6.464976364417705,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019630542742172692,
      "loss": 4.9019,
      "mean_token_accuracy": 0.22488683760166167,
      "num_tokens": 13684796.0,
      "step": 7525
    },
    {
      "entropy": 5.592291450500488,
      "epoch": 6.4692737430167595,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019599402973838854,
      "loss": 5.0083,
      "mean_token_accuracy": 0.2147836297750473,
      "num_tokens": 13693158.0,
      "step": 7530
    },
    {
      "entropy": 5.609640884399414,
      "epoch": 6.473571121615814,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001956828045725107,
      "loss": 5.0057,
      "mean_token_accuracy": 0.21335594356060028,
      "num_tokens": 13703521.0,
      "step": 7535
    },
    {
      "entropy": 5.57060022354126,
      "epoch": 6.477868500214869,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001953717526036849,
      "loss": 4.89,
      "mean_token_accuracy": 0.22477730810642244,
      "num_tokens": 13712337.0,
      "step": 7540
    },
    {
      "entropy": 5.561862468719482,
      "epoch": 6.482165878813923,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019506087451112437,
      "loss": 4.9109,
      "mean_token_accuracy": 0.21911588609218596,
      "num_tokens": 13721605.0,
      "step": 7545
    },
    {
      "entropy": 5.513821172714233,
      "epoch": 6.486463257412978,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019475017097366244,
      "loss": 4.8259,
      "mean_token_accuracy": 0.2250768333673477,
      "num_tokens": 13730827.0,
      "step": 7550
    },
    {
      "entropy": 5.618396425247193,
      "epoch": 6.490760636012032,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019443964266975156,
      "loss": 4.9888,
      "mean_token_accuracy": 0.21844715625047684,
      "num_tokens": 13740128.0,
      "step": 7555
    },
    {
      "entropy": 5.531411123275757,
      "epoch": 6.495058014611088,
      "grad_norm": 1.375,
      "learning_rate": 0.0001941292902774614,
      "loss": 4.9641,
      "mean_token_accuracy": 0.21880121380090714,
      "num_tokens": 13748428.0,
      "step": 7560
    },
    {
      "entropy": 5.601159906387329,
      "epoch": 6.499355393210142,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019381911447447742,
      "loss": 4.9408,
      "mean_token_accuracy": 0.22451299130916597,
      "num_tokens": 13757109.0,
      "step": 7565
    },
    {
      "entropy": 5.61955771446228,
      "epoch": 6.503652771809197,
      "grad_norm": 1.5,
      "learning_rate": 0.00019350911593809977,
      "loss": 5.0011,
      "mean_token_accuracy": 0.21385532021522521,
      "num_tokens": 13766281.0,
      "step": 7570
    },
    {
      "entropy": 5.619399642944336,
      "epoch": 6.507950150408251,
      "grad_norm": 1.265625,
      "learning_rate": 0.00019319929534524128,
      "loss": 4.9854,
      "mean_token_accuracy": 0.21372777372598648,
      "num_tokens": 13775535.0,
      "step": 7575
    },
    {
      "entropy": 5.584274864196777,
      "epoch": 6.512247529007306,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019288965337242636,
      "loss": 4.954,
      "mean_token_accuracy": 0.20956738889217377,
      "num_tokens": 13784099.0,
      "step": 7580
    },
    {
      "entropy": 5.658680105209351,
      "epoch": 6.51654490760636,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019258019069578924,
      "loss": 5.0936,
      "mean_token_accuracy": 0.2100210815668106,
      "num_tokens": 13793098.0,
      "step": 7585
    },
    {
      "entropy": 5.5276800155639645,
      "epoch": 6.520842286205415,
      "grad_norm": 1.375,
      "learning_rate": 0.00019227090799107266,
      "loss": 4.94,
      "mean_token_accuracy": 0.21735497415065766,
      "num_tokens": 13801847.0,
      "step": 7590
    },
    {
      "entropy": 5.544484567642212,
      "epoch": 6.5251396648044695,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001919618059336265,
      "loss": 4.8072,
      "mean_token_accuracy": 0.23526784181594848,
      "num_tokens": 13810599.0,
      "step": 7595
    },
    {
      "entropy": 5.618886518478393,
      "epoch": 6.529437043403524,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019165288519840617,
      "loss": 4.9607,
      "mean_token_accuracy": 0.22045986652374266,
      "num_tokens": 13819602.0,
      "step": 7600
    },
    {
      "entropy": 5.478440809249878,
      "epoch": 6.533734422002579,
      "grad_norm": 1.453125,
      "learning_rate": 0.000191344146459971,
      "loss": 4.9092,
      "mean_token_accuracy": 0.22710810601711273,
      "num_tokens": 13828254.0,
      "step": 7605
    },
    {
      "entropy": 5.532521200180054,
      "epoch": 6.538031800601633,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019103559039248302,
      "loss": 4.9653,
      "mean_token_accuracy": 0.21937706470489501,
      "num_tokens": 13837163.0,
      "step": 7610
    },
    {
      "entropy": 5.647298908233642,
      "epoch": 6.542329179200688,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001907272176697052,
      "loss": 4.971,
      "mean_token_accuracy": 0.2165326565504074,
      "num_tokens": 13846373.0,
      "step": 7615
    },
    {
      "entropy": 5.521724987030029,
      "epoch": 6.546626557799742,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019041902896500059,
      "loss": 4.9036,
      "mean_token_accuracy": 0.2245998725295067,
      "num_tokens": 13855846.0,
      "step": 7620
    },
    {
      "entropy": 5.613715314865113,
      "epoch": 6.550923936398797,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019011102495132993,
      "loss": 4.9303,
      "mean_token_accuracy": 0.2205586016178131,
      "num_tokens": 13864723.0,
      "step": 7625
    },
    {
      "entropy": 5.671797466278076,
      "epoch": 6.555221314997851,
      "grad_norm": 1.15625,
      "learning_rate": 0.00018980320630125104,
      "loss": 5.0842,
      "mean_token_accuracy": 0.20742758363485336,
      "num_tokens": 13873418.0,
      "step": 7630
    },
    {
      "entropy": 5.63235764503479,
      "epoch": 6.559518693596906,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00018949557368691666,
      "loss": 4.9662,
      "mean_token_accuracy": 0.2206213355064392,
      "num_tokens": 13881890.0,
      "step": 7635
    },
    {
      "entropy": 5.55091700553894,
      "epoch": 6.56381607219596,
      "grad_norm": 1.25,
      "learning_rate": 0.00018918812778007343,
      "loss": 5.0013,
      "mean_token_accuracy": 0.21714837700128556,
      "num_tokens": 13891289.0,
      "step": 7640
    },
    {
      "entropy": 5.558005285263062,
      "epoch": 6.568113450795015,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018888086925206054,
      "loss": 4.945,
      "mean_token_accuracy": 0.2151249796152115,
      "num_tokens": 13900344.0,
      "step": 7645
    },
    {
      "entropy": 5.601388931274414,
      "epoch": 6.5724108293940695,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018857379877380763,
      "loss": 4.9525,
      "mean_token_accuracy": 0.22474519461393355,
      "num_tokens": 13909108.0,
      "step": 7650
    },
    {
      "entropy": 5.612611389160156,
      "epoch": 6.576708207993124,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018826691701583404,
      "loss": 5.0031,
      "mean_token_accuracy": 0.2208331659436226,
      "num_tokens": 13918458.0,
      "step": 7655
    },
    {
      "entropy": 5.602540016174316,
      "epoch": 6.581005586592179,
      "grad_norm": 1.375,
      "learning_rate": 0.00018796022464824663,
      "loss": 4.9433,
      "mean_token_accuracy": 0.22257139533758163,
      "num_tokens": 13927186.0,
      "step": 7660
    },
    {
      "entropy": 5.631735706329346,
      "epoch": 6.585302965191233,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018765372234073912,
      "loss": 5.0318,
      "mean_token_accuracy": 0.2073415160179138,
      "num_tokens": 13936701.0,
      "step": 7665
    },
    {
      "entropy": 5.582999992370605,
      "epoch": 6.589600343790288,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00018734741076259005,
      "loss": 5.0569,
      "mean_token_accuracy": 0.20306194722652435,
      "num_tokens": 13945842.0,
      "step": 7670
    },
    {
      "entropy": 5.534829044342041,
      "epoch": 6.593897722389342,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018704129058266152,
      "loss": 4.9236,
      "mean_token_accuracy": 0.22281349152326585,
      "num_tokens": 13955675.0,
      "step": 7675
    },
    {
      "entropy": 5.637769365310669,
      "epoch": 6.598195100988397,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018673536246939743,
      "loss": 5.0189,
      "mean_token_accuracy": 0.22031795233488083,
      "num_tokens": 13964153.0,
      "step": 7680
    },
    {
      "entropy": 5.5713536739349365,
      "epoch": 6.602492479587451,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018642962709082274,
      "loss": 4.9866,
      "mean_token_accuracy": 0.22005293369293213,
      "num_tokens": 13972141.0,
      "step": 7685
    },
    {
      "entropy": 5.525777196884155,
      "epoch": 6.606789858186506,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018612408511454103,
      "loss": 4.8543,
      "mean_token_accuracy": 0.2260838642716408,
      "num_tokens": 13981369.0,
      "step": 7690
    },
    {
      "entropy": 5.547107696533203,
      "epoch": 6.61108723678556,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018581873720773423,
      "loss": 4.9132,
      "mean_token_accuracy": 0.2208578556776047,
      "num_tokens": 13990626.0,
      "step": 7695
    },
    {
      "entropy": 5.584371280670166,
      "epoch": 6.615384615384615,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018551358403715989,
      "loss": 5.0066,
      "mean_token_accuracy": 0.21585517078638078,
      "num_tokens": 13999399.0,
      "step": 7700
    },
    {
      "entropy": 5.611315202713013,
      "epoch": 6.61968199398367,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00018520862626915052,
      "loss": 4.8941,
      "mean_token_accuracy": 0.2244178131222725,
      "num_tokens": 14007487.0,
      "step": 7705
    },
    {
      "entropy": 5.5730628490448,
      "epoch": 6.623979372582724,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018490386456961223,
      "loss": 4.8903,
      "mean_token_accuracy": 0.22212093621492385,
      "num_tokens": 14016779.0,
      "step": 7710
    },
    {
      "entropy": 5.611715698242188,
      "epoch": 6.6282767511817795,
      "grad_norm": 1.375,
      "learning_rate": 0.0001845992996040224,
      "loss": 4.9847,
      "mean_token_accuracy": 0.22296055108308793,
      "num_tokens": 14025770.0,
      "step": 7715
    },
    {
      "entropy": 5.539245176315307,
      "epoch": 6.632574129780834,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018429493203742946,
      "loss": 4.8723,
      "mean_token_accuracy": 0.2331751763820648,
      "num_tokens": 14035304.0,
      "step": 7720
    },
    {
      "entropy": 5.553162527084351,
      "epoch": 6.636871508379889,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018399076253445052,
      "loss": 4.8276,
      "mean_token_accuracy": 0.22845176011323928,
      "num_tokens": 14044058.0,
      "step": 7725
    },
    {
      "entropy": 5.557445621490478,
      "epoch": 6.641168886978943,
      "grad_norm": 1.375,
      "learning_rate": 0.00018368679175927012,
      "loss": 4.8708,
      "mean_token_accuracy": 0.23002058416604995,
      "num_tokens": 14052709.0,
      "step": 7730
    },
    {
      "entropy": 5.471322631835937,
      "epoch": 6.645466265577998,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018338302037563885,
      "loss": 4.8251,
      "mean_token_accuracy": 0.24188781529664993,
      "num_tokens": 14061346.0,
      "step": 7735
    },
    {
      "entropy": 5.566992998123169,
      "epoch": 6.649763644177052,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018307944904687211,
      "loss": 4.9477,
      "mean_token_accuracy": 0.2262141153216362,
      "num_tokens": 14070409.0,
      "step": 7740
    },
    {
      "entropy": 5.612448215484619,
      "epoch": 6.654061022776107,
      "grad_norm": 1.5,
      "learning_rate": 0.0001827760784358483,
      "loss": 4.9997,
      "mean_token_accuracy": 0.20946128517389298,
      "num_tokens": 14079448.0,
      "step": 7745
    },
    {
      "entropy": 5.614980792999267,
      "epoch": 6.658358401375161,
      "grad_norm": 1.375,
      "learning_rate": 0.00018247290920500776,
      "loss": 4.997,
      "mean_token_accuracy": 0.21683522015810014,
      "num_tokens": 14088452.0,
      "step": 7750
    },
    {
      "entropy": 5.57450008392334,
      "epoch": 6.662655779974216,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018216994201635062,
      "loss": 4.9557,
      "mean_token_accuracy": 0.22259806990623474,
      "num_tokens": 14097956.0,
      "step": 7755
    },
    {
      "entropy": 5.602868890762329,
      "epoch": 6.66695315857327,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018186717753143633,
      "loss": 4.9084,
      "mean_token_accuracy": 0.2201705664396286,
      "num_tokens": 14106771.0,
      "step": 7760
    },
    {
      "entropy": 5.600787591934204,
      "epoch": 6.671250537172325,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018156461641138133,
      "loss": 4.9848,
      "mean_token_accuracy": 0.213291634619236,
      "num_tokens": 14115871.0,
      "step": 7765
    },
    {
      "entropy": 5.565445756912231,
      "epoch": 6.6755479157713795,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018126225931685836,
      "loss": 4.9376,
      "mean_token_accuracy": 0.2180083692073822,
      "num_tokens": 14125316.0,
      "step": 7770
    },
    {
      "entropy": 5.504559707641602,
      "epoch": 6.679845294370434,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018096010690809444,
      "loss": 4.8611,
      "mean_token_accuracy": 0.22735559940338135,
      "num_tokens": 14134244.0,
      "step": 7775
    },
    {
      "entropy": 5.613789987564087,
      "epoch": 6.684142672969489,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018065815984486962,
      "loss": 5.0013,
      "mean_token_accuracy": 0.21289446651935579,
      "num_tokens": 14143600.0,
      "step": 7780
    },
    {
      "entropy": 5.5780164241790775,
      "epoch": 6.688440051568543,
      "grad_norm": 1.375,
      "learning_rate": 0.00018035641878651548,
      "loss": 4.9002,
      "mean_token_accuracy": 0.22572480142116547,
      "num_tokens": 14152112.0,
      "step": 7785
    },
    {
      "entropy": 5.549471139907837,
      "epoch": 6.692737430167598,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018005488439191408,
      "loss": 4.8813,
      "mean_token_accuracy": 0.22656293213367462,
      "num_tokens": 14161044.0,
      "step": 7790
    },
    {
      "entropy": 5.585894823074341,
      "epoch": 6.697034808766652,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001797535573194959,
      "loss": 4.9898,
      "mean_token_accuracy": 0.22107121497392654,
      "num_tokens": 14169629.0,
      "step": 7795
    },
    {
      "entropy": 5.5539881706237795,
      "epoch": 6.701332187365707,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001794524382272389,
      "loss": 4.9276,
      "mean_token_accuracy": 0.22021517902612686,
      "num_tokens": 14179234.0,
      "step": 7800
    },
    {
      "entropy": 5.56066746711731,
      "epoch": 6.705629565964761,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001791515277726667,
      "loss": 4.9668,
      "mean_token_accuracy": 0.22331743389368058,
      "num_tokens": 14188887.0,
      "step": 7805
    },
    {
      "entropy": 5.515817260742187,
      "epoch": 6.709926944563816,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017885082661284763,
      "loss": 4.8574,
      "mean_token_accuracy": 0.2264365389943123,
      "num_tokens": 14197731.0,
      "step": 7810
    },
    {
      "entropy": 5.58310832977295,
      "epoch": 6.71422432316287,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00017855033540439274,
      "loss": 5.0187,
      "mean_token_accuracy": 0.21447960138320923,
      "num_tokens": 14206851.0,
      "step": 7815
    },
    {
      "entropy": 5.58062252998352,
      "epoch": 6.718521701761925,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017825005480345463,
      "loss": 5.0159,
      "mean_token_accuracy": 0.21562800109386443,
      "num_tokens": 14215879.0,
      "step": 7820
    },
    {
      "entropy": 5.580096578598022,
      "epoch": 6.7228190803609795,
      "grad_norm": 1.46875,
      "learning_rate": 0.00017794998546572627,
      "loss": 4.8388,
      "mean_token_accuracy": 0.23093472719192504,
      "num_tokens": 14223798.0,
      "step": 7825
    },
    {
      "entropy": 5.59050555229187,
      "epoch": 6.727116458960034,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001776501280464391,
      "loss": 4.9448,
      "mean_token_accuracy": 0.22316467016935349,
      "num_tokens": 14233234.0,
      "step": 7830
    },
    {
      "entropy": 5.574538469314575,
      "epoch": 6.731413837559089,
      "grad_norm": 1.5,
      "learning_rate": 0.00017735048320036197,
      "loss": 4.8972,
      "mean_token_accuracy": 0.2224713072180748,
      "num_tokens": 14241851.0,
      "step": 7835
    },
    {
      "entropy": 5.587837409973145,
      "epoch": 6.735711216158143,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017705105158179917,
      "loss": 5.0249,
      "mean_token_accuracy": 0.21009934544563294,
      "num_tokens": 14251578.0,
      "step": 7840
    },
    {
      "entropy": 5.610586404800415,
      "epoch": 6.740008594757198,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017675183384458987,
      "loss": 5.042,
      "mean_token_accuracy": 0.2101135104894638,
      "num_tokens": 14261122.0,
      "step": 7845
    },
    {
      "entropy": 5.620992374420166,
      "epoch": 6.744305973356253,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017645283064210616,
      "loss": 4.912,
      "mean_token_accuracy": 0.2277627781033516,
      "num_tokens": 14270594.0,
      "step": 7850
    },
    {
      "entropy": 5.5979420185089115,
      "epoch": 6.748603351955307,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017615404262725132,
      "loss": 4.97,
      "mean_token_accuracy": 0.21964607685804366,
      "num_tokens": 14279646.0,
      "step": 7855
    },
    {
      "entropy": 5.556132078170776,
      "epoch": 6.752900730554362,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017585547045245885,
      "loss": 4.8996,
      "mean_token_accuracy": 0.22634491622447966,
      "num_tokens": 14288555.0,
      "step": 7860
    },
    {
      "entropy": 5.548083066940308,
      "epoch": 6.757198109153417,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017555711476969138,
      "loss": 4.9449,
      "mean_token_accuracy": 0.21816781014204026,
      "num_tokens": 14297813.0,
      "step": 7865
    },
    {
      "entropy": 5.594791221618652,
      "epoch": 6.761495487752471,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017525897623043806,
      "loss": 4.9315,
      "mean_token_accuracy": 0.22951747477054596,
      "num_tokens": 14305779.0,
      "step": 7870
    },
    {
      "entropy": 5.6130509853363035,
      "epoch": 6.765792866351526,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017496105548571472,
      "loss": 4.9955,
      "mean_token_accuracy": 0.21068995743989943,
      "num_tokens": 14314419.0,
      "step": 7875
    },
    {
      "entropy": 5.57996678352356,
      "epoch": 6.77009024495058,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017466335318606086,
      "loss": 4.9597,
      "mean_token_accuracy": 0.2159273162484169,
      "num_tokens": 14322959.0,
      "step": 7880
    },
    {
      "entropy": 5.555951738357544,
      "epoch": 6.774387623549635,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017436586998153947,
      "loss": 4.9153,
      "mean_token_accuracy": 0.22613337635993958,
      "num_tokens": 14332270.0,
      "step": 7885
    },
    {
      "entropy": 5.536405754089356,
      "epoch": 6.7786850021486895,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017406860652173495,
      "loss": 4.9092,
      "mean_token_accuracy": 0.22230609953403474,
      "num_tokens": 14341701.0,
      "step": 7890
    },
    {
      "entropy": 5.6402346134185795,
      "epoch": 6.782982380747744,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017377156345575176,
      "loss": 4.9557,
      "mean_token_accuracy": 0.21746353954076766,
      "num_tokens": 14349551.0,
      "step": 7895
    },
    {
      "entropy": 5.507066106796264,
      "epoch": 6.787279759346799,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017347474143221338,
      "loss": 4.8529,
      "mean_token_accuracy": 0.23069808334112168,
      "num_tokens": 14358577.0,
      "step": 7900
    },
    {
      "entropy": 5.644873476028442,
      "epoch": 6.791577137945853,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017317814109926044,
      "loss": 4.9688,
      "mean_token_accuracy": 0.21770287454128265,
      "num_tokens": 14367862.0,
      "step": 7905
    },
    {
      "entropy": 5.472003364562989,
      "epoch": 6.795874516544908,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001728817631045495,
      "loss": 4.9529,
      "mean_token_accuracy": 0.22002813071012498,
      "num_tokens": 14377414.0,
      "step": 7910
    },
    {
      "entropy": 5.461136341094971,
      "epoch": 6.800171895143962,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001725856080952516,
      "loss": 4.8399,
      "mean_token_accuracy": 0.24165070056915283,
      "num_tokens": 14387239.0,
      "step": 7915
    },
    {
      "entropy": 5.504508304595947,
      "epoch": 6.804469273743017,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001722896767180509,
      "loss": 4.8715,
      "mean_token_accuracy": 0.23467917740345,
      "num_tokens": 14396076.0,
      "step": 7920
    },
    {
      "entropy": 5.546143865585327,
      "epoch": 6.808766652342071,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017199396961914334,
      "loss": 4.9522,
      "mean_token_accuracy": 0.22052715718746185,
      "num_tokens": 14404982.0,
      "step": 7925
    },
    {
      "entropy": 5.540010976791382,
      "epoch": 6.813064030941126,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017169848744423506,
      "loss": 4.8983,
      "mean_token_accuracy": 0.21933743357658386,
      "num_tokens": 14413364.0,
      "step": 7930
    },
    {
      "entropy": 5.597727537155151,
      "epoch": 6.81736140954018,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00017140323083854076,
      "loss": 5.0317,
      "mean_token_accuracy": 0.21132098138332367,
      "num_tokens": 14424279.0,
      "step": 7935
    },
    {
      "entropy": 5.5751330852508545,
      "epoch": 6.821658788139235,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017110820044678317,
      "loss": 5.0044,
      "mean_token_accuracy": 0.21433879137039186,
      "num_tokens": 14432931.0,
      "step": 7940
    },
    {
      "entropy": 5.57401671409607,
      "epoch": 6.8259561667382895,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017081339691319054,
      "loss": 4.9311,
      "mean_token_accuracy": 0.22126074880361557,
      "num_tokens": 14442652.0,
      "step": 7945
    },
    {
      "entropy": 5.5973313331604,
      "epoch": 6.830253545337344,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017051882088149612,
      "loss": 4.9108,
      "mean_token_accuracy": 0.21747010946273804,
      "num_tokens": 14452061.0,
      "step": 7950
    },
    {
      "entropy": 5.581775331497193,
      "epoch": 6.834550923936399,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017022447299493599,
      "loss": 4.906,
      "mean_token_accuracy": 0.223048897087574,
      "num_tokens": 14460771.0,
      "step": 7955
    },
    {
      "entropy": 5.626034927368164,
      "epoch": 6.838848302535453,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016993035389624854,
      "loss": 5.0328,
      "mean_token_accuracy": 0.20910212397575378,
      "num_tokens": 14469983.0,
      "step": 7960
    },
    {
      "entropy": 5.567556667327881,
      "epoch": 6.843145681134508,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001696364642276722,
      "loss": 4.9325,
      "mean_token_accuracy": 0.22394620031118392,
      "num_tokens": 14478641.0,
      "step": 7965
    },
    {
      "entropy": 5.570947504043579,
      "epoch": 6.847443059733562,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016934280463094448,
      "loss": 4.9957,
      "mean_token_accuracy": 0.2159895658493042,
      "num_tokens": 14487922.0,
      "step": 7970
    },
    {
      "entropy": 5.503707027435302,
      "epoch": 6.851740438332617,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00016904937574730062,
      "loss": 4.8375,
      "mean_token_accuracy": 0.23160989582538605,
      "num_tokens": 14496259.0,
      "step": 7975
    },
    {
      "entropy": 5.588885354995727,
      "epoch": 6.856037816931671,
      "grad_norm": 1.5,
      "learning_rate": 0.00016875617821747208,
      "loss": 4.9514,
      "mean_token_accuracy": 0.2178613469004631,
      "num_tokens": 14504366.0,
      "step": 7980
    },
    {
      "entropy": 5.528595113754273,
      "epoch": 6.860335195530726,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00016846321268168508,
      "loss": 4.9789,
      "mean_token_accuracy": 0.21555833965539933,
      "num_tokens": 14513561.0,
      "step": 7985
    },
    {
      "entropy": 5.556788539886474,
      "epoch": 6.8646325741297805,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016817047977965905,
      "loss": 4.9266,
      "mean_token_accuracy": 0.2233048528432846,
      "num_tokens": 14522531.0,
      "step": 7990
    },
    {
      "entropy": 5.603210020065307,
      "epoch": 6.868929952728836,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001678779801506058,
      "loss": 4.9617,
      "mean_token_accuracy": 0.21128788888454436,
      "num_tokens": 14531375.0,
      "step": 7995
    },
    {
      "entropy": 5.5853945255279545,
      "epoch": 6.8732273313278895,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016758571443322774,
      "loss": 4.9628,
      "mean_token_accuracy": 0.21971309632062913,
      "num_tokens": 14541081.0,
      "step": 8000
    },
    {
      "epoch": 6.8732273313278895,
      "eval_entropy": 5.467374265730918,
      "eval_loss": 5.934433937072754,
      "eval_mean_token_accuracy": 0.17857106053480157,
      "eval_num_tokens": 14541081.0,
      "eval_runtime": 2.6439,
      "eval_samples_per_second": 1342.357,
      "eval_steps_per_second": 167.936,
      "step": 8000
    },
    {
      "entropy": 5.52855167388916,
      "epoch": 6.877524709926945,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001672936832657162,
      "loss": 4.8854,
      "mean_token_accuracy": 0.2273966148495674,
      "num_tokens": 14550129.0,
      "step": 8005
    },
    {
      "entropy": 5.6391089916229244,
      "epoch": 6.8818220885259995,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016700188728575047,
      "loss": 5.0577,
      "mean_token_accuracy": 0.21500970125198365,
      "num_tokens": 14559051.0,
      "step": 8010
    },
    {
      "entropy": 5.5877820491790775,
      "epoch": 6.886119467125054,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016671032713049655,
      "loss": 4.9632,
      "mean_token_accuracy": 0.2292462855577469,
      "num_tokens": 14567719.0,
      "step": 8015
    },
    {
      "entropy": 5.611098527908325,
      "epoch": 6.890416845724109,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016641900343660515,
      "loss": 4.989,
      "mean_token_accuracy": 0.2201634094119072,
      "num_tokens": 14576256.0,
      "step": 8020
    },
    {
      "entropy": 5.573671865463257,
      "epoch": 6.894714224323163,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001661279168402107,
      "loss": 4.9822,
      "mean_token_accuracy": 0.21778583973646165,
      "num_tokens": 14586392.0,
      "step": 8025
    },
    {
      "entropy": 5.603881740570069,
      "epoch": 6.899011602922218,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016583706797693008,
      "loss": 4.9743,
      "mean_token_accuracy": 0.21982818096876144,
      "num_tokens": 14595448.0,
      "step": 8030
    },
    {
      "entropy": 5.626660442352295,
      "epoch": 6.903308981521272,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016554645748186105,
      "loss": 4.9973,
      "mean_token_accuracy": 0.22556039690971375,
      "num_tokens": 14604242.0,
      "step": 8035
    },
    {
      "entropy": 5.652194213867188,
      "epoch": 6.907606360120327,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016525608598958063,
      "loss": 5.0721,
      "mean_token_accuracy": 0.2080199047923088,
      "num_tokens": 14614983.0,
      "step": 8040
    },
    {
      "entropy": 5.570321369171142,
      "epoch": 6.911903738719381,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00016496595413414421,
      "loss": 4.9338,
      "mean_token_accuracy": 0.2235411897301674,
      "num_tokens": 14624748.0,
      "step": 8045
    },
    {
      "entropy": 5.548950624465943,
      "epoch": 6.916201117318436,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016467606254908355,
      "loss": 4.9047,
      "mean_token_accuracy": 0.22795995622873305,
      "num_tokens": 14633642.0,
      "step": 8050
    },
    {
      "entropy": 5.60926570892334,
      "epoch": 6.9204984959174904,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016438641186740632,
      "loss": 4.9815,
      "mean_token_accuracy": 0.2164205640554428,
      "num_tokens": 14642549.0,
      "step": 8055
    },
    {
      "entropy": 5.5641517162323,
      "epoch": 6.924795874516545,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016409700272159371,
      "loss": 4.9761,
      "mean_token_accuracy": 0.21618568599224092,
      "num_tokens": 14651642.0,
      "step": 8060
    },
    {
      "entropy": 5.613296985626221,
      "epoch": 6.9290932531155995,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016380783574359957,
      "loss": 4.9451,
      "mean_token_accuracy": 0.22404495775699615,
      "num_tokens": 14661052.0,
      "step": 8065
    },
    {
      "entropy": 5.579953289031982,
      "epoch": 6.933390631714654,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001635189115648491,
      "loss": 4.9601,
      "mean_token_accuracy": 0.21903859823942184,
      "num_tokens": 14670292.0,
      "step": 8070
    },
    {
      "entropy": 5.492034435272217,
      "epoch": 6.937688010313709,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016323023081623705,
      "loss": 4.8717,
      "mean_token_accuracy": 0.2316209077835083,
      "num_tokens": 14679735.0,
      "step": 8075
    },
    {
      "entropy": 5.597985410690308,
      "epoch": 6.941985388912763,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016294179412812702,
      "loss": 4.9985,
      "mean_token_accuracy": 0.2139566257596016,
      "num_tokens": 14688710.0,
      "step": 8080
    },
    {
      "entropy": 5.6177679061889645,
      "epoch": 6.946282767511818,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016265360213034923,
      "loss": 5.0674,
      "mean_token_accuracy": 0.20269976258277894,
      "num_tokens": 14698523.0,
      "step": 8085
    },
    {
      "entropy": 5.5545045852661135,
      "epoch": 6.950580146110872,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016236565545220007,
      "loss": 4.951,
      "mean_token_accuracy": 0.22034550011157988,
      "num_tokens": 14707674.0,
      "step": 8090
    },
    {
      "entropy": 5.561030626296997,
      "epoch": 6.954877524709927,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016207795472243975,
      "loss": 4.8655,
      "mean_token_accuracy": 0.23134921193122865,
      "num_tokens": 14716600.0,
      "step": 8095
    },
    {
      "entropy": 5.631087779998779,
      "epoch": 6.959174903308981,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00016179050056929173,
      "loss": 5.0394,
      "mean_token_accuracy": 0.2120370462536812,
      "num_tokens": 14726112.0,
      "step": 8100
    },
    {
      "entropy": 5.568548107147217,
      "epoch": 6.963472281908036,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00016150329362044102,
      "loss": 4.9299,
      "mean_token_accuracy": 0.22501958459615706,
      "num_tokens": 14735126.0,
      "step": 8105
    },
    {
      "entropy": 5.5859311580657955,
      "epoch": 6.9677696605070905,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016121633450303285,
      "loss": 4.9333,
      "mean_token_accuracy": 0.2186357468366623,
      "num_tokens": 14744346.0,
      "step": 8110
    },
    {
      "entropy": 5.684606218338013,
      "epoch": 6.972067039106145,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016092962384367122,
      "loss": 5.0365,
      "mean_token_accuracy": 0.22210466116666794,
      "num_tokens": 14753322.0,
      "step": 8115
    },
    {
      "entropy": 5.544355058670044,
      "epoch": 6.9763644177051995,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001606431622684176,
      "loss": 4.9604,
      "mean_token_accuracy": 0.22419198900461196,
      "num_tokens": 14762384.0,
      "step": 8120
    },
    {
      "entropy": 5.527321624755859,
      "epoch": 6.980661796304254,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016035695040278935,
      "loss": 4.9839,
      "mean_token_accuracy": 0.2183494120836258,
      "num_tokens": 14771451.0,
      "step": 8125
    },
    {
      "entropy": 5.606777143478394,
      "epoch": 6.984959174903309,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016007098887175914,
      "loss": 5.0041,
      "mean_token_accuracy": 0.20941489487886428,
      "num_tokens": 14780662.0,
      "step": 8130
    },
    {
      "entropy": 5.611530685424805,
      "epoch": 6.989256553502363,
      "grad_norm": 1.46875,
      "learning_rate": 0.00015978527829975254,
      "loss": 4.9109,
      "mean_token_accuracy": 0.22887244820594788,
      "num_tokens": 14789201.0,
      "step": 8135
    },
    {
      "entropy": 5.561617708206176,
      "epoch": 6.993553932101419,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015949981931064714,
      "loss": 4.9619,
      "mean_token_accuracy": 0.22160696536302565,
      "num_tokens": 14797857.0,
      "step": 8140
    },
    {
      "entropy": 5.5851739883422855,
      "epoch": 6.997851310700472,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001592146125277714,
      "loss": 4.9434,
      "mean_token_accuracy": 0.21917157471179963,
      "num_tokens": 14806271.0,
      "step": 8145
    },
    {
      "entropy": 5.619498835669623,
      "epoch": 7.001718951439622,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00015892965857390278,
      "loss": 5.0305,
      "mean_token_accuracy": 0.2121889723671807,
      "num_tokens": 14815568.0,
      "step": 8150
    },
    {
      "entropy": 5.6152424812316895,
      "epoch": 7.006016330038676,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015864495807126704,
      "loss": 4.8755,
      "mean_token_accuracy": 0.23051389455795288,
      "num_tokens": 14825140.0,
      "step": 8155
    },
    {
      "entropy": 5.638502931594848,
      "epoch": 7.010313708637731,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015836051164153602,
      "loss": 4.887,
      "mean_token_accuracy": 0.22762184143066405,
      "num_tokens": 14834459.0,
      "step": 8160
    },
    {
      "entropy": 5.640847396850586,
      "epoch": 7.014611087236785,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015807631990582733,
      "loss": 4.8458,
      "mean_token_accuracy": 0.22552157193422318,
      "num_tokens": 14843632.0,
      "step": 8165
    },
    {
      "entropy": 5.6428046226501465,
      "epoch": 7.01890846583584,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015779238348470192,
      "loss": 4.818,
      "mean_token_accuracy": 0.23315514922142028,
      "num_tokens": 14852626.0,
      "step": 8170
    },
    {
      "entropy": 5.506103467941284,
      "epoch": 7.023205844434894,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00015750870299816345,
      "loss": 4.7804,
      "mean_token_accuracy": 0.22414395809173585,
      "num_tokens": 14861571.0,
      "step": 8175
    },
    {
      "entropy": 5.548079252243042,
      "epoch": 7.027503223033949,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015722527906565672,
      "loss": 4.7457,
      "mean_token_accuracy": 0.24087063372135162,
      "num_tokens": 14870383.0,
      "step": 8180
    },
    {
      "entropy": 5.6116972923278805,
      "epoch": 7.0318006016330035,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015694211230606647,
      "loss": 4.8542,
      "mean_token_accuracy": 0.22502037584781648,
      "num_tokens": 14880212.0,
      "step": 8185
    },
    {
      "entropy": 5.554553699493408,
      "epoch": 7.036097980232058,
      "grad_norm": 1.375,
      "learning_rate": 0.00015665920333771564,
      "loss": 4.8081,
      "mean_token_accuracy": 0.2318879932165146,
      "num_tokens": 14889347.0,
      "step": 8190
    },
    {
      "entropy": 5.602101850509643,
      "epoch": 7.0403953588311134,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015637655277836427,
      "loss": 4.8424,
      "mean_token_accuracy": 0.22176593989133836,
      "num_tokens": 14898553.0,
      "step": 8195
    },
    {
      "entropy": 5.60179295539856,
      "epoch": 7.044692737430168,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001560941612452081,
      "loss": 4.835,
      "mean_token_accuracy": 0.2274198442697525,
      "num_tokens": 14907275.0,
      "step": 8200
    },
    {
      "entropy": 5.593973827362061,
      "epoch": 7.0489901160292225,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001558120293548777,
      "loss": 4.8546,
      "mean_token_accuracy": 0.22675100713968277,
      "num_tokens": 14916409.0,
      "step": 8205
    },
    {
      "entropy": 5.634195184707641,
      "epoch": 7.053287494628277,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015553015772343614,
      "loss": 4.9222,
      "mean_token_accuracy": 0.21540625840425492,
      "num_tokens": 14927144.0,
      "step": 8210
    },
    {
      "entropy": 5.532114934921265,
      "epoch": 7.057584873227332,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015524854696637847,
      "loss": 4.7644,
      "mean_token_accuracy": 0.23972903341054916,
      "num_tokens": 14936310.0,
      "step": 8215
    },
    {
      "entropy": 5.526848554611206,
      "epoch": 7.061882251826386,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015496719769862981,
      "loss": 4.7624,
      "mean_token_accuracy": 0.2374246448278427,
      "num_tokens": 14945571.0,
      "step": 8220
    },
    {
      "entropy": 5.517508888244629,
      "epoch": 7.066179630425441,
      "grad_norm": 1.46875,
      "learning_rate": 0.00015468611053454478,
      "loss": 4.7602,
      "mean_token_accuracy": 0.23772317916154861,
      "num_tokens": 14954586.0,
      "step": 8225
    },
    {
      "entropy": 5.565473413467407,
      "epoch": 7.070477009024495,
      "grad_norm": 1.515625,
      "learning_rate": 0.00015440528608790533,
      "loss": 4.8417,
      "mean_token_accuracy": 0.23188529163599014,
      "num_tokens": 14963048.0,
      "step": 8230
    },
    {
      "entropy": 5.62564058303833,
      "epoch": 7.07477438762355,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001541247249719197,
      "loss": 4.8576,
      "mean_token_accuracy": 0.22911639660596847,
      "num_tokens": 14972158.0,
      "step": 8235
    },
    {
      "entropy": 5.559882926940918,
      "epoch": 7.079071766222604,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015384442779922135,
      "loss": 4.8109,
      "mean_token_accuracy": 0.2286899447441101,
      "num_tokens": 14980869.0,
      "step": 8240
    },
    {
      "entropy": 5.556419467926025,
      "epoch": 7.083369144821659,
      "grad_norm": 1.5,
      "learning_rate": 0.00015356439518186726,
      "loss": 4.7882,
      "mean_token_accuracy": 0.2323804795742035,
      "num_tokens": 14989103.0,
      "step": 8245
    },
    {
      "entropy": 5.543722009658813,
      "epoch": 7.0876665234207135,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015328462773133672,
      "loss": 4.8263,
      "mean_token_accuracy": 0.23760733604431153,
      "num_tokens": 14999378.0,
      "step": 8250
    },
    {
      "entropy": 5.604150438308716,
      "epoch": 7.091963902019768,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015300512605852977,
      "loss": 4.8879,
      "mean_token_accuracy": 0.2223494365811348,
      "num_tokens": 15007971.0,
      "step": 8255
    },
    {
      "entropy": 5.65213189125061,
      "epoch": 7.0962612806188226,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001527258907737668,
      "loss": 4.8854,
      "mean_token_accuracy": 0.2223148986697197,
      "num_tokens": 15017537.0,
      "step": 8260
    },
    {
      "entropy": 5.5658944606781,
      "epoch": 7.100558659217877,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00015244692248678586,
      "loss": 4.7836,
      "mean_token_accuracy": 0.23811090737581253,
      "num_tokens": 15025684.0,
      "step": 8265
    },
    {
      "entropy": 5.568011999130249,
      "epoch": 7.104856037816932,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001521682218067421,
      "loss": 4.813,
      "mean_token_accuracy": 0.22853953242301941,
      "num_tokens": 15034753.0,
      "step": 8270
    },
    {
      "entropy": 5.582677841186523,
      "epoch": 7.109153416415986,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015188978934220642,
      "loss": 4.8381,
      "mean_token_accuracy": 0.22904444634914398,
      "num_tokens": 15044685.0,
      "step": 8275
    },
    {
      "entropy": 5.593039083480835,
      "epoch": 7.113450795015041,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001516116257011641,
      "loss": 4.8678,
      "mean_token_accuracy": 0.22881769239902497,
      "num_tokens": 15054853.0,
      "step": 8280
    },
    {
      "entropy": 5.5440826416015625,
      "epoch": 7.117748173614095,
      "grad_norm": 1.25,
      "learning_rate": 0.0001513337314910134,
      "loss": 4.8269,
      "mean_token_accuracy": 0.2262088805437088,
      "num_tokens": 15065244.0,
      "step": 8285
    },
    {
      "entropy": 5.543696880340576,
      "epoch": 7.12204555221315,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015105610731856416,
      "loss": 4.7814,
      "mean_token_accuracy": 0.23860660791397095,
      "num_tokens": 15074046.0,
      "step": 8290
    },
    {
      "entropy": 5.536691665649414,
      "epoch": 7.126342930812204,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015077875379003653,
      "loss": 4.8226,
      "mean_token_accuracy": 0.22963619232177734,
      "num_tokens": 15083518.0,
      "step": 8295
    },
    {
      "entropy": 5.600537109375,
      "epoch": 7.130640309411259,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015050167151105988,
      "loss": 4.8769,
      "mean_token_accuracy": 0.22942472100257874,
      "num_tokens": 15092512.0,
      "step": 8300
    },
    {
      "entropy": 5.601132440567016,
      "epoch": 7.1349376880103135,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000150224861086671,
      "loss": 4.8419,
      "mean_token_accuracy": 0.22564840018749238,
      "num_tokens": 15101722.0,
      "step": 8305
    },
    {
      "entropy": 5.559588050842285,
      "epoch": 7.139235066609368,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014994832312131332,
      "loss": 4.8055,
      "mean_token_accuracy": 0.23476933538913727,
      "num_tokens": 15110114.0,
      "step": 8310
    },
    {
      "entropy": 5.593437194824219,
      "epoch": 7.143532445208423,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014967205821883532,
      "loss": 4.8829,
      "mean_token_accuracy": 0.22446369826793672,
      "num_tokens": 15119461.0,
      "step": 8315
    },
    {
      "entropy": 5.576046276092529,
      "epoch": 7.147829823807477,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000149396066982489,
      "loss": 4.7984,
      "mean_token_accuracy": 0.23506859540939332,
      "num_tokens": 15127518.0,
      "step": 8320
    },
    {
      "entropy": 5.584294271469116,
      "epoch": 7.152127202406532,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014912035001492897,
      "loss": 4.9032,
      "mean_token_accuracy": 0.2214670792222023,
      "num_tokens": 15136741.0,
      "step": 8325
    },
    {
      "entropy": 5.5741575241088865,
      "epoch": 7.156424581005586,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014884490791821058,
      "loss": 4.8296,
      "mean_token_accuracy": 0.23365769237279893,
      "num_tokens": 15145193.0,
      "step": 8330
    },
    {
      "entropy": 5.575591707229615,
      "epoch": 7.160721959604641,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014856974129378981,
      "loss": 4.8463,
      "mean_token_accuracy": 0.2283509209752083,
      "num_tokens": 15154117.0,
      "step": 8335
    },
    {
      "entropy": 5.584064912796021,
      "epoch": 7.165019338203695,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001482948507425203,
      "loss": 4.8584,
      "mean_token_accuracy": 0.2270919382572174,
      "num_tokens": 15163221.0,
      "step": 8340
    },
    {
      "entropy": 5.594315862655639,
      "epoch": 7.169316716802751,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014802023686465314,
      "loss": 4.9283,
      "mean_token_accuracy": 0.21705838590860366,
      "num_tokens": 15173234.0,
      "step": 8345
    },
    {
      "entropy": 5.559284973144531,
      "epoch": 7.173614095401805,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014774590025983523,
      "loss": 4.7969,
      "mean_token_accuracy": 0.23195214569568634,
      "num_tokens": 15181436.0,
      "step": 8350
    },
    {
      "entropy": 5.534706735610962,
      "epoch": 7.17791147400086,
      "grad_norm": 1.515625,
      "learning_rate": 0.00014747184152710807,
      "loss": 4.8244,
      "mean_token_accuracy": 0.236300465464592,
      "num_tokens": 15191697.0,
      "step": 8355
    },
    {
      "entropy": 5.540991115570068,
      "epoch": 7.182208852599914,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014719806126490658,
      "loss": 4.8293,
      "mean_token_accuracy": 0.235206139087677,
      "num_tokens": 15201563.0,
      "step": 8360
    },
    {
      "entropy": 5.564880418777466,
      "epoch": 7.186506231198969,
      "grad_norm": 1.5,
      "learning_rate": 0.0001469245600710573,
      "loss": 4.8077,
      "mean_token_accuracy": 0.23383856266736985,
      "num_tokens": 15210886.0,
      "step": 8365
    },
    {
      "entropy": 5.578991317749024,
      "epoch": 7.1908036097980235,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014665133854277742,
      "loss": 4.8765,
      "mean_token_accuracy": 0.22310879826545715,
      "num_tokens": 15219254.0,
      "step": 8370
    },
    {
      "entropy": 5.541424179077149,
      "epoch": 7.195100988397078,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001463783972766737,
      "loss": 4.8282,
      "mean_token_accuracy": 0.23382658809423446,
      "num_tokens": 15228117.0,
      "step": 8375
    },
    {
      "entropy": 5.505723714828491,
      "epoch": 7.199398366996133,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001461057368687407,
      "loss": 4.8224,
      "mean_token_accuracy": 0.23662276566028595,
      "num_tokens": 15236621.0,
      "step": 8380
    },
    {
      "entropy": 5.513776397705078,
      "epoch": 7.203695745595187,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014583335791435971,
      "loss": 4.7552,
      "mean_token_accuracy": 0.23776894807815552,
      "num_tokens": 15245487.0,
      "step": 8385
    },
    {
      "entropy": 5.551266717910766,
      "epoch": 7.207993124194242,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014556126100829774,
      "loss": 4.8141,
      "mean_token_accuracy": 0.2299559786915779,
      "num_tokens": 15255321.0,
      "step": 8390
    },
    {
      "entropy": 5.495361280441284,
      "epoch": 7.212290502793296,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014528944674470546,
      "loss": 4.7238,
      "mean_token_accuracy": 0.24432678073644637,
      "num_tokens": 15264788.0,
      "step": 8395
    },
    {
      "entropy": 5.586590385437011,
      "epoch": 7.216587881392351,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001450179157171166,
      "loss": 4.8604,
      "mean_token_accuracy": 0.2202071949839592,
      "num_tokens": 15273448.0,
      "step": 8400
    },
    {
      "entropy": 5.622144174575806,
      "epoch": 7.220885259991405,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014474666851844632,
      "loss": 4.9269,
      "mean_token_accuracy": 0.2213926136493683,
      "num_tokens": 15283071.0,
      "step": 8405
    },
    {
      "entropy": 5.521999263763428,
      "epoch": 7.22518263859046,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014447570574099028,
      "loss": 4.7523,
      "mean_token_accuracy": 0.23776609897613527,
      "num_tokens": 15291537.0,
      "step": 8410
    },
    {
      "entropy": 5.602297449111939,
      "epoch": 7.229480017189514,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014420502797642283,
      "loss": 4.8106,
      "mean_token_accuracy": 0.23558389097452165,
      "num_tokens": 15300531.0,
      "step": 8415
    },
    {
      "entropy": 5.57661280632019,
      "epoch": 7.233777395788569,
      "grad_norm": 1.421875,
      "learning_rate": 0.000143934635815796,
      "loss": 4.8989,
      "mean_token_accuracy": 0.22630394846200944,
      "num_tokens": 15309820.0,
      "step": 8420
    },
    {
      "entropy": 5.546637344360351,
      "epoch": 7.2380747743876235,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001436645298495381,
      "loss": 4.8585,
      "mean_token_accuracy": 0.2299433633685112,
      "num_tokens": 15318604.0,
      "step": 8425
    },
    {
      "entropy": 5.527101802825928,
      "epoch": 7.242372152986678,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014339471066745262,
      "loss": 4.8441,
      "mean_token_accuracy": 0.22634208649396897,
      "num_tokens": 15327737.0,
      "step": 8430
    },
    {
      "entropy": 5.566941642761231,
      "epoch": 7.246669531585733,
      "grad_norm": 1.421875,
      "learning_rate": 0.000143125178858717,
      "loss": 4.9019,
      "mean_token_accuracy": 0.223842254281044,
      "num_tokens": 15336663.0,
      "step": 8435
    },
    {
      "entropy": 5.622009801864624,
      "epoch": 7.250966910184787,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014285593501188083,
      "loss": 4.9119,
      "mean_token_accuracy": 0.22033586353063583,
      "num_tokens": 15345278.0,
      "step": 8440
    },
    {
      "entropy": 5.619880151748657,
      "epoch": 7.255264288783842,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014258697971486492,
      "loss": 4.8591,
      "mean_token_accuracy": 0.22582849115133286,
      "num_tokens": 15354230.0,
      "step": 8445
    },
    {
      "entropy": 5.58750696182251,
      "epoch": 7.259561667382896,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014231831355496045,
      "loss": 4.8682,
      "mean_token_accuracy": 0.23125071525573732,
      "num_tokens": 15362838.0,
      "step": 8450
    },
    {
      "entropy": 5.506886005401611,
      "epoch": 7.263859045981951,
      "grad_norm": 1.21875,
      "learning_rate": 0.00014204993711882662,
      "loss": 4.8487,
      "mean_token_accuracy": 0.22859285026788712,
      "num_tokens": 15372593.0,
      "step": 8455
    },
    {
      "entropy": 5.594306993484497,
      "epoch": 7.268156424581005,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001417818509924906,
      "loss": 4.9056,
      "mean_token_accuracy": 0.22276172190904617,
      "num_tokens": 15381945.0,
      "step": 8460
    },
    {
      "entropy": 5.541630268096924,
      "epoch": 7.27245380318006,
      "grad_norm": 1.296875,
      "learning_rate": 0.000141514055761345,
      "loss": 4.8173,
      "mean_token_accuracy": 0.22973333299160004,
      "num_tokens": 15391487.0,
      "step": 8465
    },
    {
      "entropy": 5.591726016998291,
      "epoch": 7.276751181779114,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014124655201014786,
      "loss": 4.7836,
      "mean_token_accuracy": 0.23431181758642197,
      "num_tokens": 15399891.0,
      "step": 8470
    },
    {
      "entropy": 5.57764573097229,
      "epoch": 7.281048560378169,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014097934032302037,
      "loss": 4.7899,
      "mean_token_accuracy": 0.23358217179775237,
      "num_tokens": 15408693.0,
      "step": 8475
    },
    {
      "entropy": 5.5628900051116945,
      "epoch": 7.2853459389772235,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014071242128344593,
      "loss": 4.8927,
      "mean_token_accuracy": 0.2317359060049057,
      "num_tokens": 15417779.0,
      "step": 8480
    },
    {
      "entropy": 5.533954048156739,
      "epoch": 7.289643317576278,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001404457954742691,
      "loss": 4.7861,
      "mean_token_accuracy": 0.2404788017272949,
      "num_tokens": 15425826.0,
      "step": 8485
    },
    {
      "entropy": 5.561101675033569,
      "epoch": 7.2939406961753335,
      "grad_norm": 1.5625,
      "learning_rate": 0.00014017946347769423,
      "loss": 4.8518,
      "mean_token_accuracy": 0.2285791128873825,
      "num_tokens": 15435811.0,
      "step": 8490
    },
    {
      "entropy": 5.509159708023072,
      "epoch": 7.298238074774388,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013991342587528377,
      "loss": 4.765,
      "mean_token_accuracy": 0.23475974649190903,
      "num_tokens": 15444949.0,
      "step": 8495
    },
    {
      "entropy": 5.484537172317505,
      "epoch": 7.302535453373443,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013964768324795752,
      "loss": 4.7819,
      "mean_token_accuracy": 0.2423618733882904,
      "num_tokens": 15453398.0,
      "step": 8500
    },
    {
      "epoch": 7.302535453373443,
      "eval_entropy": 5.418461682023229,
      "eval_loss": 5.940352439880371,
      "eval_mean_token_accuracy": 0.17928062282032795,
      "eval_num_tokens": 15453398.0,
      "eval_runtime": 2.641,
      "eval_samples_per_second": 1343.805,
      "eval_steps_per_second": 168.118,
      "step": 8500
    },
    {
      "entropy": 5.534810400009155,
      "epoch": 7.306832831972497,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013938223617599124,
      "loss": 4.8636,
      "mean_token_accuracy": 0.22623300850391387,
      "num_tokens": 15462785.0,
      "step": 8505
    },
    {
      "entropy": 5.575749683380127,
      "epoch": 7.311130210571552,
      "grad_norm": 1.5,
      "learning_rate": 0.00013911708523901514,
      "loss": 4.8787,
      "mean_token_accuracy": 0.23186656832695007,
      "num_tokens": 15471718.0,
      "step": 8510
    },
    {
      "entropy": 5.580225086212158,
      "epoch": 7.315427589170606,
      "grad_norm": 1.375,
      "learning_rate": 0.00013885223101601303,
      "loss": 4.8253,
      "mean_token_accuracy": 0.23118109554052352,
      "num_tokens": 15480204.0,
      "step": 8515
    },
    {
      "entropy": 5.532940483093261,
      "epoch": 7.319724967769661,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013858767408532051,
      "loss": 4.779,
      "mean_token_accuracy": 0.23626572489738465,
      "num_tokens": 15489388.0,
      "step": 8520
    },
    {
      "entropy": 5.517295503616333,
      "epoch": 7.324022346368715,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013832341502462432,
      "loss": 4.7788,
      "mean_token_accuracy": 0.23014454245567323,
      "num_tokens": 15498028.0,
      "step": 8525
    },
    {
      "entropy": 5.569064092636109,
      "epoch": 7.32831972496777,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00013805945441096057,
      "loss": 4.8395,
      "mean_token_accuracy": 0.2291802555322647,
      "num_tokens": 15506382.0,
      "step": 8530
    },
    {
      "entropy": 5.554863977432251,
      "epoch": 7.332617103566824,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013779579282071364,
      "loss": 4.8346,
      "mean_token_accuracy": 0.23108793050050735,
      "num_tokens": 15515271.0,
      "step": 8535
    },
    {
      "entropy": 5.5129218101501465,
      "epoch": 7.336914482165879,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013753243082961512,
      "loss": 4.7922,
      "mean_token_accuracy": 0.23765636533498763,
      "num_tokens": 15524396.0,
      "step": 8540
    },
    {
      "entropy": 5.535199928283691,
      "epoch": 7.3412118607649335,
      "grad_norm": 1.46875,
      "learning_rate": 0.00013726936901274246,
      "loss": 4.7561,
      "mean_token_accuracy": 0.24567289054393768,
      "num_tokens": 15532829.0,
      "step": 8545
    },
    {
      "entropy": 5.59448504447937,
      "epoch": 7.345509239363988,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001370066079445174,
      "loss": 4.8774,
      "mean_token_accuracy": 0.22510895133018494,
      "num_tokens": 15541726.0,
      "step": 8550
    },
    {
      "entropy": 5.605797958374024,
      "epoch": 7.349806617963043,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013674414819870502,
      "loss": 4.994,
      "mean_token_accuracy": 0.21879294216632844,
      "num_tokens": 15551539.0,
      "step": 8555
    },
    {
      "entropy": 5.552589416503906,
      "epoch": 7.354103996562097,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013648199034841264,
      "loss": 4.8261,
      "mean_token_accuracy": 0.23322057873010635,
      "num_tokens": 15560147.0,
      "step": 8560
    },
    {
      "entropy": 5.550334692001343,
      "epoch": 7.358401375161152,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001362201349660882,
      "loss": 4.8127,
      "mean_token_accuracy": 0.24100401103496552,
      "num_tokens": 15568983.0,
      "step": 8565
    },
    {
      "entropy": 5.509068441390991,
      "epoch": 7.362698753760206,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001359585826235192,
      "loss": 4.8097,
      "mean_token_accuracy": 0.23465747982263566,
      "num_tokens": 15578065.0,
      "step": 8570
    },
    {
      "entropy": 5.613720560073853,
      "epoch": 7.366996132359261,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013569733389183126,
      "loss": 4.8995,
      "mean_token_accuracy": 0.22496907860040666,
      "num_tokens": 15587181.0,
      "step": 8575
    },
    {
      "entropy": 5.55947732925415,
      "epoch": 7.371293510958315,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013543638934148736,
      "loss": 4.8274,
      "mean_token_accuracy": 0.23065395951271056,
      "num_tokens": 15596602.0,
      "step": 8580
    },
    {
      "entropy": 5.637532711029053,
      "epoch": 7.37559088955737,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000135175749542286,
      "loss": 4.9016,
      "mean_token_accuracy": 0.21847975105047227,
      "num_tokens": 15605857.0,
      "step": 8585
    },
    {
      "entropy": 5.529611825942993,
      "epoch": 7.379888268156424,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001349154150633604,
      "loss": 4.7819,
      "mean_token_accuracy": 0.2396458774805069,
      "num_tokens": 15615320.0,
      "step": 8590
    },
    {
      "entropy": 5.599874448776245,
      "epoch": 7.384185646755479,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000134655386473177,
      "loss": 4.9344,
      "mean_token_accuracy": 0.21727651059627534,
      "num_tokens": 15624193.0,
      "step": 8595
    },
    {
      "entropy": 5.464236688613892,
      "epoch": 7.3884830253545335,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00013439566433953427,
      "loss": 4.7906,
      "mean_token_accuracy": 0.23281699866056443,
      "num_tokens": 15632924.0,
      "step": 8600
    },
    {
      "entropy": 5.544660139083862,
      "epoch": 7.392780403953588,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001341362492295616,
      "loss": 4.8005,
      "mean_token_accuracy": 0.2357855573296547,
      "num_tokens": 15642201.0,
      "step": 8605
    },
    {
      "entropy": 5.585439729690552,
      "epoch": 7.397077782552643,
      "grad_norm": 1.546875,
      "learning_rate": 0.00013387714170971776,
      "loss": 4.8274,
      "mean_token_accuracy": 0.23000081032514572,
      "num_tokens": 15651608.0,
      "step": 8610
    },
    {
      "entropy": 5.604709148406982,
      "epoch": 7.401375161151697,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013361834234579012,
      "loss": 4.9392,
      "mean_token_accuracy": 0.2254776641726494,
      "num_tokens": 15661768.0,
      "step": 8615
    },
    {
      "entropy": 5.505479192733764,
      "epoch": 7.405672539750752,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001333598517028931,
      "loss": 4.8145,
      "mean_token_accuracy": 0.22790594398975372,
      "num_tokens": 15670270.0,
      "step": 8620
    },
    {
      "entropy": 5.541030788421631,
      "epoch": 7.409969918349806,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00013310167034546688,
      "loss": 4.81,
      "mean_token_accuracy": 0.23079528510570527,
      "num_tokens": 15679587.0,
      "step": 8625
    },
    {
      "entropy": 5.560227918624878,
      "epoch": 7.414267296948861,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001328437988372763,
      "loss": 4.8617,
      "mean_token_accuracy": 0.22552231550216675,
      "num_tokens": 15688838.0,
      "step": 8630
    },
    {
      "entropy": 5.569988155364991,
      "epoch": 7.418564675547916,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013258623774140967,
      "loss": 4.8311,
      "mean_token_accuracy": 0.23475466370582582,
      "num_tokens": 15697744.0,
      "step": 8635
    },
    {
      "entropy": 5.533057928085327,
      "epoch": 7.422862054146971,
      "grad_norm": 1.53125,
      "learning_rate": 0.00013232898762027766,
      "loss": 4.8004,
      "mean_token_accuracy": 0.2368886798620224,
      "num_tokens": 15707643.0,
      "step": 8640
    },
    {
      "entropy": 5.568829727172852,
      "epoch": 7.427159432746025,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013207204903561154,
      "loss": 4.8869,
      "mean_token_accuracy": 0.22737328708171844,
      "num_tokens": 15717568.0,
      "step": 8645
    },
    {
      "entropy": 5.51854133605957,
      "epoch": 7.43145681134508,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013181542254846247,
      "loss": 4.7625,
      "mean_token_accuracy": 0.2366994708776474,
      "num_tokens": 15726467.0,
      "step": 8650
    },
    {
      "entropy": 5.567011833190918,
      "epoch": 7.435754189944134,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001315591087192002,
      "loss": 4.8962,
      "mean_token_accuracy": 0.2260964944958687,
      "num_tokens": 15736533.0,
      "step": 8655
    },
    {
      "entropy": 5.477988195419312,
      "epoch": 7.440051568543189,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013130310810751162,
      "loss": 4.7993,
      "mean_token_accuracy": 0.22829737961292268,
      "num_tokens": 15745853.0,
      "step": 8660
    },
    {
      "entropy": 5.61041030883789,
      "epoch": 7.4443489471422435,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013104742127239983,
      "loss": 4.925,
      "mean_token_accuracy": 0.21945925205945968,
      "num_tokens": 15755534.0,
      "step": 8665
    },
    {
      "entropy": 5.611725568771362,
      "epoch": 7.448646325741298,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001307920487721826,
      "loss": 4.9035,
      "mean_token_accuracy": 0.21820500493049622,
      "num_tokens": 15766182.0,
      "step": 8670
    },
    {
      "entropy": 5.548447847366333,
      "epoch": 7.452943704340353,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013053699116449144,
      "loss": 4.8311,
      "mean_token_accuracy": 0.2286737695336342,
      "num_tokens": 15775454.0,
      "step": 8675
    },
    {
      "entropy": 5.6966629981994625,
      "epoch": 7.457241082939407,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013028224900627026,
      "loss": 4.9015,
      "mean_token_accuracy": 0.2242224857211113,
      "num_tokens": 15784768.0,
      "step": 8680
    },
    {
      "entropy": 5.667233657836914,
      "epoch": 7.461538461538462,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00013002782285377395,
      "loss": 4.8797,
      "mean_token_accuracy": 0.22435784637928008,
      "num_tokens": 15794255.0,
      "step": 8685
    },
    {
      "entropy": 5.583252811431885,
      "epoch": 7.465835840137516,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001297737132625677,
      "loss": 4.8885,
      "mean_token_accuracy": 0.2244955152273178,
      "num_tokens": 15803722.0,
      "step": 8690
    },
    {
      "entropy": 5.608900785446167,
      "epoch": 7.470133218736571,
      "grad_norm": 1.703125,
      "learning_rate": 0.00012951992078752528,
      "loss": 4.8329,
      "mean_token_accuracy": 0.22491242438554765,
      "num_tokens": 15811819.0,
      "step": 8695
    },
    {
      "entropy": 5.555014562606812,
      "epoch": 7.474430597335625,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012926644598282798,
      "loss": 4.8482,
      "mean_token_accuracy": 0.22133827954530716,
      "num_tokens": 15821446.0,
      "step": 8700
    },
    {
      "entropy": 5.592018413543701,
      "epoch": 7.47872797593468,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001290132894019634,
      "loss": 4.868,
      "mean_token_accuracy": 0.2351159542798996,
      "num_tokens": 15830585.0,
      "step": 8705
    },
    {
      "entropy": 5.62072377204895,
      "epoch": 7.483025354533734,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012876045159772442,
      "loss": 4.8513,
      "mean_token_accuracy": 0.2290963351726532,
      "num_tokens": 15838872.0,
      "step": 8710
    },
    {
      "entropy": 5.607170057296753,
      "epoch": 7.487322733132789,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012850793312220766,
      "loss": 4.8396,
      "mean_token_accuracy": 0.2252478614449501,
      "num_tokens": 15847561.0,
      "step": 8715
    },
    {
      "entropy": 5.516605234146118,
      "epoch": 7.4916201117318435,
      "grad_norm": 1.296875,
      "learning_rate": 0.00012825573452681266,
      "loss": 4.7882,
      "mean_token_accuracy": 0.240067820250988,
      "num_tokens": 15856405.0,
      "step": 8720
    },
    {
      "entropy": 5.574163389205933,
      "epoch": 7.495917490330898,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012800385636224017,
      "loss": 4.8592,
      "mean_token_accuracy": 0.22380622625350952,
      "num_tokens": 15865856.0,
      "step": 8725
    },
    {
      "entropy": 5.614547967910767,
      "epoch": 7.500214868929953,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012775229917849162,
      "loss": 4.799,
      "mean_token_accuracy": 0.23128498643636702,
      "num_tokens": 15873605.0,
      "step": 8730
    },
    {
      "entropy": 5.592815160751343,
      "epoch": 7.504512247529007,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012750106352486728,
      "loss": 4.7973,
      "mean_token_accuracy": 0.2359904482960701,
      "num_tokens": 15883123.0,
      "step": 8735
    },
    {
      "entropy": 5.5159807205200195,
      "epoch": 7.508809626128062,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012725014994996534,
      "loss": 4.8343,
      "mean_token_accuracy": 0.23460762500762938,
      "num_tokens": 15892713.0,
      "step": 8740
    },
    {
      "entropy": 5.566962862014771,
      "epoch": 7.513107004727116,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012699955900168075,
      "loss": 4.8355,
      "mean_token_accuracy": 0.2303671807050705,
      "num_tokens": 15902913.0,
      "step": 8745
    },
    {
      "entropy": 5.553069067001343,
      "epoch": 7.517404383326171,
      "grad_norm": 1.375,
      "learning_rate": 0.00012674929122720414,
      "loss": 4.8781,
      "mean_token_accuracy": 0.22082593888044358,
      "num_tokens": 15912721.0,
      "step": 8750
    },
    {
      "entropy": 5.484966802597046,
      "epoch": 7.521701761925225,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001264993471730202,
      "loss": 4.7771,
      "mean_token_accuracy": 0.23455059081315993,
      "num_tokens": 15921520.0,
      "step": 8755
    },
    {
      "entropy": 5.609423875808716,
      "epoch": 7.52599914052428,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012624972738490675,
      "loss": 4.9138,
      "mean_token_accuracy": 0.21497470289468765,
      "num_tokens": 15930753.0,
      "step": 8760
    },
    {
      "entropy": 5.636999130249023,
      "epoch": 7.530296519123334,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012600043240793368,
      "loss": 4.8701,
      "mean_token_accuracy": 0.22687977701425552,
      "num_tokens": 15939957.0,
      "step": 8765
    },
    {
      "entropy": 5.563047266006469,
      "epoch": 7.534593897722389,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012575146278646175,
      "loss": 4.797,
      "mean_token_accuracy": 0.23644523024559022,
      "num_tokens": 15949555.0,
      "step": 8770
    },
    {
      "entropy": 5.522928619384766,
      "epoch": 7.5388912763214435,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012550281906414097,
      "loss": 4.8092,
      "mean_token_accuracy": 0.23000673949718475,
      "num_tokens": 15958395.0,
      "step": 8775
    },
    {
      "entropy": 5.583071756362915,
      "epoch": 7.543188654920499,
      "grad_norm": 1.375,
      "learning_rate": 0.00012525450178390972,
      "loss": 4.8633,
      "mean_token_accuracy": 0.2339993715286255,
      "num_tokens": 15967522.0,
      "step": 8780
    },
    {
      "entropy": 5.587414169311524,
      "epoch": 7.547486033519553,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001250065114879939,
      "loss": 4.8373,
      "mean_token_accuracy": 0.23017602264881135,
      "num_tokens": 15976311.0,
      "step": 8785
    },
    {
      "entropy": 5.541785144805909,
      "epoch": 7.551783412118608,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012475884871790505,
      "loss": 4.782,
      "mean_token_accuracy": 0.23119063228368758,
      "num_tokens": 15985202.0,
      "step": 8790
    },
    {
      "entropy": 5.644336605072022,
      "epoch": 7.556080790717663,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012451151401443982,
      "loss": 4.9187,
      "mean_token_accuracy": 0.21566017866134643,
      "num_tokens": 15995043.0,
      "step": 8795
    },
    {
      "entropy": 5.465802717208862,
      "epoch": 7.560378169316717,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012426450791767815,
      "loss": 4.7555,
      "mean_token_accuracy": 0.24007294625043868,
      "num_tokens": 16004355.0,
      "step": 8800
    },
    {
      "entropy": 5.543409776687622,
      "epoch": 7.564675547915772,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012401783096698283,
      "loss": 4.6898,
      "mean_token_accuracy": 0.23701778799295425,
      "num_tokens": 16013069.0,
      "step": 8805
    },
    {
      "entropy": 5.568853044509888,
      "epoch": 7.568972926514826,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012377148370099764,
      "loss": 4.8676,
      "mean_token_accuracy": 0.22199283838272094,
      "num_tokens": 16023757.0,
      "step": 8810
    },
    {
      "entropy": 5.572279739379883,
      "epoch": 7.573270305113881,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012352546665764642,
      "loss": 4.8801,
      "mean_token_accuracy": 0.22454654574394226,
      "num_tokens": 16032550.0,
      "step": 8815
    },
    {
      "entropy": 5.471732950210571,
      "epoch": 7.577567683712935,
      "grad_norm": 1.3125,
      "learning_rate": 0.00012327978037413219,
      "loss": 4.7249,
      "mean_token_accuracy": 0.24093529880046843,
      "num_tokens": 16041580.0,
      "step": 8820
    },
    {
      "entropy": 5.564283943176269,
      "epoch": 7.58186506231199,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00012303442538693564,
      "loss": 4.8605,
      "mean_token_accuracy": 0.23017286211252214,
      "num_tokens": 16049845.0,
      "step": 8825
    },
    {
      "entropy": 5.467882394790649,
      "epoch": 7.586162440911044,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012278940223181393,
      "loss": 4.7328,
      "mean_token_accuracy": 0.23491140455007553,
      "num_tokens": 16059703.0,
      "step": 8830
    },
    {
      "entropy": 5.516620731353759,
      "epoch": 7.590459819510099,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012254471144379964,
      "loss": 4.7195,
      "mean_token_accuracy": 0.24515746682882308,
      "num_tokens": 16068416.0,
      "step": 8835
    },
    {
      "entropy": 5.5409619331359865,
      "epoch": 7.5947571981091535,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012230035355719968,
      "loss": 4.8735,
      "mean_token_accuracy": 0.22853126674890517,
      "num_tokens": 16078067.0,
      "step": 8840
    },
    {
      "entropy": 5.604869794845581,
      "epoch": 7.599054576708208,
      "grad_norm": 1.375,
      "learning_rate": 0.0001220563291055941,
      "loss": 4.8628,
      "mean_token_accuracy": 0.22932736426591874,
      "num_tokens": 16086591.0,
      "step": 8845
    },
    {
      "entropy": 5.6001232147216795,
      "epoch": 7.603351955307263,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001218126386218347,
      "loss": 4.8535,
      "mean_token_accuracy": 0.22812440246343613,
      "num_tokens": 16096138.0,
      "step": 8850
    },
    {
      "entropy": 5.515623235702515,
      "epoch": 7.607649333906317,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012156928263804403,
      "loss": 4.7886,
      "mean_token_accuracy": 0.23354648649692536,
      "num_tokens": 16105182.0,
      "step": 8855
    },
    {
      "entropy": 5.568499946594239,
      "epoch": 7.611946712505372,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001213262616856144,
      "loss": 4.9025,
      "mean_token_accuracy": 0.22694600820541383,
      "num_tokens": 16113940.0,
      "step": 8860
    },
    {
      "entropy": 5.562524366378784,
      "epoch": 7.616244091104426,
      "grad_norm": 1.546875,
      "learning_rate": 0.00012108357629520635,
      "loss": 4.8046,
      "mean_token_accuracy": 0.23113808333873748,
      "num_tokens": 16123036.0,
      "step": 8865
    },
    {
      "entropy": 5.561686563491821,
      "epoch": 7.620541469703481,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012084122699674785,
      "loss": 4.8121,
      "mean_token_accuracy": 0.23125939220190048,
      "num_tokens": 16131057.0,
      "step": 8870
    },
    {
      "entropy": 5.506033182144165,
      "epoch": 7.624838848302535,
      "grad_norm": 1.53125,
      "learning_rate": 0.00012059921431943278,
      "loss": 4.8631,
      "mean_token_accuracy": 0.22523228228092193,
      "num_tokens": 16140259.0,
      "step": 8875
    },
    {
      "entropy": 5.623447608947754,
      "epoch": 7.62913622690159,
      "grad_norm": 1.5625,
      "learning_rate": 0.00012035753879172026,
      "loss": 5.0524,
      "mean_token_accuracy": 0.2060105249285698,
      "num_tokens": 16149585.0,
      "step": 8880
    },
    {
      "entropy": 5.506596660614013,
      "epoch": 7.633433605500644,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012011620094133296,
      "loss": 4.6716,
      "mean_token_accuracy": 0.24745012819766998,
      "num_tokens": 16157656.0,
      "step": 8885
    },
    {
      "entropy": 5.49847297668457,
      "epoch": 7.637730984099699,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011987520129525622,
      "loss": 4.8197,
      "mean_token_accuracy": 0.23195276260375977,
      "num_tokens": 16166900.0,
      "step": 8890
    },
    {
      "entropy": 5.562042284011841,
      "epoch": 7.6420283626987535,
      "grad_norm": 1.546875,
      "learning_rate": 0.000119634540379737,
      "loss": 4.8676,
      "mean_token_accuracy": 0.2259794741868973,
      "num_tokens": 16174859.0,
      "step": 8895
    },
    {
      "entropy": 5.6275067806243895,
      "epoch": 7.646325741297808,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011939421872028262,
      "loss": 4.8533,
      "mean_token_accuracy": 0.23243143707513808,
      "num_tokens": 16183660.0,
      "step": 8900
    },
    {
      "entropy": 5.538630867004395,
      "epoch": 7.650623119896863,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011915423684165948,
      "loss": 4.7873,
      "mean_token_accuracy": 0.23280810415744782,
      "num_tokens": 16192344.0,
      "step": 8905
    },
    {
      "entropy": 5.607868671417236,
      "epoch": 7.654920498495917,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011891459526789198,
      "loss": 4.873,
      "mean_token_accuracy": 0.22652816027402878,
      "num_tokens": 16202060.0,
      "step": 8910
    },
    {
      "entropy": 5.5807970523834225,
      "epoch": 7.659217877094972,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001186752945222616,
      "loss": 4.8515,
      "mean_token_accuracy": 0.22743048071861266,
      "num_tokens": 16211297.0,
      "step": 8915
    },
    {
      "entropy": 5.57390866279602,
      "epoch": 7.663515255694026,
      "grad_norm": 1.375,
      "learning_rate": 0.00011843633512730562,
      "loss": 4.8068,
      "mean_token_accuracy": 0.23263307362794877,
      "num_tokens": 16219812.0,
      "step": 8920
    },
    {
      "entropy": 5.54047532081604,
      "epoch": 7.667812634293082,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011819771760481576,
      "loss": 4.8192,
      "mean_token_accuracy": 0.23949947208166122,
      "num_tokens": 16229197.0,
      "step": 8925
    },
    {
      "entropy": 5.483737325668335,
      "epoch": 7.672110012892135,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011795944247583725,
      "loss": 4.7433,
      "mean_token_accuracy": 0.23419629484415055,
      "num_tokens": 16238154.0,
      "step": 8930
    },
    {
      "entropy": 5.506278038024902,
      "epoch": 7.676407391491191,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011772151026066789,
      "loss": 4.795,
      "mean_token_accuracy": 0.22703787535429001,
      "num_tokens": 16247206.0,
      "step": 8935
    },
    {
      "entropy": 5.612638998031616,
      "epoch": 7.680704770090245,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011748392147885642,
      "loss": 4.9047,
      "mean_token_accuracy": 0.22023337036371232,
      "num_tokens": 16256571.0,
      "step": 8940
    },
    {
      "entropy": 5.571756410598755,
      "epoch": 7.6850021486893,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00011724667664920177,
      "loss": 4.8311,
      "mean_token_accuracy": 0.2272126019001007,
      "num_tokens": 16265429.0,
      "step": 8945
    },
    {
      "entropy": 5.590095806121826,
      "epoch": 7.689299527288354,
      "grad_norm": 1.5,
      "learning_rate": 0.00011700977628975183,
      "loss": 4.9385,
      "mean_token_accuracy": 0.2217423975467682,
      "num_tokens": 16273804.0,
      "step": 8950
    },
    {
      "entropy": 5.592393159866333,
      "epoch": 7.693596905887409,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011677322091780243,
      "loss": 4.8681,
      "mean_token_accuracy": 0.22543271034955978,
      "num_tokens": 16282894.0,
      "step": 8955
    },
    {
      "entropy": 5.623022603988647,
      "epoch": 7.6978942844864635,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001165370110498958,
      "loss": 4.8887,
      "mean_token_accuracy": 0.2259175643324852,
      "num_tokens": 16291568.0,
      "step": 8960
    },
    {
      "entropy": 5.65300350189209,
      "epoch": 7.702191663085518,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011630114720181989,
      "loss": 4.8814,
      "mean_token_accuracy": 0.22854696661233903,
      "num_tokens": 16300650.0,
      "step": 8965
    },
    {
      "entropy": 5.576943254470825,
      "epoch": 7.706489041684573,
      "grad_norm": 1.375,
      "learning_rate": 0.00011606562988860711,
      "loss": 4.823,
      "mean_token_accuracy": 0.2229461058974266,
      "num_tokens": 16309712.0,
      "step": 8970
    },
    {
      "entropy": 5.540242147445679,
      "epoch": 7.710786420283627,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001158304596245332,
      "loss": 4.8002,
      "mean_token_accuracy": 0.23573469370603561,
      "num_tokens": 16319440.0,
      "step": 8975
    },
    {
      "entropy": 5.469468116760254,
      "epoch": 7.715083798882682,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011559563692311595,
      "loss": 4.7976,
      "mean_token_accuracy": 0.240695957839489,
      "num_tokens": 16328752.0,
      "step": 8980
    },
    {
      "entropy": 5.5501690864562985,
      "epoch": 7.719381177481736,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011536116229711422,
      "loss": 4.7848,
      "mean_token_accuracy": 0.23823384940624237,
      "num_tokens": 16338045.0,
      "step": 8985
    },
    {
      "entropy": 5.616193532943726,
      "epoch": 7.723678556080791,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000115127036258527,
      "loss": 4.8623,
      "mean_token_accuracy": 0.22796519994735717,
      "num_tokens": 16347174.0,
      "step": 8990
    },
    {
      "entropy": 5.541440486907959,
      "epoch": 7.727975934679845,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011489325931859185,
      "loss": 4.7337,
      "mean_token_accuracy": 0.2423200160264969,
      "num_tokens": 16355371.0,
      "step": 8995
    },
    {
      "entropy": 5.496782302856445,
      "epoch": 7.7322733132789,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001146598319877843,
      "loss": 4.8006,
      "mean_token_accuracy": 0.23760196566581726,
      "num_tokens": 16363938.0,
      "step": 9000
    },
    {
      "epoch": 7.7322733132789,
      "eval_entropy": 5.414116493216506,
      "eval_loss": 5.934047222137451,
      "eval_mean_token_accuracy": 0.1796834310768424,
      "eval_num_tokens": 16363938.0,
      "eval_runtime": 2.8525,
      "eval_samples_per_second": 1244.169,
      "eval_steps_per_second": 155.653,
      "step": 9000
    },
    {
      "entropy": 5.551463937759399,
      "epoch": 7.736570691877954,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011442675477581621,
      "loss": 4.8601,
      "mean_token_accuracy": 0.23094746321439744,
      "num_tokens": 16373110.0,
      "step": 9005
    },
    {
      "entropy": 5.571599769592285,
      "epoch": 7.740868070477009,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001141940281916352,
      "loss": 4.7754,
      "mean_token_accuracy": 0.22749817818403245,
      "num_tokens": 16381521.0,
      "step": 9010
    },
    {
      "entropy": 5.6218891620635985,
      "epoch": 7.7451654490760635,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011396165274342304,
      "loss": 4.8856,
      "mean_token_accuracy": 0.21985125690698623,
      "num_tokens": 16391322.0,
      "step": 9015
    },
    {
      "entropy": 5.555123996734619,
      "epoch": 7.749462827675118,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011372962893859471,
      "loss": 4.8419,
      "mean_token_accuracy": 0.22425442039966584,
      "num_tokens": 16400653.0,
      "step": 9020
    },
    {
      "entropy": 5.56007604598999,
      "epoch": 7.753760206274173,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00011349795728379759,
      "loss": 4.854,
      "mean_token_accuracy": 0.22200659960508345,
      "num_tokens": 16410133.0,
      "step": 9025
    },
    {
      "entropy": 5.56913948059082,
      "epoch": 7.758057584873227,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011326663828491,
      "loss": 4.8895,
      "mean_token_accuracy": 0.22325430661439896,
      "num_tokens": 16419302.0,
      "step": 9030
    },
    {
      "entropy": 5.562041902542115,
      "epoch": 7.762354963472282,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011303567244704015,
      "loss": 4.8981,
      "mean_token_accuracy": 0.22844958752393724,
      "num_tokens": 16428020.0,
      "step": 9035
    },
    {
      "entropy": 5.62833080291748,
      "epoch": 7.766652342071336,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011280506027452502,
      "loss": 4.9626,
      "mean_token_accuracy": 0.21808964759111404,
      "num_tokens": 16438033.0,
      "step": 9040
    },
    {
      "entropy": 5.617523097991944,
      "epoch": 7.770949720670391,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001125748022709295,
      "loss": 4.8764,
      "mean_token_accuracy": 0.23195703476667404,
      "num_tokens": 16447067.0,
      "step": 9045
    },
    {
      "entropy": 5.618326997756958,
      "epoch": 7.775247099269445,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011234489893904509,
      "loss": 4.899,
      "mean_token_accuracy": 0.22247299402952195,
      "num_tokens": 16457146.0,
      "step": 9050
    },
    {
      "entropy": 5.554581642150879,
      "epoch": 7.7795444778685,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011211535078088869,
      "loss": 4.7486,
      "mean_token_accuracy": 0.23768365234136582,
      "num_tokens": 16466428.0,
      "step": 9055
    },
    {
      "entropy": 5.538989162445068,
      "epoch": 7.783841856467554,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011188615829770171,
      "loss": 4.8163,
      "mean_token_accuracy": 0.23359301090240478,
      "num_tokens": 16474198.0,
      "step": 9060
    },
    {
      "entropy": 5.538145923614502,
      "epoch": 7.788139235066609,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011165732198994905,
      "loss": 4.8531,
      "mean_token_accuracy": 0.23090007901191711,
      "num_tokens": 16483464.0,
      "step": 9065
    },
    {
      "entropy": 5.554736948013305,
      "epoch": 7.792436613665664,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011142884235731756,
      "loss": 4.8179,
      "mean_token_accuracy": 0.23584447503089906,
      "num_tokens": 16492619.0,
      "step": 9070
    },
    {
      "entropy": 5.619088459014892,
      "epoch": 7.796733992264718,
      "grad_norm": 1.734375,
      "learning_rate": 0.00011120071989871564,
      "loss": 4.9417,
      "mean_token_accuracy": 0.22552303969860077,
      "num_tokens": 16501690.0,
      "step": 9075
    },
    {
      "entropy": 5.5142174243927,
      "epoch": 7.8010313708637735,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011097295511227134,
      "loss": 4.8115,
      "mean_token_accuracy": 0.23689529448747634,
      "num_tokens": 16510158.0,
      "step": 9080
    },
    {
      "entropy": 5.5451208591461185,
      "epoch": 7.805328749462827,
      "grad_norm": 1.5,
      "learning_rate": 0.0001107455484953321,
      "loss": 4.8112,
      "mean_token_accuracy": 0.22746312767267227,
      "num_tokens": 16518722.0,
      "step": 9085
    },
    {
      "entropy": 5.512939929962158,
      "epoch": 7.809626128061883,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011051850054446306,
      "loss": 4.8328,
      "mean_token_accuracy": 0.22806360721588134,
      "num_tokens": 16527404.0,
      "step": 9090
    },
    {
      "entropy": 5.49378981590271,
      "epoch": 7.813923506660937,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00011029181175544603,
      "loss": 4.7566,
      "mean_token_accuracy": 0.24142335802316667,
      "num_tokens": 16536210.0,
      "step": 9095
    },
    {
      "entropy": 5.617502737045288,
      "epoch": 7.818220885259992,
      "grad_norm": 1.53125,
      "learning_rate": 0.00011006548262327884,
      "loss": 4.9121,
      "mean_token_accuracy": 0.21973130404949187,
      "num_tokens": 16544707.0,
      "step": 9100
    },
    {
      "entropy": 5.616050672531128,
      "epoch": 7.822518263859046,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001098395136421739,
      "loss": 4.8865,
      "mean_token_accuracy": 0.22247736901044846,
      "num_tokens": 16553883.0,
      "step": 9105
    },
    {
      "entropy": 5.574926900863647,
      "epoch": 7.826815642458101,
      "grad_norm": 1.3125,
      "learning_rate": 0.00010961390530555712,
      "loss": 4.8989,
      "mean_token_accuracy": 0.2185367614030838,
      "num_tokens": 16562537.0,
      "step": 9110
    },
    {
      "entropy": 5.580214214324951,
      "epoch": 7.831113021057155,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010938865810606682,
      "loss": 4.8258,
      "mean_token_accuracy": 0.23526800274848939,
      "num_tokens": 16571665.0,
      "step": 9115
    },
    {
      "entropy": 5.564577722549439,
      "epoch": 7.83541039965621,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010916377253555293,
      "loss": 4.8426,
      "mean_token_accuracy": 0.22370180785655974,
      "num_tokens": 16581102.0,
      "step": 9120
    },
    {
      "entropy": 5.557686614990234,
      "epoch": 7.839707778255264,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010893924908507573,
      "loss": 4.8077,
      "mean_token_accuracy": 0.2326553836464882,
      "num_tokens": 16589958.0,
      "step": 9125
    },
    {
      "entropy": 5.585665321350097,
      "epoch": 7.844005156854319,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001087150882449046,
      "loss": 4.8782,
      "mean_token_accuracy": 0.23088639080524445,
      "num_tokens": 16598800.0,
      "step": 9130
    },
    {
      "entropy": 5.580033445358277,
      "epoch": 7.8483025354533735,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010849129050451717,
      "loss": 4.8375,
      "mean_token_accuracy": 0.22930959463119507,
      "num_tokens": 16607751.0,
      "step": 9135
    },
    {
      "entropy": 5.525313949584961,
      "epoch": 7.852599914052428,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00010826785635259842,
      "loss": 4.7915,
      "mean_token_accuracy": 0.23290392458438874,
      "num_tokens": 16616041.0,
      "step": 9140
    },
    {
      "entropy": 5.519612550735474,
      "epoch": 7.856897292651483,
      "grad_norm": 1.5,
      "learning_rate": 0.00010804478627703903,
      "loss": 4.7592,
      "mean_token_accuracy": 0.24496115744113922,
      "num_tokens": 16624800.0,
      "step": 9145
    },
    {
      "entropy": 5.622063636779785,
      "epoch": 7.861194671250537,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00010782208076493508,
      "loss": 4.8763,
      "mean_token_accuracy": 0.22427395582199097,
      "num_tokens": 16632808.0,
      "step": 9150
    },
    {
      "entropy": 5.585298156738281,
      "epoch": 7.865492049849592,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010759974030258621,
      "loss": 4.8628,
      "mean_token_accuracy": 0.22683680802583694,
      "num_tokens": 16641179.0,
      "step": 9155
    },
    {
      "entropy": 5.5642650604248045,
      "epoch": 7.869789428448646,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010737776537549531,
      "loss": 4.8448,
      "mean_token_accuracy": 0.23744607716798782,
      "num_tokens": 16650402.0,
      "step": 9160
    },
    {
      "entropy": 5.496195459365845,
      "epoch": 7.874086807047701,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010715615646836679,
      "loss": 4.806,
      "mean_token_accuracy": 0.22923666536808013,
      "num_tokens": 16659661.0,
      "step": 9165
    },
    {
      "entropy": 5.503462553024292,
      "epoch": 7.878384185646755,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00010693491406510585,
      "loss": 4.8052,
      "mean_token_accuracy": 0.23879680037498474,
      "num_tokens": 16668630.0,
      "step": 9170
    },
    {
      "entropy": 5.577972650527954,
      "epoch": 7.88268156424581,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010671403864881757,
      "loss": 4.8793,
      "mean_token_accuracy": 0.22622952461242676,
      "num_tokens": 16678023.0,
      "step": 9175
    },
    {
      "entropy": 5.582621240615845,
      "epoch": 7.8869789428448644,
      "grad_norm": 1.359375,
      "learning_rate": 0.00010649353070180562,
      "loss": 4.8567,
      "mean_token_accuracy": 0.23242652118206025,
      "num_tokens": 16686751.0,
      "step": 9180
    },
    {
      "entropy": 5.554329776763916,
      "epoch": 7.891276321443919,
      "grad_norm": 1.375,
      "learning_rate": 0.00010627339070557118,
      "loss": 4.8432,
      "mean_token_accuracy": 0.23419446647167205,
      "num_tokens": 16696672.0,
      "step": 9185
    },
    {
      "entropy": 5.5002720832824705,
      "epoch": 7.8955737000429735,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010605361914081194,
      "loss": 4.7446,
      "mean_token_accuracy": 0.24070673137903215,
      "num_tokens": 16706018.0,
      "step": 9190
    },
    {
      "entropy": 5.589330959320068,
      "epoch": 7.899871078642028,
      "grad_norm": 1.296875,
      "learning_rate": 0.00010583421648742125,
      "loss": 4.7954,
      "mean_token_accuracy": 0.2352122813463211,
      "num_tokens": 16715206.0,
      "step": 9195
    },
    {
      "entropy": 5.572130298614502,
      "epoch": 7.904168457241083,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010561518322448673,
      "loss": 4.9278,
      "mean_token_accuracy": 0.22245595306158067,
      "num_tokens": 16724479.0,
      "step": 9200
    },
    {
      "entropy": 5.504499864578247,
      "epoch": 7.908465835840137,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00010539651983028955,
      "loss": 4.7518,
      "mean_token_accuracy": 0.24365971833467484,
      "num_tokens": 16733304.0,
      "step": 9205
    },
    {
      "entropy": 5.536966466903687,
      "epoch": 7.912763214439192,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001051782267823031,
      "loss": 4.7668,
      "mean_token_accuracy": 0.23533472418785095,
      "num_tokens": 16741447.0,
      "step": 9210
    },
    {
      "entropy": 5.531696462631226,
      "epoch": 7.917060593038247,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010496030455719225,
      "loss": 4.8224,
      "mean_token_accuracy": 0.22648018598556519,
      "num_tokens": 16751487.0,
      "step": 9215
    },
    {
      "entropy": 5.569997072219849,
      "epoch": 7.921357971637301,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010474275363081193,
      "loss": 4.8521,
      "mean_token_accuracy": 0.22589143216609955,
      "num_tokens": 16760795.0,
      "step": 9220
    },
    {
      "entropy": 5.5185792446136475,
      "epoch": 7.925655350236356,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001045255744782064,
      "loss": 4.8165,
      "mean_token_accuracy": 0.23142348378896713,
      "num_tokens": 16769639.0,
      "step": 9225
    },
    {
      "entropy": 5.5929947853088375,
      "epoch": 7.92995272883541,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010430876757360817,
      "loss": 4.8893,
      "mean_token_accuracy": 0.22292661517858506,
      "num_tokens": 16779195.0,
      "step": 9230
    },
    {
      "entropy": 5.4721239566802975,
      "epoch": 7.934250107434465,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010409233339043694,
      "loss": 4.7165,
      "mean_token_accuracy": 0.24348629266023636,
      "num_tokens": 16787531.0,
      "step": 9235
    },
    {
      "entropy": 5.481355285644531,
      "epoch": 7.93854748603352,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010387627240129838,
      "loss": 4.7775,
      "mean_token_accuracy": 0.23772444874048232,
      "num_tokens": 16796392.0,
      "step": 9240
    },
    {
      "entropy": 5.562499237060547,
      "epoch": 7.942844864632574,
      "grad_norm": 1.296875,
      "learning_rate": 0.00010366058507798326,
      "loss": 4.8255,
      "mean_token_accuracy": 0.23046404421329497,
      "num_tokens": 16804942.0,
      "step": 9245
    },
    {
      "entropy": 5.574090671539307,
      "epoch": 7.947142243231629,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010344527189146655,
      "loss": 4.8925,
      "mean_token_accuracy": 0.2183934599161148,
      "num_tokens": 16813754.0,
      "step": 9250
    },
    {
      "entropy": 5.5677299976348875,
      "epoch": 7.9514396218306835,
      "grad_norm": 1.265625,
      "learning_rate": 0.00010323033331190626,
      "loss": 4.8289,
      "mean_token_accuracy": 0.23214813619852065,
      "num_tokens": 16823010.0,
      "step": 9255
    },
    {
      "entropy": 5.5534709930419925,
      "epoch": 7.955737000429738,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010301576980864228,
      "loss": 4.7524,
      "mean_token_accuracy": 0.24431314766407014,
      "num_tokens": 16831909.0,
      "step": 9260
    },
    {
      "entropy": 5.482468748092652,
      "epoch": 7.960034379028793,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010280158185019547,
      "loss": 4.8071,
      "mean_token_accuracy": 0.23906526267528533,
      "num_tokens": 16841460.0,
      "step": 9265
    },
    {
      "entropy": 5.518536186218261,
      "epoch": 7.964331757627847,
      "grad_norm": 1.28125,
      "learning_rate": 0.00010258776990426686,
      "loss": 4.82,
      "mean_token_accuracy": 0.23808008134365083,
      "num_tokens": 16850592.0,
      "step": 9270
    },
    {
      "entropy": 5.536955404281616,
      "epoch": 7.968629136226902,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010237433443773612,
      "loss": 4.8367,
      "mean_token_accuracy": 0.23342074006795882,
      "num_tokens": 16859736.0,
      "step": 9275
    },
    {
      "entropy": 5.473064279556274,
      "epoch": 7.972926514825956,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00010216127591666115,
      "loss": 4.8316,
      "mean_token_accuracy": 0.23818496912717818,
      "num_tokens": 16870084.0,
      "step": 9280
    },
    {
      "entropy": 5.5625152587890625,
      "epoch": 7.977223893425011,
      "grad_norm": 1.640625,
      "learning_rate": 0.00010194859480627648,
      "loss": 4.8499,
      "mean_token_accuracy": 0.23696989715099334,
      "num_tokens": 16877771.0,
      "step": 9285
    },
    {
      "entropy": 5.532229614257813,
      "epoch": 7.981521272024065,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010173629157099279,
      "loss": 4.8632,
      "mean_token_accuracy": 0.23381411284208298,
      "num_tokens": 16887487.0,
      "step": 9290
    },
    {
      "entropy": 5.540496873855591,
      "epoch": 7.98581865062312,
      "grad_norm": 1.3125,
      "learning_rate": 0.00010152436667439537,
      "loss": 4.8649,
      "mean_token_accuracy": 0.23214023560285568,
      "num_tokens": 16897286.0,
      "step": 9295
    },
    {
      "entropy": 5.559723711013794,
      "epoch": 7.9901160292221745,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010131282057924345,
      "loss": 4.7738,
      "mean_token_accuracy": 0.236821249127388,
      "num_tokens": 16905968.0,
      "step": 9300
    },
    {
      "entropy": 5.597210121154785,
      "epoch": 7.994413407821229,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010110165374746924,
      "loss": 4.8118,
      "mean_token_accuracy": 0.23468527644872667,
      "num_tokens": 16914604.0,
      "step": 9305
    },
    {
      "entropy": 5.580416059494018,
      "epoch": 7.9987107864202835,
      "grad_norm": 1.359375,
      "learning_rate": 0.00010089086664017674,
      "loss": 4.9495,
      "mean_token_accuracy": 0.22792178094387056,
      "num_tokens": 16925085.0,
      "step": 9310
    },
    {
      "entropy": 5.569480843014187,
      "epoch": 8.002578427159433,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010068045971764067,
      "loss": 4.7624,
      "mean_token_accuracy": 0.24095100661118826,
      "num_tokens": 16932717.0,
      "step": 9315
    },
    {
      "entropy": 5.616175127029419,
      "epoch": 8.006875805758487,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010047043343930561,
      "loss": 4.8425,
      "mean_token_accuracy": 0.2277050420641899,
      "num_tokens": 16941332.0,
      "step": 9320
    },
    {
      "entropy": 5.499203252792358,
      "epoch": 8.011173184357542,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010026078826378502,
      "loss": 4.6707,
      "mean_token_accuracy": 0.240666164457798,
      "num_tokens": 16949732.0,
      "step": 9325
    },
    {
      "entropy": 5.610618162155151,
      "epoch": 8.015470562956596,
      "grad_norm": 1.359375,
      "learning_rate": 0.00010005152464886031,
      "loss": 4.7443,
      "mean_token_accuracy": 0.23739273250102996,
      "num_tokens": 16958013.0,
      "step": 9330
    },
    {
      "entropy": 5.51518611907959,
      "epoch": 8.019767941555651,
      "grad_norm": 1.3671875,
      "learning_rate": 9.984264305147941e-05,
      "loss": 4.6799,
      "mean_token_accuracy": 0.2413589432835579,
      "num_tokens": 16966050.0,
      "step": 9335
    },
    {
      "entropy": 5.502107381820679,
      "epoch": 8.024065320154705,
      "grad_norm": 1.390625,
      "learning_rate": 9.963414392775627e-05,
      "loss": 4.6908,
      "mean_token_accuracy": 0.24662898927927018,
      "num_tokens": 16975178.0,
      "step": 9340
    },
    {
      "entropy": 5.423726749420166,
      "epoch": 8.02836269875376,
      "grad_norm": 1.453125,
      "learning_rate": 9.942602773296971e-05,
      "loss": 4.6548,
      "mean_token_accuracy": 0.2561335489153862,
      "num_tokens": 16984247.0,
      "step": 9345
    },
    {
      "entropy": 5.593319082260132,
      "epoch": 8.032660077352816,
      "grad_norm": 1.40625,
      "learning_rate": 9.921829492156223e-05,
      "loss": 4.8429,
      "mean_token_accuracy": 0.2291508287191391,
      "num_tokens": 16995048.0,
      "step": 9350
    },
    {
      "entropy": 5.611334419250488,
      "epoch": 8.03695745595187,
      "grad_norm": 1.5859375,
      "learning_rate": 9.901094594713933e-05,
      "loss": 4.8282,
      "mean_token_accuracy": 0.23207295536994935,
      "num_tokens": 17003748.0,
      "step": 9355
    },
    {
      "entropy": 5.616546773910523,
      "epoch": 8.041254834550925,
      "grad_norm": 1.4296875,
      "learning_rate": 9.88039812624682e-05,
      "loss": 4.7982,
      "mean_token_accuracy": 0.2320023998618126,
      "num_tokens": 17011639.0,
      "step": 9360
    },
    {
      "entropy": 5.643500900268554,
      "epoch": 8.045552213149978,
      "grad_norm": 1.2578125,
      "learning_rate": 9.859740131947715e-05,
      "loss": 4.7798,
      "mean_token_accuracy": 0.23057130575180054,
      "num_tokens": 17021056.0,
      "step": 9365
    },
    {
      "entropy": 5.553992128372192,
      "epoch": 8.049849591749034,
      "grad_norm": 1.375,
      "learning_rate": 9.839120656925407e-05,
      "loss": 4.8048,
      "mean_token_accuracy": 0.22752827405929565,
      "num_tokens": 17030944.0,
      "step": 9370
    },
    {
      "entropy": 5.541428470611573,
      "epoch": 8.054146970348087,
      "grad_norm": 1.40625,
      "learning_rate": 9.818539746204588e-05,
      "loss": 4.7734,
      "mean_token_accuracy": 0.2304042398929596,
      "num_tokens": 17040127.0,
      "step": 9375
    },
    {
      "entropy": 5.601216650009155,
      "epoch": 8.058444348947143,
      "grad_norm": 1.578125,
      "learning_rate": 9.797997444725745e-05,
      "loss": 4.7617,
      "mean_token_accuracy": 0.23736438453197478,
      "num_tokens": 17049418.0,
      "step": 9380
    },
    {
      "entropy": 5.5397251605987545,
      "epoch": 8.062741727546197,
      "grad_norm": 1.359375,
      "learning_rate": 9.77749379734506e-05,
      "loss": 4.7712,
      "mean_token_accuracy": 0.2405427426099777,
      "num_tokens": 17058270.0,
      "step": 9385
    },
    {
      "entropy": 5.558532285690307,
      "epoch": 8.067039106145252,
      "grad_norm": 1.453125,
      "learning_rate": 9.757028848834293e-05,
      "loss": 4.7781,
      "mean_token_accuracy": 0.23632340580224992,
      "num_tokens": 17068011.0,
      "step": 9390
    },
    {
      "entropy": 5.545119667053223,
      "epoch": 8.071336484744306,
      "grad_norm": 1.3984375,
      "learning_rate": 9.736602643880712e-05,
      "loss": 4.7938,
      "mean_token_accuracy": 0.23605266958475113,
      "num_tokens": 17077356.0,
      "step": 9395
    },
    {
      "entropy": 5.5596250057220455,
      "epoch": 8.075633863343361,
      "grad_norm": 1.3515625,
      "learning_rate": 9.716215227086997e-05,
      "loss": 4.7007,
      "mean_token_accuracy": 0.24419404268264772,
      "num_tokens": 17085679.0,
      "step": 9400
    },
    {
      "entropy": 5.586064052581787,
      "epoch": 8.079931241942415,
      "grad_norm": 1.3359375,
      "learning_rate": 9.695866642971098e-05,
      "loss": 4.761,
      "mean_token_accuracy": 0.23189135044813156,
      "num_tokens": 17094925.0,
      "step": 9405
    },
    {
      "entropy": 5.556618738174438,
      "epoch": 8.08422862054147,
      "grad_norm": 1.3828125,
      "learning_rate": 9.67555693596621e-05,
      "loss": 4.8135,
      "mean_token_accuracy": 0.2339961513876915,
      "num_tokens": 17105278.0,
      "step": 9410
    },
    {
      "entropy": 5.507935237884522,
      "epoch": 8.088525999140524,
      "grad_norm": 1.375,
      "learning_rate": 9.655286150420595e-05,
      "loss": 4.7298,
      "mean_token_accuracy": 0.24389908611774444,
      "num_tokens": 17114070.0,
      "step": 9415
    },
    {
      "entropy": 5.435633993148803,
      "epoch": 8.09282337773958,
      "grad_norm": 1.3125,
      "learning_rate": 9.635054330597565e-05,
      "loss": 4.6532,
      "mean_token_accuracy": 0.25158399641513823,
      "num_tokens": 17122862.0,
      "step": 9420
    },
    {
      "entropy": 5.57655439376831,
      "epoch": 8.097120756338633,
      "grad_norm": 1.4921875,
      "learning_rate": 9.614861520675322e-05,
      "loss": 4.7495,
      "mean_token_accuracy": 0.24384848177433013,
      "num_tokens": 17131555.0,
      "step": 9425
    },
    {
      "entropy": 5.518671560287475,
      "epoch": 8.101418134937688,
      "grad_norm": 1.4375,
      "learning_rate": 9.594707764746881e-05,
      "loss": 4.6999,
      "mean_token_accuracy": 0.24201369881629944,
      "num_tokens": 17140841.0,
      "step": 9430
    },
    {
      "entropy": 5.52758412361145,
      "epoch": 8.105715513536742,
      "grad_norm": 1.3515625,
      "learning_rate": 9.57459310682e-05,
      "loss": 4.7778,
      "mean_token_accuracy": 0.2363356426358223,
      "num_tokens": 17150027.0,
      "step": 9435
    },
    {
      "entropy": 5.565421772003174,
      "epoch": 8.110012892135797,
      "grad_norm": 1.3828125,
      "learning_rate": 9.554517590817055e-05,
      "loss": 4.7933,
      "mean_token_accuracy": 0.22599780261516572,
      "num_tokens": 17159589.0,
      "step": 9440
    },
    {
      "entropy": 5.599954509735108,
      "epoch": 8.114310270734851,
      "grad_norm": 1.34375,
      "learning_rate": 9.534481260574944e-05,
      "loss": 4.8025,
      "mean_token_accuracy": 0.22512343525886536,
      "num_tokens": 17168219.0,
      "step": 9445
    },
    {
      "entropy": 5.579455947875976,
      "epoch": 8.118607649333907,
      "grad_norm": 1.3515625,
      "learning_rate": 9.514484159844997e-05,
      "loss": 4.7863,
      "mean_token_accuracy": 0.23330431133508683,
      "num_tokens": 17177364.0,
      "step": 9450
    },
    {
      "entropy": 5.618521499633789,
      "epoch": 8.12290502793296,
      "grad_norm": 1.46875,
      "learning_rate": 9.494526332292899e-05,
      "loss": 4.8253,
      "mean_token_accuracy": 0.22921546101570128,
      "num_tokens": 17186572.0,
      "step": 9455
    },
    {
      "entropy": 5.576432752609253,
      "epoch": 8.127202406532016,
      "grad_norm": 1.375,
      "learning_rate": 9.47460782149857e-05,
      "loss": 4.8148,
      "mean_token_accuracy": 0.23114224672317504,
      "num_tokens": 17195645.0,
      "step": 9460
    },
    {
      "entropy": 5.5786761283874515,
      "epoch": 8.13149978513107,
      "grad_norm": 1.3671875,
      "learning_rate": 9.454728670956073e-05,
      "loss": 4.7902,
      "mean_token_accuracy": 0.22771504372358323,
      "num_tokens": 17205279.0,
      "step": 9465
    },
    {
      "entropy": 5.485693502426147,
      "epoch": 8.135797163730125,
      "grad_norm": 1.4140625,
      "learning_rate": 9.43488892407352e-05,
      "loss": 4.7004,
      "mean_token_accuracy": 0.2360362872481346,
      "num_tokens": 17214536.0,
      "step": 9470
    },
    {
      "entropy": 5.619683456420899,
      "epoch": 8.140094542329178,
      "grad_norm": 1.375,
      "learning_rate": 9.415088624172997e-05,
      "loss": 4.8332,
      "mean_token_accuracy": 0.2294527620077133,
      "num_tokens": 17223336.0,
      "step": 9475
    },
    {
      "entropy": 5.583427238464355,
      "epoch": 8.144391920928234,
      "grad_norm": 1.3515625,
      "learning_rate": 9.395327814490439e-05,
      "loss": 4.8219,
      "mean_token_accuracy": 0.224343965947628,
      "num_tokens": 17232991.0,
      "step": 9480
    },
    {
      "entropy": 5.567170715332031,
      "epoch": 8.148689299527287,
      "grad_norm": 1.4609375,
      "learning_rate": 9.375606538175566e-05,
      "loss": 4.7292,
      "mean_token_accuracy": 0.2407409816980362,
      "num_tokens": 17241760.0,
      "step": 9485
    },
    {
      "entropy": 5.578453874588012,
      "epoch": 8.152986678126343,
      "grad_norm": 1.375,
      "learning_rate": 9.35592483829175e-05,
      "loss": 4.818,
      "mean_token_accuracy": 0.22747791707515716,
      "num_tokens": 17251514.0,
      "step": 9490
    },
    {
      "entropy": 5.553469657897949,
      "epoch": 8.157284056725398,
      "grad_norm": 1.3984375,
      "learning_rate": 9.336282757815964e-05,
      "loss": 4.768,
      "mean_token_accuracy": 0.23889153599739074,
      "num_tokens": 17260876.0,
      "step": 9495
    },
    {
      "entropy": 5.562651348114014,
      "epoch": 8.161581435324452,
      "grad_norm": 1.40625,
      "learning_rate": 9.316680339638664e-05,
      "loss": 4.8402,
      "mean_token_accuracy": 0.22985865026712418,
      "num_tokens": 17270051.0,
      "step": 9500
    },
    {
      "epoch": 8.161581435324452,
      "eval_entropy": 5.447511271850483,
      "eval_loss": 5.934919834136963,
      "eval_mean_token_accuracy": 0.17958645829679193,
      "eval_num_tokens": 17270051.0,
      "eval_runtime": 2.6439,
      "eval_samples_per_second": 1342.312,
      "eval_steps_per_second": 167.931,
      "step": 9500
    },
    {
      "entropy": 5.6030192375183105,
      "epoch": 8.165878813923507,
      "grad_norm": 1.5703125,
      "learning_rate": 9.297117626563687e-05,
      "loss": 4.8334,
      "mean_token_accuracy": 0.23247114419937134,
      "num_tokens": 17279744.0,
      "step": 9505
    },
    {
      "entropy": 5.53428430557251,
      "epoch": 8.170176192522561,
      "grad_norm": 1.4140625,
      "learning_rate": 9.27759466130818e-05,
      "loss": 4.7431,
      "mean_token_accuracy": 0.23856568783521653,
      "num_tokens": 17289107.0,
      "step": 9510
    },
    {
      "entropy": 5.494548416137695,
      "epoch": 8.174473571121617,
      "grad_norm": 1.3125,
      "learning_rate": 9.25811148650251e-05,
      "loss": 4.6571,
      "mean_token_accuracy": 0.2468874931335449,
      "num_tokens": 17298385.0,
      "step": 9515
    },
    {
      "entropy": 5.563962984085083,
      "epoch": 8.17877094972067,
      "grad_norm": 1.359375,
      "learning_rate": 9.238668144690133e-05,
      "loss": 4.7608,
      "mean_token_accuracy": 0.23405558466911316,
      "num_tokens": 17307759.0,
      "step": 9520
    },
    {
      "entropy": 5.52780122756958,
      "epoch": 8.183068328319726,
      "grad_norm": 1.3515625,
      "learning_rate": 9.219264678327527e-05,
      "loss": 4.7656,
      "mean_token_accuracy": 0.23638618886470794,
      "num_tokens": 17317135.0,
      "step": 9525
    },
    {
      "entropy": 5.507844161987305,
      "epoch": 8.18736570691878,
      "grad_norm": 1.3984375,
      "learning_rate": 9.199901129784121e-05,
      "loss": 4.7259,
      "mean_token_accuracy": 0.24346019327640533,
      "num_tokens": 17326950.0,
      "step": 9530
    },
    {
      "entropy": 5.646039342880249,
      "epoch": 8.191663085517835,
      "grad_norm": 1.3125,
      "learning_rate": 9.180577541342164e-05,
      "loss": 4.8643,
      "mean_token_accuracy": 0.22529096901416779,
      "num_tokens": 17335873.0,
      "step": 9535
    },
    {
      "entropy": 5.574686765670776,
      "epoch": 8.195960464116888,
      "grad_norm": 1.4921875,
      "learning_rate": 9.161293955196648e-05,
      "loss": 4.7614,
      "mean_token_accuracy": 0.23744102865457534,
      "num_tokens": 17344659.0,
      "step": 9540
    },
    {
      "entropy": 5.617691993713379,
      "epoch": 8.200257842715944,
      "grad_norm": 1.40625,
      "learning_rate": 9.142050413455214e-05,
      "loss": 4.8195,
      "mean_token_accuracy": 0.23108586966991423,
      "num_tokens": 17353274.0,
      "step": 9545
    },
    {
      "entropy": 5.574840879440307,
      "epoch": 8.204555221314997,
      "grad_norm": 1.296875,
      "learning_rate": 9.12284695813807e-05,
      "loss": 4.7958,
      "mean_token_accuracy": 0.2372947335243225,
      "num_tokens": 17362212.0,
      "step": 9550
    },
    {
      "entropy": 5.564355325698853,
      "epoch": 8.208852599914053,
      "grad_norm": 1.4609375,
      "learning_rate": 9.103683631177878e-05,
      "loss": 4.7609,
      "mean_token_accuracy": 0.24562520533800125,
      "num_tokens": 17370567.0,
      "step": 9555
    },
    {
      "entropy": 5.574023962020874,
      "epoch": 8.213149978513107,
      "grad_norm": 1.7109375,
      "learning_rate": 9.084560474419701e-05,
      "loss": 4.7501,
      "mean_token_accuracy": 0.23746486455202104,
      "num_tokens": 17378521.0,
      "step": 9560
    },
    {
      "entropy": 5.580052232742309,
      "epoch": 8.217447357112162,
      "grad_norm": 1.484375,
      "learning_rate": 9.065477529620852e-05,
      "loss": 4.7779,
      "mean_token_accuracy": 0.23159873932600022,
      "num_tokens": 17386988.0,
      "step": 9565
    },
    {
      "entropy": 5.554970407485962,
      "epoch": 8.221744735711216,
      "grad_norm": 1.53125,
      "learning_rate": 9.046434838450868e-05,
      "loss": 4.8039,
      "mean_token_accuracy": 0.234009450674057,
      "num_tokens": 17396317.0,
      "step": 9570
    },
    {
      "entropy": 5.620703172683716,
      "epoch": 8.226042114310271,
      "grad_norm": 1.484375,
      "learning_rate": 9.027432442491369e-05,
      "loss": 4.8885,
      "mean_token_accuracy": 0.21945073157548906,
      "num_tokens": 17405768.0,
      "step": 9575
    },
    {
      "entropy": 5.59485821723938,
      "epoch": 8.230339492909325,
      "grad_norm": 1.484375,
      "learning_rate": 9.008470383235991e-05,
      "loss": 4.8758,
      "mean_token_accuracy": 0.22929517924785614,
      "num_tokens": 17416098.0,
      "step": 9580
    },
    {
      "entropy": 5.542322635650635,
      "epoch": 8.23463687150838,
      "grad_norm": 1.46875,
      "learning_rate": 8.989548702090295e-05,
      "loss": 4.786,
      "mean_token_accuracy": 0.23930941820144652,
      "num_tokens": 17426091.0,
      "step": 9585
    },
    {
      "entropy": 5.517304944992065,
      "epoch": 8.238934250107434,
      "grad_norm": 1.4453125,
      "learning_rate": 8.970667440371676e-05,
      "loss": 4.7482,
      "mean_token_accuracy": 0.2393985003232956,
      "num_tokens": 17434956.0,
      "step": 9590
    },
    {
      "entropy": 5.565658760070801,
      "epoch": 8.24323162870649,
      "grad_norm": 1.40625,
      "learning_rate": 8.951826639309257e-05,
      "loss": 4.8112,
      "mean_token_accuracy": 0.23133782893419266,
      "num_tokens": 17443980.0,
      "step": 9595
    },
    {
      "entropy": 5.523622751235962,
      "epoch": 8.247529007305543,
      "grad_norm": 1.6015625,
      "learning_rate": 8.933026340043811e-05,
      "loss": 4.7493,
      "mean_token_accuracy": 0.23938990086317063,
      "num_tokens": 17452982.0,
      "step": 9600
    },
    {
      "entropy": 5.525299739837647,
      "epoch": 8.251826385904598,
      "grad_norm": 1.4609375,
      "learning_rate": 8.914266583627684e-05,
      "loss": 4.7163,
      "mean_token_accuracy": 0.24762487709522246,
      "num_tokens": 17461946.0,
      "step": 9605
    },
    {
      "entropy": 5.640941858291626,
      "epoch": 8.256123764503652,
      "grad_norm": 1.5703125,
      "learning_rate": 8.89554741102469e-05,
      "loss": 4.8142,
      "mean_token_accuracy": 0.2222455248236656,
      "num_tokens": 17470125.0,
      "step": 9610
    },
    {
      "entropy": 5.566991138458252,
      "epoch": 8.260421143102707,
      "grad_norm": 1.3984375,
      "learning_rate": 8.876868863110013e-05,
      "loss": 4.8039,
      "mean_token_accuracy": 0.2298445910215378,
      "num_tokens": 17479536.0,
      "step": 9615
    },
    {
      "entropy": 5.590273094177246,
      "epoch": 8.264718521701761,
      "grad_norm": 1.3671875,
      "learning_rate": 8.858230980670134e-05,
      "loss": 4.759,
      "mean_token_accuracy": 0.23215475231409072,
      "num_tokens": 17488709.0,
      "step": 9620
    },
    {
      "entropy": 5.570983743667602,
      "epoch": 8.269015900300817,
      "grad_norm": 1.3515625,
      "learning_rate": 8.839633804402747e-05,
      "loss": 4.8075,
      "mean_token_accuracy": 0.23217775970697402,
      "num_tokens": 17498196.0,
      "step": 9625
    },
    {
      "entropy": 5.577793741226197,
      "epoch": 8.27331327889987,
      "grad_norm": 1.5546875,
      "learning_rate": 8.821077374916647e-05,
      "loss": 4.7813,
      "mean_token_accuracy": 0.237832909822464,
      "num_tokens": 17507374.0,
      "step": 9630
    },
    {
      "entropy": 5.511423826217651,
      "epoch": 8.277610657498926,
      "grad_norm": 1.453125,
      "learning_rate": 8.802561732731654e-05,
      "loss": 4.7477,
      "mean_token_accuracy": 0.23412707895040513,
      "num_tokens": 17516163.0,
      "step": 9635
    },
    {
      "entropy": 5.622076559066772,
      "epoch": 8.281908036097981,
      "grad_norm": 1.3359375,
      "learning_rate": 8.784086918278534e-05,
      "loss": 4.8224,
      "mean_token_accuracy": 0.2291496589779854,
      "num_tokens": 17525567.0,
      "step": 9640
    },
    {
      "entropy": 5.493152332305908,
      "epoch": 8.286205414697035,
      "grad_norm": 1.3828125,
      "learning_rate": 8.765652971898908e-05,
      "loss": 4.6983,
      "mean_token_accuracy": 0.24549489021301268,
      "num_tokens": 17534433.0,
      "step": 9645
    },
    {
      "entropy": 5.578742265701294,
      "epoch": 8.29050279329609,
      "grad_norm": 1.515625,
      "learning_rate": 8.747259933845134e-05,
      "loss": 4.8314,
      "mean_token_accuracy": 0.24186994284391403,
      "num_tokens": 17542981.0,
      "step": 9650
    },
    {
      "entropy": 5.63447380065918,
      "epoch": 8.294800171895144,
      "grad_norm": 1.6015625,
      "learning_rate": 8.728907844280254e-05,
      "loss": 4.8364,
      "mean_token_accuracy": 0.2241244927048683,
      "num_tokens": 17552425.0,
      "step": 9655
    },
    {
      "entropy": 5.553133726119995,
      "epoch": 8.2990975504942,
      "grad_norm": 1.5078125,
      "learning_rate": 8.710596743277901e-05,
      "loss": 4.773,
      "mean_token_accuracy": 0.24261627048254014,
      "num_tokens": 17560954.0,
      "step": 9660
    },
    {
      "entropy": 5.546798944473267,
      "epoch": 8.303394929093253,
      "grad_norm": 1.4296875,
      "learning_rate": 8.692326670822207e-05,
      "loss": 4.721,
      "mean_token_accuracy": 0.24569681882858277,
      "num_tokens": 17569769.0,
      "step": 9665
    },
    {
      "entropy": 5.52406120300293,
      "epoch": 8.307692307692308,
      "grad_norm": 1.40625,
      "learning_rate": 8.6740976668077e-05,
      "loss": 4.7419,
      "mean_token_accuracy": 0.23903406411409378,
      "num_tokens": 17578875.0,
      "step": 9670
    },
    {
      "entropy": 5.499690818786621,
      "epoch": 8.311989686291362,
      "grad_norm": 1.5234375,
      "learning_rate": 8.655909771039239e-05,
      "loss": 4.7394,
      "mean_token_accuracy": 0.23793799579143524,
      "num_tokens": 17587556.0,
      "step": 9675
    },
    {
      "entropy": 5.592768716812134,
      "epoch": 8.316287064890417,
      "grad_norm": 1.671875,
      "learning_rate": 8.637763023231922e-05,
      "loss": 4.7941,
      "mean_token_accuracy": 0.234085750579834,
      "num_tokens": 17595077.0,
      "step": 9680
    },
    {
      "entropy": 5.464831256866455,
      "epoch": 8.320584443489471,
      "grad_norm": 1.4921875,
      "learning_rate": 8.619657463010987e-05,
      "loss": 4.6741,
      "mean_token_accuracy": 0.2478677436709404,
      "num_tokens": 17603864.0,
      "step": 9685
    },
    {
      "entropy": 5.619125652313232,
      "epoch": 8.324881822088527,
      "grad_norm": 1.5078125,
      "learning_rate": 8.60159312991175e-05,
      "loss": 4.8693,
      "mean_token_accuracy": 0.22640979439020156,
      "num_tokens": 17613570.0,
      "step": 9690
    },
    {
      "entropy": 5.595480489730835,
      "epoch": 8.32917920068758,
      "grad_norm": 1.3984375,
      "learning_rate": 8.583570063379487e-05,
      "loss": 4.7993,
      "mean_token_accuracy": 0.23202421516180038,
      "num_tokens": 17622352.0,
      "step": 9695
    },
    {
      "entropy": 5.546668863296508,
      "epoch": 8.333476579286636,
      "grad_norm": 1.46875,
      "learning_rate": 8.565588302769374e-05,
      "loss": 4.7406,
      "mean_token_accuracy": 0.23265265822410583,
      "num_tokens": 17631556.0,
      "step": 9700
    },
    {
      "entropy": 5.51851716041565,
      "epoch": 8.33777395788569,
      "grad_norm": 1.4921875,
      "learning_rate": 8.54764788734639e-05,
      "loss": 4.7511,
      "mean_token_accuracy": 0.2405464231967926,
      "num_tokens": 17641182.0,
      "step": 9705
    },
    {
      "entropy": 5.620053958892822,
      "epoch": 8.342071336484745,
      "grad_norm": 1.3984375,
      "learning_rate": 8.529748856285228e-05,
      "loss": 4.7553,
      "mean_token_accuracy": 0.23552594035863877,
      "num_tokens": 17650842.0,
      "step": 9710
    },
    {
      "entropy": 5.546793222427368,
      "epoch": 8.346368715083798,
      "grad_norm": 1.4375,
      "learning_rate": 8.511891248670217e-05,
      "loss": 4.7531,
      "mean_token_accuracy": 0.23959094732999803,
      "num_tokens": 17659963.0,
      "step": 9715
    },
    {
      "entropy": 5.610232830047607,
      "epoch": 8.350666093682854,
      "grad_norm": 1.65625,
      "learning_rate": 8.494075103495245e-05,
      "loss": 4.8155,
      "mean_token_accuracy": 0.23657935559749604,
      "num_tokens": 17668637.0,
      "step": 9720
    },
    {
      "entropy": 5.629941034317016,
      "epoch": 8.354963472281907,
      "grad_norm": 1.40625,
      "learning_rate": 8.476300459663643e-05,
      "loss": 4.801,
      "mean_token_accuracy": 0.22663483619689942,
      "num_tokens": 17678212.0,
      "step": 9725
    },
    {
      "entropy": 5.600242614746094,
      "epoch": 8.359260850880963,
      "grad_norm": 1.421875,
      "learning_rate": 8.458567355988123e-05,
      "loss": 4.8535,
      "mean_token_accuracy": 0.22737656235694886,
      "num_tokens": 17686766.0,
      "step": 9730
    },
    {
      "entropy": 5.51709361076355,
      "epoch": 8.363558229480017,
      "grad_norm": 1.3125,
      "learning_rate": 8.440875831190704e-05,
      "loss": 4.7641,
      "mean_token_accuracy": 0.23619571030139924,
      "num_tokens": 17696846.0,
      "step": 9735
    },
    {
      "entropy": 5.529401206970215,
      "epoch": 8.367855608079072,
      "grad_norm": 1.328125,
      "learning_rate": 8.423225923902608e-05,
      "loss": 4.754,
      "mean_token_accuracy": 0.23966118693351746,
      "num_tokens": 17705846.0,
      "step": 9740
    },
    {
      "entropy": 5.54409122467041,
      "epoch": 8.372152986678126,
      "grad_norm": 1.5,
      "learning_rate": 8.405617672664175e-05,
      "loss": 4.701,
      "mean_token_accuracy": 0.24875909239053726,
      "num_tokens": 17714255.0,
      "step": 9745
    },
    {
      "entropy": 5.619966840744018,
      "epoch": 8.376450365277181,
      "grad_norm": 1.359375,
      "learning_rate": 8.388051115924781e-05,
      "loss": 4.8728,
      "mean_token_accuracy": 0.22623682171106338,
      "num_tokens": 17724246.0,
      "step": 9750
    },
    {
      "entropy": 5.548094177246094,
      "epoch": 8.380747743876235,
      "grad_norm": 1.4140625,
      "learning_rate": 8.370526292042774e-05,
      "loss": 4.7526,
      "mean_token_accuracy": 0.24092641919851304,
      "num_tokens": 17733081.0,
      "step": 9755
    },
    {
      "entropy": 5.5072181701660154,
      "epoch": 8.38504512247529,
      "grad_norm": 1.6015625,
      "learning_rate": 8.353043239285357e-05,
      "loss": 4.7559,
      "mean_token_accuracy": 0.23757712692022323,
      "num_tokens": 17741220.0,
      "step": 9760
    },
    {
      "entropy": 5.561735343933106,
      "epoch": 8.389342501074344,
      "grad_norm": 1.5625,
      "learning_rate": 8.335601995828534e-05,
      "loss": 4.7815,
      "mean_token_accuracy": 0.2381587266921997,
      "num_tokens": 17750057.0,
      "step": 9765
    },
    {
      "entropy": 5.5584488868713375,
      "epoch": 8.3936398796734,
      "grad_norm": 1.34375,
      "learning_rate": 8.318202599757008e-05,
      "loss": 4.8051,
      "mean_token_accuracy": 0.23406596779823302,
      "num_tokens": 17759016.0,
      "step": 9770
    },
    {
      "entropy": 5.5196608543396,
      "epoch": 8.397937258272453,
      "grad_norm": 1.4453125,
      "learning_rate": 8.30084508906411e-05,
      "loss": 4.7107,
      "mean_token_accuracy": 0.23793317675590514,
      "num_tokens": 17767170.0,
      "step": 9775
    },
    {
      "entropy": 5.521137952804565,
      "epoch": 8.402234636871508,
      "grad_norm": 1.46875,
      "learning_rate": 8.283529501651698e-05,
      "loss": 4.7947,
      "mean_token_accuracy": 0.22975805401802063,
      "num_tokens": 17776508.0,
      "step": 9780
    },
    {
      "entropy": 5.52327880859375,
      "epoch": 8.406532015470564,
      "grad_norm": 1.3984375,
      "learning_rate": 8.266255875330095e-05,
      "loss": 4.7397,
      "mean_token_accuracy": 0.23804617971181868,
      "num_tokens": 17785494.0,
      "step": 9785
    },
    {
      "entropy": 5.536149787902832,
      "epoch": 8.410829394069617,
      "grad_norm": 1.4375,
      "learning_rate": 8.249024247817998e-05,
      "loss": 4.7927,
      "mean_token_accuracy": 0.2346429705619812,
      "num_tokens": 17795158.0,
      "step": 9790
    },
    {
      "entropy": 5.5473369598388675,
      "epoch": 8.415126772668673,
      "grad_norm": 1.2734375,
      "learning_rate": 8.231834656742402e-05,
      "loss": 4.7958,
      "mean_token_accuracy": 0.23063292801380159,
      "num_tokens": 17805130.0,
      "step": 9795
    },
    {
      "entropy": 5.579331111907959,
      "epoch": 8.419424151267727,
      "grad_norm": 1.46875,
      "learning_rate": 8.214687139638494e-05,
      "loss": 4.8002,
      "mean_token_accuracy": 0.23527741581201553,
      "num_tokens": 17813775.0,
      "step": 9800
    },
    {
      "entropy": 5.54561128616333,
      "epoch": 8.423721529866782,
      "grad_norm": 1.484375,
      "learning_rate": 8.197581733949597e-05,
      "loss": 4.7423,
      "mean_token_accuracy": 0.24413740485906602,
      "num_tokens": 17822599.0,
      "step": 9805
    },
    {
      "entropy": 5.604226303100586,
      "epoch": 8.428018908465836,
      "grad_norm": 1.5390625,
      "learning_rate": 8.180518477027083e-05,
      "loss": 4.7787,
      "mean_token_accuracy": 0.23703248351812362,
      "num_tokens": 17830742.0,
      "step": 9810
    },
    {
      "entropy": 5.449470853805542,
      "epoch": 8.432316287064891,
      "grad_norm": 1.390625,
      "learning_rate": 8.163497406130287e-05,
      "loss": 4.6679,
      "mean_token_accuracy": 0.24530456960201263,
      "num_tokens": 17840101.0,
      "step": 9815
    },
    {
      "entropy": 5.64233365058899,
      "epoch": 8.436613665663945,
      "grad_norm": 1.390625,
      "learning_rate": 8.146518558426426e-05,
      "loss": 4.8633,
      "mean_token_accuracy": 0.22387629449367524,
      "num_tokens": 17850464.0,
      "step": 9820
    },
    {
      "entropy": 5.5736902236938475,
      "epoch": 8.440911044263,
      "grad_norm": 1.5,
      "learning_rate": 8.129581970990507e-05,
      "loss": 4.7452,
      "mean_token_accuracy": 0.2469510778784752,
      "num_tokens": 17858792.0,
      "step": 9825
    },
    {
      "entropy": 5.5450379848480225,
      "epoch": 8.445208422862054,
      "grad_norm": 1.53125,
      "learning_rate": 8.11268768080528e-05,
      "loss": 4.7614,
      "mean_token_accuracy": 0.23657285869121553,
      "num_tokens": 17867225.0,
      "step": 9830
    },
    {
      "entropy": 5.610232591629028,
      "epoch": 8.44950580146111,
      "grad_norm": 1.5234375,
      "learning_rate": 8.09583572476111e-05,
      "loss": 4.8461,
      "mean_token_accuracy": 0.2263411745429039,
      "num_tokens": 17876423.0,
      "step": 9835
    },
    {
      "entropy": 5.563453197479248,
      "epoch": 8.453803180060163,
      "grad_norm": 1.6171875,
      "learning_rate": 8.079026139655946e-05,
      "loss": 4.8152,
      "mean_token_accuracy": 0.22774849832057953,
      "num_tokens": 17886847.0,
      "step": 9840
    },
    {
      "entropy": 5.482745265960693,
      "epoch": 8.458100558659218,
      "grad_norm": 1.390625,
      "learning_rate": 8.062258962195192e-05,
      "loss": 4.716,
      "mean_token_accuracy": 0.2414958268404007,
      "num_tokens": 17897395.0,
      "step": 9845
    },
    {
      "entropy": 5.551933002471924,
      "epoch": 8.462397937258272,
      "grad_norm": 1.5625,
      "learning_rate": 8.04553422899167e-05,
      "loss": 4.8128,
      "mean_token_accuracy": 0.2252075880765915,
      "num_tokens": 17906190.0,
      "step": 9850
    },
    {
      "entropy": 5.544550991058349,
      "epoch": 8.466695315857327,
      "grad_norm": 1.3359375,
      "learning_rate": 8.028851976565508e-05,
      "loss": 4.8377,
      "mean_token_accuracy": 0.23465893119573594,
      "num_tokens": 17916056.0,
      "step": 9855
    },
    {
      "entropy": 5.533039903640747,
      "epoch": 8.470992694456381,
      "grad_norm": 1.390625,
      "learning_rate": 8.01221224134408e-05,
      "loss": 4.7286,
      "mean_token_accuracy": 0.24108494967222213,
      "num_tokens": 17925790.0,
      "step": 9860
    },
    {
      "entropy": 5.5521807193756105,
      "epoch": 8.475290073055437,
      "grad_norm": 1.4609375,
      "learning_rate": 7.995615059661907e-05,
      "loss": 4.8077,
      "mean_token_accuracy": 0.22946373820304872,
      "num_tokens": 17935007.0,
      "step": 9865
    },
    {
      "entropy": 5.569910383224487,
      "epoch": 8.47958745165449,
      "grad_norm": 1.4140625,
      "learning_rate": 7.979060467760616e-05,
      "loss": 4.7987,
      "mean_token_accuracy": 0.23664053082466124,
      "num_tokens": 17943363.0,
      "step": 9870
    },
    {
      "entropy": 5.514348983764648,
      "epoch": 8.483884830253546,
      "grad_norm": 1.53125,
      "learning_rate": 7.962548501788811e-05,
      "loss": 4.7209,
      "mean_token_accuracy": 0.23829310685396193,
      "num_tokens": 17951667.0,
      "step": 9875
    },
    {
      "entropy": 5.54012451171875,
      "epoch": 8.4881822088526,
      "grad_norm": 1.5,
      "learning_rate": 7.946079197802028e-05,
      "loss": 4.7981,
      "mean_token_accuracy": 0.23005436211824418,
      "num_tokens": 17960920.0,
      "step": 9880
    },
    {
      "entropy": 5.592625713348388,
      "epoch": 8.492479587451655,
      "grad_norm": 1.5390625,
      "learning_rate": 7.929652591762636e-05,
      "loss": 4.8151,
      "mean_token_accuracy": 0.233868071436882,
      "num_tokens": 17969462.0,
      "step": 9885
    },
    {
      "entropy": 5.539914941787719,
      "epoch": 8.496776966050708,
      "grad_norm": 1.515625,
      "learning_rate": 7.913268719539785e-05,
      "loss": 4.8077,
      "mean_token_accuracy": 0.23866284042596816,
      "num_tokens": 17978608.0,
      "step": 9890
    },
    {
      "entropy": 5.544913053512573,
      "epoch": 8.501074344649764,
      "grad_norm": 1.3984375,
      "learning_rate": 7.896927616909304e-05,
      "loss": 4.6985,
      "mean_token_accuracy": 0.24125376343727112,
      "num_tokens": 17987563.0,
      "step": 9895
    },
    {
      "entropy": 5.574773597717285,
      "epoch": 8.505371723248818,
      "grad_norm": 1.3359375,
      "learning_rate": 7.880629319553623e-05,
      "loss": 4.7929,
      "mean_token_accuracy": 0.238761068880558,
      "num_tokens": 17996493.0,
      "step": 9900
    },
    {
      "entropy": 5.5586256980896,
      "epoch": 8.509669101847873,
      "grad_norm": 1.40625,
      "learning_rate": 7.864373863061717e-05,
      "loss": 4.7734,
      "mean_token_accuracy": 0.24194271117448807,
      "num_tokens": 18004578.0,
      "step": 9905
    },
    {
      "entropy": 5.542367792129516,
      "epoch": 8.513966480446927,
      "grad_norm": 1.546875,
      "learning_rate": 7.848161282929006e-05,
      "loss": 4.7847,
      "mean_token_accuracy": 0.2362925812602043,
      "num_tokens": 18013366.0,
      "step": 9910
    },
    {
      "entropy": 5.57535457611084,
      "epoch": 8.518263859045982,
      "grad_norm": 1.515625,
      "learning_rate": 7.831991614557274e-05,
      "loss": 4.7958,
      "mean_token_accuracy": 0.2271557107567787,
      "num_tokens": 18021718.0,
      "step": 9915
    },
    {
      "entropy": 5.566425228118897,
      "epoch": 8.522561237645036,
      "grad_norm": 1.46875,
      "learning_rate": 7.815864893254619e-05,
      "loss": 4.7994,
      "mean_token_accuracy": 0.2345098614692688,
      "num_tokens": 18030770.0,
      "step": 9920
    },
    {
      "entropy": 5.583288097381592,
      "epoch": 8.526858616244091,
      "grad_norm": 1.4453125,
      "learning_rate": 7.799781154235361e-05,
      "loss": 4.8246,
      "mean_token_accuracy": 0.23079057335853576,
      "num_tokens": 18039009.0,
      "step": 9925
    },
    {
      "entropy": 5.636148166656494,
      "epoch": 8.531155994843147,
      "grad_norm": 1.5859375,
      "learning_rate": 7.783740432619954e-05,
      "loss": 4.8571,
      "mean_token_accuracy": 0.2233716145157814,
      "num_tokens": 18046677.0,
      "step": 9930
    },
    {
      "entropy": 5.552960777282715,
      "epoch": 8.5354533734422,
      "grad_norm": 1.421875,
      "learning_rate": 7.767742763434922e-05,
      "loss": 4.8078,
      "mean_token_accuracy": 0.23056557029485703,
      "num_tokens": 18056772.0,
      "step": 9935
    },
    {
      "entropy": 5.497981166839599,
      "epoch": 8.539750752041256,
      "grad_norm": 1.40625,
      "learning_rate": 7.75178818161277e-05,
      "loss": 4.6964,
      "mean_token_accuracy": 0.235154028236866,
      "num_tokens": 18065387.0,
      "step": 9940
    },
    {
      "entropy": 5.575107192993164,
      "epoch": 8.54404813064031,
      "grad_norm": 1.515625,
      "learning_rate": 7.735876721991945e-05,
      "loss": 4.8563,
      "mean_token_accuracy": 0.22708459049463273,
      "num_tokens": 18074495.0,
      "step": 9945
    },
    {
      "entropy": 5.582318687438965,
      "epoch": 8.548345509239365,
      "grad_norm": 1.3203125,
      "learning_rate": 7.720008419316708e-05,
      "loss": 4.8231,
      "mean_token_accuracy": 0.2292675718665123,
      "num_tokens": 18083351.0,
      "step": 9950
    },
    {
      "entropy": 5.526089096069336,
      "epoch": 8.552642887838418,
      "grad_norm": 1.390625,
      "learning_rate": 7.704183308237089e-05,
      "loss": 4.7237,
      "mean_token_accuracy": 0.23848740011453629,
      "num_tokens": 18092632.0,
      "step": 9955
    },
    {
      "entropy": 5.498479557037354,
      "epoch": 8.556940266437474,
      "grad_norm": 1.359375,
      "learning_rate": 7.688401423308799e-05,
      "loss": 4.7123,
      "mean_token_accuracy": 0.24167543947696685,
      "num_tokens": 18102380.0,
      "step": 9960
    },
    {
      "entropy": 5.487370920181275,
      "epoch": 8.561237645036528,
      "grad_norm": 1.5234375,
      "learning_rate": 7.672662798993174e-05,
      "loss": 4.7094,
      "mean_token_accuracy": 0.24901142567396164,
      "num_tokens": 18111254.0,
      "step": 9965
    },
    {
      "entropy": 5.554504776000977,
      "epoch": 8.565535023635583,
      "grad_norm": 1.4140625,
      "learning_rate": 7.656967469657083e-05,
      "loss": 4.787,
      "mean_token_accuracy": 0.23393507301807404,
      "num_tokens": 18119998.0,
      "step": 9970
    },
    {
      "entropy": 5.532280969619751,
      "epoch": 8.569832402234637,
      "grad_norm": 1.46875,
      "learning_rate": 7.641315469572841e-05,
      "loss": 4.7217,
      "mean_token_accuracy": 0.2369483768939972,
      "num_tokens": 18129237.0,
      "step": 9975
    },
    {
      "entropy": 5.612370347976684,
      "epoch": 8.574129780833692,
      "grad_norm": 1.359375,
      "learning_rate": 7.625706832918172e-05,
      "loss": 4.8006,
      "mean_token_accuracy": 0.23784585744142533,
      "num_tokens": 18138440.0,
      "step": 9980
    },
    {
      "entropy": 5.584556293487549,
      "epoch": 8.578427159432746,
      "grad_norm": 1.5,
      "learning_rate": 7.610141593776091e-05,
      "loss": 4.8878,
      "mean_token_accuracy": 0.22648182064294814,
      "num_tokens": 18147712.0,
      "step": 9985
    },
    {
      "entropy": 5.583337593078613,
      "epoch": 8.582724538031801,
      "grad_norm": 1.5078125,
      "learning_rate": 7.59461978613486e-05,
      "loss": 4.7912,
      "mean_token_accuracy": 0.22834626585245132,
      "num_tokens": 18156710.0,
      "step": 9990
    },
    {
      "entropy": 5.543131160736084,
      "epoch": 8.587021916630855,
      "grad_norm": 1.546875,
      "learning_rate": 7.579141443887901e-05,
      "loss": 4.7399,
      "mean_token_accuracy": 0.23841555416584015,
      "num_tokens": 18165412.0,
      "step": 9995
    },
    {
      "entropy": 5.550899600982666,
      "epoch": 8.59131929522991,
      "grad_norm": 1.53125,
      "learning_rate": 7.563706600833737e-05,
      "loss": 4.8072,
      "mean_token_accuracy": 0.22851362377405166,
      "num_tokens": 18174415.0,
      "step": 10000
    },
    {
      "epoch": 8.59131929522991,
      "eval_entropy": 5.419305076470247,
      "eval_loss": 5.931044578552246,
      "eval_mean_token_accuracy": 0.17962968263875795,
      "eval_num_tokens": 18174415.0,
      "eval_runtime": 2.6277,
      "eval_samples_per_second": 1350.598,
      "eval_steps_per_second": 168.968,
      "step": 10000
    },
    {
      "entropy": 5.4642997741699215,
      "epoch": 8.595616673828964,
      "grad_norm": 1.578125,
      "learning_rate": 7.548315290675886e-05,
      "loss": 4.7207,
      "mean_token_accuracy": 0.2390255317091942,
      "num_tokens": 18183468.0,
      "step": 10005
    },
    {
      "entropy": 5.506887912750244,
      "epoch": 8.59991405242802,
      "grad_norm": 1.3828125,
      "learning_rate": 7.532967547022825e-05,
      "loss": 4.7303,
      "mean_token_accuracy": 0.24207512736320497,
      "num_tokens": 18192552.0,
      "step": 10010
    },
    {
      "entropy": 5.555754041671753,
      "epoch": 8.604211431027073,
      "grad_norm": 1.5234375,
      "learning_rate": 7.517663403387874e-05,
      "loss": 4.8131,
      "mean_token_accuracy": 0.2332708477973938,
      "num_tokens": 18202056.0,
      "step": 10015
    },
    {
      "entropy": 5.52849760055542,
      "epoch": 8.608508809626128,
      "grad_norm": 1.484375,
      "learning_rate": 7.502402893189191e-05,
      "loss": 4.6941,
      "mean_token_accuracy": 0.24236228168010712,
      "num_tokens": 18210329.0,
      "step": 10020
    },
    {
      "entropy": 5.624890518188477,
      "epoch": 8.612806188225182,
      "grad_norm": 1.3203125,
      "learning_rate": 7.487186049749618e-05,
      "loss": 4.8229,
      "mean_token_accuracy": 0.22929491698741913,
      "num_tokens": 18219143.0,
      "step": 10025
    },
    {
      "entropy": 5.559404802322388,
      "epoch": 8.617103566824238,
      "grad_norm": 1.4296875,
      "learning_rate": 7.472012906296658e-05,
      "loss": 4.782,
      "mean_token_accuracy": 0.235048146545887,
      "num_tokens": 18227679.0,
      "step": 10030
    },
    {
      "entropy": 5.507843112945556,
      "epoch": 8.621400945423291,
      "grad_norm": 1.2890625,
      "learning_rate": 7.45688349596239e-05,
      "loss": 4.731,
      "mean_token_accuracy": 0.2376631513237953,
      "num_tokens": 18237522.0,
      "step": 10035
    },
    {
      "entropy": 5.561053037643433,
      "epoch": 8.625698324022347,
      "grad_norm": 1.4609375,
      "learning_rate": 7.441797851783402e-05,
      "loss": 4.7993,
      "mean_token_accuracy": 0.2326131835579872,
      "num_tokens": 18245969.0,
      "step": 10040
    },
    {
      "entropy": 5.542596960067749,
      "epoch": 8.6299957026214,
      "grad_norm": 1.484375,
      "learning_rate": 7.426756006700716e-05,
      "loss": 4.7428,
      "mean_token_accuracy": 0.2408544361591339,
      "num_tokens": 18254886.0,
      "step": 10045
    },
    {
      "entropy": 5.580687236785889,
      "epoch": 8.634293081220456,
      "grad_norm": 1.5546875,
      "learning_rate": 7.411757993559702e-05,
      "loss": 4.8962,
      "mean_token_accuracy": 0.23125398606061937,
      "num_tokens": 18264222.0,
      "step": 10050
    },
    {
      "entropy": 5.605154037475586,
      "epoch": 8.63859045981951,
      "grad_norm": 1.46875,
      "learning_rate": 7.396803845110032e-05,
      "loss": 4.8079,
      "mean_token_accuracy": 0.23628308922052382,
      "num_tokens": 18273548.0,
      "step": 10055
    },
    {
      "entropy": 5.561076211929321,
      "epoch": 8.642887838418565,
      "grad_norm": 1.484375,
      "learning_rate": 7.381893594005585e-05,
      "loss": 4.7711,
      "mean_token_accuracy": 0.2374502331018448,
      "num_tokens": 18282842.0,
      "step": 10060
    },
    {
      "entropy": 5.581733512878418,
      "epoch": 8.647185217017618,
      "grad_norm": 1.4375,
      "learning_rate": 7.367027272804387e-05,
      "loss": 4.7522,
      "mean_token_accuracy": 0.23518631607294083,
      "num_tokens": 18291133.0,
      "step": 10065
    },
    {
      "entropy": 5.5557349681854244,
      "epoch": 8.651482595616674,
      "grad_norm": 1.40625,
      "learning_rate": 7.352204913968546e-05,
      "loss": 4.8062,
      "mean_token_accuracy": 0.24141866862773895,
      "num_tokens": 18300936.0,
      "step": 10070
    },
    {
      "entropy": 5.583840274810791,
      "epoch": 8.65577997421573,
      "grad_norm": 1.3046875,
      "learning_rate": 7.337426549864175e-05,
      "loss": 4.8232,
      "mean_token_accuracy": 0.23305517584085464,
      "num_tokens": 18311372.0,
      "step": 10075
    },
    {
      "entropy": 5.577575635910034,
      "epoch": 8.660077352814783,
      "grad_norm": 1.3828125,
      "learning_rate": 7.322692212761306e-05,
      "loss": 4.8428,
      "mean_token_accuracy": 0.23797387927770614,
      "num_tokens": 18321166.0,
      "step": 10080
    },
    {
      "entropy": 5.581952047348023,
      "epoch": 8.664374731413837,
      "grad_norm": 1.5859375,
      "learning_rate": 7.308001934833844e-05,
      "loss": 4.7919,
      "mean_token_accuracy": 0.23726631402969361,
      "num_tokens": 18330399.0,
      "step": 10085
    },
    {
      "entropy": 5.587984275817871,
      "epoch": 8.668672110012892,
      "grad_norm": 1.46875,
      "learning_rate": 7.29335574815948e-05,
      "loss": 4.7734,
      "mean_token_accuracy": 0.23511392921209334,
      "num_tokens": 18338920.0,
      "step": 10090
    },
    {
      "entropy": 5.533380937576294,
      "epoch": 8.672969488611947,
      "grad_norm": 1.328125,
      "learning_rate": 7.278753684719636e-05,
      "loss": 4.7906,
      "mean_token_accuracy": 0.2325812667608261,
      "num_tokens": 18348515.0,
      "step": 10095
    },
    {
      "entropy": 5.553415250778198,
      "epoch": 8.677266867211001,
      "grad_norm": 1.515625,
      "learning_rate": 7.264195776399386e-05,
      "loss": 4.8026,
      "mean_token_accuracy": 0.22855228781700135,
      "num_tokens": 18357353.0,
      "step": 10100
    },
    {
      "entropy": 5.542114925384522,
      "epoch": 8.681564245810057,
      "grad_norm": 1.4375,
      "learning_rate": 7.249682054987381e-05,
      "loss": 4.7188,
      "mean_token_accuracy": 0.24218224883079528,
      "num_tokens": 18366434.0,
      "step": 10105
    },
    {
      "entropy": 5.5978456974029545,
      "epoch": 8.68586162440911,
      "grad_norm": 1.3828125,
      "learning_rate": 7.23521255217578e-05,
      "loss": 4.8673,
      "mean_token_accuracy": 0.21942984461784362,
      "num_tokens": 18375589.0,
      "step": 10110
    },
    {
      "entropy": 5.548715257644654,
      "epoch": 8.690159003008166,
      "grad_norm": 1.3984375,
      "learning_rate": 7.220787299560205e-05,
      "loss": 4.7976,
      "mean_token_accuracy": 0.22914974987506867,
      "num_tokens": 18384904.0,
      "step": 10115
    },
    {
      "entropy": 5.505613470077515,
      "epoch": 8.69445638160722,
      "grad_norm": 1.46875,
      "learning_rate": 7.20640632863963e-05,
      "loss": 4.6389,
      "mean_token_accuracy": 0.2520404323935509,
      "num_tokens": 18393833.0,
      "step": 10120
    },
    {
      "entropy": 5.558655595779419,
      "epoch": 8.698753760206275,
      "grad_norm": 1.4609375,
      "learning_rate": 7.192069670816359e-05,
      "loss": 4.8435,
      "mean_token_accuracy": 0.22806500047445297,
      "num_tokens": 18403444.0,
      "step": 10125
    },
    {
      "entropy": 5.520944976806641,
      "epoch": 8.703051138805328,
      "grad_norm": 1.5546875,
      "learning_rate": 7.177777357395912e-05,
      "loss": 4.7596,
      "mean_token_accuracy": 0.237407124042511,
      "num_tokens": 18412091.0,
      "step": 10130
    },
    {
      "entropy": 5.538614940643311,
      "epoch": 8.707348517404384,
      "grad_norm": 1.5,
      "learning_rate": 7.163529419587002e-05,
      "loss": 4.765,
      "mean_token_accuracy": 0.23865150511264802,
      "num_tokens": 18421655.0,
      "step": 10135
    },
    {
      "entropy": 5.622085475921631,
      "epoch": 8.711645896003438,
      "grad_norm": 1.4375,
      "learning_rate": 7.149325888501418e-05,
      "loss": 4.8524,
      "mean_token_accuracy": 0.22857151180505753,
      "num_tokens": 18431239.0,
      "step": 10140
    },
    {
      "entropy": 5.549133062362671,
      "epoch": 8.715943274602493,
      "grad_norm": 1.34375,
      "learning_rate": 7.135166795153992e-05,
      "loss": 4.7368,
      "mean_token_accuracy": 0.24104002565145494,
      "num_tokens": 18440572.0,
      "step": 10145
    },
    {
      "entropy": 5.600282621383667,
      "epoch": 8.720240653201547,
      "grad_norm": 1.46875,
      "learning_rate": 7.121052170462541e-05,
      "loss": 4.8425,
      "mean_token_accuracy": 0.23224472105503083,
      "num_tokens": 18449245.0,
      "step": 10150
    },
    {
      "entropy": 5.495789575576782,
      "epoch": 8.724538031800602,
      "grad_norm": 1.5859375,
      "learning_rate": 7.106982045247754e-05,
      "loss": 4.6717,
      "mean_token_accuracy": 0.2571649789810181,
      "num_tokens": 18457961.0,
      "step": 10155
    },
    {
      "entropy": 5.536230659484863,
      "epoch": 8.728835410399656,
      "grad_norm": 1.375,
      "learning_rate": 7.092956450233162e-05,
      "loss": 4.8305,
      "mean_token_accuracy": 0.23536324203014375,
      "num_tokens": 18467956.0,
      "step": 10160
    },
    {
      "entropy": 5.5926182746887205,
      "epoch": 8.733132788998711,
      "grad_norm": 1.3828125,
      "learning_rate": 7.078975416045055e-05,
      "loss": 4.7763,
      "mean_token_accuracy": 0.23370456397533418,
      "num_tokens": 18477313.0,
      "step": 10165
    },
    {
      "entropy": 5.516347503662109,
      "epoch": 8.737430167597765,
      "grad_norm": 1.5625,
      "learning_rate": 7.065038973212424e-05,
      "loss": 4.6522,
      "mean_token_accuracy": 0.24469332844018937,
      "num_tokens": 18485690.0,
      "step": 10170
    },
    {
      "entropy": 5.541161823272705,
      "epoch": 8.74172754619682,
      "grad_norm": 1.5,
      "learning_rate": 7.051147152166896e-05,
      "loss": 4.7915,
      "mean_token_accuracy": 0.2405724734067917,
      "num_tokens": 18494150.0,
      "step": 10175
    },
    {
      "entropy": 5.472743940353394,
      "epoch": 8.746024924795874,
      "grad_norm": 1.25,
      "learning_rate": 7.037299983242652e-05,
      "loss": 4.7229,
      "mean_token_accuracy": 0.24257556200027466,
      "num_tokens": 18504091.0,
      "step": 10180
    },
    {
      "entropy": 5.5226599216461185,
      "epoch": 8.75032230339493,
      "grad_norm": 1.4140625,
      "learning_rate": 7.023497496676371e-05,
      "loss": 4.799,
      "mean_token_accuracy": 0.23655738979578017,
      "num_tokens": 18513695.0,
      "step": 10185
    },
    {
      "entropy": 5.548027229309082,
      "epoch": 8.754619681993983,
      "grad_norm": 1.3203125,
      "learning_rate": 7.009739722607173e-05,
      "loss": 4.7388,
      "mean_token_accuracy": 0.24277422130107879,
      "num_tokens": 18522551.0,
      "step": 10190
    },
    {
      "entropy": 5.577977514266967,
      "epoch": 8.758917060593038,
      "grad_norm": 1.421875,
      "learning_rate": 6.996026691076531e-05,
      "loss": 4.8985,
      "mean_token_accuracy": 0.22043292075395585,
      "num_tokens": 18533290.0,
      "step": 10195
    },
    {
      "entropy": 5.539222717285156,
      "epoch": 8.763214439192092,
      "grad_norm": 1.3515625,
      "learning_rate": 6.982358432028234e-05,
      "loss": 4.682,
      "mean_token_accuracy": 0.24007729440927505,
      "num_tokens": 18542990.0,
      "step": 10200
    },
    {
      "entropy": 5.65463809967041,
      "epoch": 8.767511817791148,
      "grad_norm": 1.4296875,
      "learning_rate": 6.968734975308283e-05,
      "loss": 4.8972,
      "mean_token_accuracy": 0.21504464298486708,
      "num_tokens": 18552525.0,
      "step": 10205
    },
    {
      "entropy": 5.507696342468262,
      "epoch": 8.771809196390201,
      "grad_norm": 1.421875,
      "learning_rate": 6.955156350664876e-05,
      "loss": 4.6976,
      "mean_token_accuracy": 0.2437577575445175,
      "num_tokens": 18561741.0,
      "step": 10210
    },
    {
      "entropy": 5.604904890060425,
      "epoch": 8.776106574989257,
      "grad_norm": 1.46875,
      "learning_rate": 6.941622587748298e-05,
      "loss": 4.8435,
      "mean_token_accuracy": 0.22804163992404938,
      "num_tokens": 18570130.0,
      "step": 10215
    },
    {
      "entropy": 5.508373355865478,
      "epoch": 8.780403953588312,
      "grad_norm": 1.5234375,
      "learning_rate": 6.928133716110863e-05,
      "loss": 4.7897,
      "mean_token_accuracy": 0.23433953821659087,
      "num_tokens": 18579149.0,
      "step": 10220
    },
    {
      "entropy": 5.538003396987915,
      "epoch": 8.784701332187366,
      "grad_norm": 1.4921875,
      "learning_rate": 6.914689765206899e-05,
      "loss": 4.6884,
      "mean_token_accuracy": 0.24082439243793488,
      "num_tokens": 18587604.0,
      "step": 10225
    },
    {
      "entropy": 5.583019971847534,
      "epoch": 8.78899871078642,
      "grad_norm": 1.5,
      "learning_rate": 6.901290764392609e-05,
      "loss": 4.7757,
      "mean_token_accuracy": 0.23450582027435302,
      "num_tokens": 18596056.0,
      "step": 10230
    },
    {
      "entropy": 5.5406513690948485,
      "epoch": 8.793296089385475,
      "grad_norm": 1.234375,
      "learning_rate": 6.887936742926058e-05,
      "loss": 4.8029,
      "mean_token_accuracy": 0.2300175353884697,
      "num_tokens": 18605575.0,
      "step": 10235
    },
    {
      "entropy": 5.516643142700195,
      "epoch": 8.79759346798453,
      "grad_norm": 1.40625,
      "learning_rate": 6.874627729967086e-05,
      "loss": 4.7224,
      "mean_token_accuracy": 0.24312316179275512,
      "num_tokens": 18614206.0,
      "step": 10240
    },
    {
      "entropy": 5.550478887557984,
      "epoch": 8.801890846583584,
      "grad_norm": 1.3125,
      "learning_rate": 6.861363754577262e-05,
      "loss": 4.7514,
      "mean_token_accuracy": 0.23815749138593673,
      "num_tokens": 18623067.0,
      "step": 10245
    },
    {
      "entropy": 5.483479261398315,
      "epoch": 8.80618822518264,
      "grad_norm": 1.5,
      "learning_rate": 6.848144845719808e-05,
      "loss": 4.724,
      "mean_token_accuracy": 0.24007823318243027,
      "num_tokens": 18632266.0,
      "step": 10250
    },
    {
      "entropy": 5.500712871551514,
      "epoch": 8.810485603781693,
      "grad_norm": 1.46875,
      "learning_rate": 6.834971032259537e-05,
      "loss": 4.721,
      "mean_token_accuracy": 0.24015006572008132,
      "num_tokens": 18640516.0,
      "step": 10255
    },
    {
      "entropy": 5.541616773605346,
      "epoch": 8.814782982380748,
      "grad_norm": 1.5,
      "learning_rate": 6.821842342962786e-05,
      "loss": 4.7504,
      "mean_token_accuracy": 0.244784015417099,
      "num_tokens": 18649368.0,
      "step": 10260
    },
    {
      "entropy": 5.504744291305542,
      "epoch": 8.819080360979802,
      "grad_norm": 1.453125,
      "learning_rate": 6.808758806497375e-05,
      "loss": 4.7333,
      "mean_token_accuracy": 0.2442265957593918,
      "num_tokens": 18658306.0,
      "step": 10265
    },
    {
      "entropy": 5.542925691604614,
      "epoch": 8.823377739578858,
      "grad_norm": 1.4609375,
      "learning_rate": 6.795720451432509e-05,
      "loss": 4.7234,
      "mean_token_accuracy": 0.24485713094472886,
      "num_tokens": 18668676.0,
      "step": 10270
    },
    {
      "entropy": 5.561149978637696,
      "epoch": 8.827675118177911,
      "grad_norm": 1.3984375,
      "learning_rate": 6.782727306238749e-05,
      "loss": 4.8725,
      "mean_token_accuracy": 0.22890693992376326,
      "num_tokens": 18677840.0,
      "step": 10275
    },
    {
      "entropy": 5.474625778198242,
      "epoch": 8.831972496776967,
      "grad_norm": 1.5078125,
      "learning_rate": 6.769779399287928e-05,
      "loss": 4.6831,
      "mean_token_accuracy": 0.23624491095542907,
      "num_tokens": 18685733.0,
      "step": 10280
    },
    {
      "entropy": 5.5367601871490475,
      "epoch": 8.83626987537602,
      "grad_norm": 1.3671875,
      "learning_rate": 6.756876758853104e-05,
      "loss": 4.7806,
      "mean_token_accuracy": 0.23651949912309647,
      "num_tokens": 18694648.0,
      "step": 10285
    },
    {
      "entropy": 5.5252786636352536,
      "epoch": 8.840567253975076,
      "grad_norm": 1.640625,
      "learning_rate": 6.744019413108486e-05,
      "loss": 4.7692,
      "mean_token_accuracy": 0.23644654005765914,
      "num_tokens": 18702758.0,
      "step": 10290
    },
    {
      "entropy": 5.581271696090698,
      "epoch": 8.84486463257413,
      "grad_norm": 1.484375,
      "learning_rate": 6.731207390129366e-05,
      "loss": 4.8254,
      "mean_token_accuracy": 0.23393062949180604,
      "num_tokens": 18712257.0,
      "step": 10295
    },
    {
      "entropy": 5.5495195388793945,
      "epoch": 8.849162011173185,
      "grad_norm": 1.3671875,
      "learning_rate": 6.7184407178921e-05,
      "loss": 4.7887,
      "mean_token_accuracy": 0.23641925901174546,
      "num_tokens": 18722337.0,
      "step": 10300
    },
    {
      "entropy": 5.611654329299927,
      "epoch": 8.853459389772238,
      "grad_norm": 1.5390625,
      "learning_rate": 6.70571942427399e-05,
      "loss": 4.8409,
      "mean_token_accuracy": 0.23058146685361863,
      "num_tokens": 18731260.0,
      "step": 10305
    },
    {
      "entropy": 5.61999192237854,
      "epoch": 8.857756768371294,
      "grad_norm": 1.5,
      "learning_rate": 6.693043537053254e-05,
      "loss": 4.873,
      "mean_token_accuracy": 0.2283632665872574,
      "num_tokens": 18741357.0,
      "step": 10310
    },
    {
      "entropy": 5.534827661514282,
      "epoch": 8.862054146970348,
      "grad_norm": 1.4296875,
      "learning_rate": 6.68041308390896e-05,
      "loss": 4.8224,
      "mean_token_accuracy": 0.23197623193264008,
      "num_tokens": 18751694.0,
      "step": 10315
    },
    {
      "entropy": 5.604840326309204,
      "epoch": 8.866351525569403,
      "grad_norm": 1.3671875,
      "learning_rate": 6.667828092420977e-05,
      "loss": 4.8255,
      "mean_token_accuracy": 0.23734348118305207,
      "num_tokens": 18760679.0,
      "step": 10320
    },
    {
      "entropy": 5.529338407516479,
      "epoch": 8.870648904168457,
      "grad_norm": 1.390625,
      "learning_rate": 6.655288590069889e-05,
      "loss": 4.7709,
      "mean_token_accuracy": 0.23920615911483764,
      "num_tokens": 18769650.0,
      "step": 10325
    },
    {
      "entropy": 5.487248182296753,
      "epoch": 8.874946282767512,
      "grad_norm": 1.390625,
      "learning_rate": 6.642794604236965e-05,
      "loss": 4.7831,
      "mean_token_accuracy": 0.23692469149827958,
      "num_tokens": 18780076.0,
      "step": 10330
    },
    {
      "entropy": 5.535627222061157,
      "epoch": 8.879243661366566,
      "grad_norm": 1.3984375,
      "learning_rate": 6.630346162204069e-05,
      "loss": 4.7387,
      "mean_token_accuracy": 0.23790280818939208,
      "num_tokens": 18789034.0,
      "step": 10335
    },
    {
      "entropy": 5.498855018615723,
      "epoch": 8.883541039965621,
      "grad_norm": 1.5390625,
      "learning_rate": 6.617943291153631e-05,
      "loss": 4.6558,
      "mean_token_accuracy": 0.2464265063405037,
      "num_tokens": 18797916.0,
      "step": 10340
    },
    {
      "entropy": 5.603287553787231,
      "epoch": 8.887838418564675,
      "grad_norm": 1.5859375,
      "learning_rate": 6.605586018168558e-05,
      "loss": 4.7784,
      "mean_token_accuracy": 0.23790173381567,
      "num_tokens": 18806895.0,
      "step": 10345
    },
    {
      "entropy": 5.585084724426269,
      "epoch": 8.89213579716373,
      "grad_norm": 1.34375,
      "learning_rate": 6.593274370232191e-05,
      "loss": 4.7834,
      "mean_token_accuracy": 0.23531922698020935,
      "num_tokens": 18816049.0,
      "step": 10350
    },
    {
      "entropy": 5.528953647613525,
      "epoch": 8.896433175762784,
      "grad_norm": 1.4453125,
      "learning_rate": 6.581008374228255e-05,
      "loss": 4.7579,
      "mean_token_accuracy": 0.2362537294626236,
      "num_tokens": 18825491.0,
      "step": 10355
    },
    {
      "entropy": 5.550234937667847,
      "epoch": 8.90073055436184,
      "grad_norm": 1.4765625,
      "learning_rate": 6.568788056940785e-05,
      "loss": 4.7965,
      "mean_token_accuracy": 0.23366497606039047,
      "num_tokens": 18834677.0,
      "step": 10360
    },
    {
      "entropy": 5.481243133544922,
      "epoch": 8.905027932960895,
      "grad_norm": 1.40625,
      "learning_rate": 6.556613445054063e-05,
      "loss": 4.6965,
      "mean_token_accuracy": 0.24398072361946105,
      "num_tokens": 18844283.0,
      "step": 10365
    },
    {
      "entropy": 5.558114814758301,
      "epoch": 8.909325311559948,
      "grad_norm": 1.390625,
      "learning_rate": 6.544484565152577e-05,
      "loss": 4.7544,
      "mean_token_accuracy": 0.24853970110416412,
      "num_tokens": 18853403.0,
      "step": 10370
    },
    {
      "entropy": 5.535591125488281,
      "epoch": 8.913622690159002,
      "grad_norm": 1.609375,
      "learning_rate": 6.532401443720951e-05,
      "loss": 4.7667,
      "mean_token_accuracy": 0.23787880688905716,
      "num_tokens": 18861530.0,
      "step": 10375
    },
    {
      "entropy": 5.5535601615905765,
      "epoch": 8.917920068758058,
      "grad_norm": 1.2890625,
      "learning_rate": 6.520364107143898e-05,
      "loss": 4.7894,
      "mean_token_accuracy": 0.2410448521375656,
      "num_tokens": 18871495.0,
      "step": 10380
    },
    {
      "entropy": 5.573381567001343,
      "epoch": 8.922217447357113,
      "grad_norm": 1.4140625,
      "learning_rate": 6.50837258170615e-05,
      "loss": 4.8014,
      "mean_token_accuracy": 0.235243321955204,
      "num_tokens": 18881099.0,
      "step": 10385
    },
    {
      "entropy": 5.611511135101319,
      "epoch": 8.926514825956167,
      "grad_norm": 1.390625,
      "learning_rate": 6.496426893592397e-05,
      "loss": 4.8314,
      "mean_token_accuracy": 0.22942321598529816,
      "num_tokens": 18890368.0,
      "step": 10390
    },
    {
      "entropy": 5.5862115859985355,
      "epoch": 8.930812204555222,
      "grad_norm": 1.4765625,
      "learning_rate": 6.484527068887258e-05,
      "loss": 4.8021,
      "mean_token_accuracy": 0.2278003066778183,
      "num_tokens": 18898576.0,
      "step": 10395
    },
    {
      "entropy": 5.53840069770813,
      "epoch": 8.935109583154276,
      "grad_norm": 1.53125,
      "learning_rate": 6.472673133575181e-05,
      "loss": 4.7432,
      "mean_token_accuracy": 0.23103292882442475,
      "num_tokens": 18907276.0,
      "step": 10400
    },
    {
      "entropy": 5.55468602180481,
      "epoch": 8.939406961753331,
      "grad_norm": 1.4765625,
      "learning_rate": 6.460865113540437e-05,
      "loss": 4.7316,
      "mean_token_accuracy": 0.23958467543125153,
      "num_tokens": 18915920.0,
      "step": 10405
    },
    {
      "entropy": 5.562465620040894,
      "epoch": 8.943704340352385,
      "grad_norm": 1.375,
      "learning_rate": 6.449103034567011e-05,
      "loss": 4.7304,
      "mean_token_accuracy": 0.2437051236629486,
      "num_tokens": 18925529.0,
      "step": 10410
    },
    {
      "entropy": 5.578067064285278,
      "epoch": 8.94800171895144,
      "grad_norm": 1.4765625,
      "learning_rate": 6.437386922338591e-05,
      "loss": 4.8393,
      "mean_token_accuracy": 0.23440202623605727,
      "num_tokens": 18934859.0,
      "step": 10415
    },
    {
      "entropy": 5.581465387344361,
      "epoch": 8.952299097550494,
      "grad_norm": 1.5,
      "learning_rate": 6.425716802438479e-05,
      "loss": 4.8016,
      "mean_token_accuracy": 0.2400070011615753,
      "num_tokens": 18943804.0,
      "step": 10420
    },
    {
      "entropy": 5.520764303207398,
      "epoch": 8.95659647614955,
      "grad_norm": 1.4765625,
      "learning_rate": 6.414092700349548e-05,
      "loss": 4.7755,
      "mean_token_accuracy": 0.23708345890045165,
      "num_tokens": 18954290.0,
      "step": 10425
    },
    {
      "entropy": 5.563492155075073,
      "epoch": 8.960893854748603,
      "grad_norm": 1.5,
      "learning_rate": 6.402514641454192e-05,
      "loss": 4.8776,
      "mean_token_accuracy": 0.2191103681921959,
      "num_tokens": 18963448.0,
      "step": 10430
    },
    {
      "entropy": 5.564414644241333,
      "epoch": 8.965191233347658,
      "grad_norm": 1.3671875,
      "learning_rate": 6.390982651034274e-05,
      "loss": 4.7992,
      "mean_token_accuracy": 0.2313682720065117,
      "num_tokens": 18972314.0,
      "step": 10435
    },
    {
      "entropy": 5.461703062057495,
      "epoch": 8.969488611946712,
      "grad_norm": 1.5703125,
      "learning_rate": 6.379496754271044e-05,
      "loss": 4.6049,
      "mean_token_accuracy": 0.2550671100616455,
      "num_tokens": 18980720.0,
      "step": 10440
    },
    {
      "entropy": 5.482317733764648,
      "epoch": 8.973785990545768,
      "grad_norm": 1.53125,
      "learning_rate": 6.368056976245107e-05,
      "loss": 4.6517,
      "mean_token_accuracy": 0.2550403669476509,
      "num_tokens": 18988267.0,
      "step": 10445
    },
    {
      "entropy": 5.558830881118775,
      "epoch": 8.978083369144821,
      "grad_norm": 1.5234375,
      "learning_rate": 6.356663341936368e-05,
      "loss": 4.7668,
      "mean_token_accuracy": 0.2394379660487175,
      "num_tokens": 18997036.0,
      "step": 10450
    },
    {
      "entropy": 5.5316205501556395,
      "epoch": 8.982380747743877,
      "grad_norm": 1.4765625,
      "learning_rate": 6.345315876223977e-05,
      "loss": 4.713,
      "mean_token_accuracy": 0.24060868471860886,
      "num_tokens": 19005440.0,
      "step": 10455
    },
    {
      "entropy": 5.540014791488647,
      "epoch": 8.98667812634293,
      "grad_norm": 1.453125,
      "learning_rate": 6.334014603886256e-05,
      "loss": 4.7693,
      "mean_token_accuracy": 0.23835282474756242,
      "num_tokens": 19014449.0,
      "step": 10460
    },
    {
      "entropy": 5.492503023147583,
      "epoch": 8.990975504941986,
      "grad_norm": 1.5625,
      "learning_rate": 6.322759549600665e-05,
      "loss": 4.742,
      "mean_token_accuracy": 0.23789905607700348,
      "num_tokens": 19024061.0,
      "step": 10465
    },
    {
      "entropy": 5.48447585105896,
      "epoch": 8.99527288354104,
      "grad_norm": 1.4609375,
      "learning_rate": 6.311550737943753e-05,
      "loss": 4.7191,
      "mean_token_accuracy": 0.24493125826120377,
      "num_tokens": 19033400.0,
      "step": 10470
    },
    {
      "entropy": 5.55003252029419,
      "epoch": 8.999570262140095,
      "grad_norm": 1.5234375,
      "learning_rate": 6.300388193391075e-05,
      "loss": 4.784,
      "mean_token_accuracy": 0.23504850268363953,
      "num_tokens": 19042633.0,
      "step": 10475
    },
    {
      "entropy": 5.557694170210096,
      "epoch": 9.003437902879243,
      "grad_norm": 1.40625,
      "learning_rate": 6.289271940317174e-05,
      "loss": 4.8424,
      "mean_token_accuracy": 0.22482394675413767,
      "num_tokens": 19051120.0,
      "step": 10480
    },
    {
      "entropy": 5.534358024597168,
      "epoch": 9.007735281478299,
      "grad_norm": 1.421875,
      "learning_rate": 6.278202002995497e-05,
      "loss": 4.7655,
      "mean_token_accuracy": 0.23592307567596435,
      "num_tokens": 19060755.0,
      "step": 10485
    },
    {
      "entropy": 5.540455627441406,
      "epoch": 9.012032660077352,
      "grad_norm": 1.5078125,
      "learning_rate": 6.267178405598371e-05,
      "loss": 4.6747,
      "mean_token_accuracy": 0.24799301773309707,
      "num_tokens": 19069197.0,
      "step": 10490
    },
    {
      "entropy": 5.590549802780151,
      "epoch": 9.016330038676408,
      "grad_norm": 1.4609375,
      "learning_rate": 6.256201172196921e-05,
      "loss": 4.7264,
      "mean_token_accuracy": 0.2417898178100586,
      "num_tokens": 19078452.0,
      "step": 10495
    },
    {
      "entropy": 5.566554021835327,
      "epoch": 9.020627417275461,
      "grad_norm": 1.390625,
      "learning_rate": 6.245270326761034e-05,
      "loss": 4.7481,
      "mean_token_accuracy": 0.23358300924301148,
      "num_tokens": 19087127.0,
      "step": 10500
    },
    {
      "epoch": 9.020627417275461,
      "eval_entropy": 5.446591773011663,
      "eval_loss": 5.928488731384277,
      "eval_mean_token_accuracy": 0.1801695137343428,
      "eval_num_tokens": 19087127.0,
      "eval_runtime": 2.8542,
      "eval_samples_per_second": 1243.428,
      "eval_steps_per_second": 155.56,
      "step": 10500
    },
    {
      "entropy": 5.590010643005371,
      "epoch": 9.024924795874517,
      "grad_norm": 1.484375,
      "learning_rate": 6.234385893159311e-05,
      "loss": 4.7516,
      "mean_token_accuracy": 0.23886649012565614,
      "num_tokens": 19095610.0,
      "step": 10505
    },
    {
      "entropy": 5.518437051773072,
      "epoch": 9.02922217447357,
      "grad_norm": 1.546875,
      "learning_rate": 6.223547895159009e-05,
      "loss": 4.6808,
      "mean_token_accuracy": 0.24538882672786713,
      "num_tokens": 19103649.0,
      "step": 10510
    },
    {
      "entropy": 5.471475505828858,
      "epoch": 9.033519553072626,
      "grad_norm": 1.4375,
      "learning_rate": 6.212756356425978e-05,
      "loss": 4.6276,
      "mean_token_accuracy": 0.2507261961698532,
      "num_tokens": 19112426.0,
      "step": 10515
    },
    {
      "entropy": 5.567093086242676,
      "epoch": 9.03781693167168,
      "grad_norm": 1.4140625,
      "learning_rate": 6.202011300524623e-05,
      "loss": 4.7672,
      "mean_token_accuracy": 0.24036678522825242,
      "num_tokens": 19121307.0,
      "step": 10520
    },
    {
      "entropy": 5.60212721824646,
      "epoch": 9.042114310270735,
      "grad_norm": 1.3671875,
      "learning_rate": 6.191312750917855e-05,
      "loss": 4.7785,
      "mean_token_accuracy": 0.24381277710199356,
      "num_tokens": 19129728.0,
      "step": 10525
    },
    {
      "entropy": 5.537043714523316,
      "epoch": 9.046411688869789,
      "grad_norm": 1.4765625,
      "learning_rate": 6.180660730967036e-05,
      "loss": 4.7154,
      "mean_token_accuracy": 0.236446350812912,
      "num_tokens": 19139931.0,
      "step": 10530
    },
    {
      "entropy": 5.599968433380127,
      "epoch": 9.050709067468844,
      "grad_norm": 1.4375,
      "learning_rate": 6.170055263931912e-05,
      "loss": 4.7135,
      "mean_token_accuracy": 0.23799506425857545,
      "num_tokens": 19148711.0,
      "step": 10535
    },
    {
      "entropy": 5.596178150177002,
      "epoch": 9.055006446067898,
      "grad_norm": 1.5,
      "learning_rate": 6.159496372970584e-05,
      "loss": 4.6814,
      "mean_token_accuracy": 0.24057221561670303,
      "num_tokens": 19156937.0,
      "step": 10540
    },
    {
      "entropy": 5.516369009017945,
      "epoch": 9.059303824666953,
      "grad_norm": 1.359375,
      "learning_rate": 6.148984081139454e-05,
      "loss": 4.7273,
      "mean_token_accuracy": 0.2371464863419533,
      "num_tokens": 19166449.0,
      "step": 10545
    },
    {
      "entropy": 5.598357343673706,
      "epoch": 9.063601203266007,
      "grad_norm": 1.3046875,
      "learning_rate": 6.138518411393163e-05,
      "loss": 4.8579,
      "mean_token_accuracy": 0.2288147896528244,
      "num_tokens": 19175757.0,
      "step": 10550
    },
    {
      "entropy": 5.564884662628174,
      "epoch": 9.067898581865062,
      "grad_norm": 1.4453125,
      "learning_rate": 6.128099386584548e-05,
      "loss": 4.7401,
      "mean_token_accuracy": 0.24756367206573487,
      "num_tokens": 19184638.0,
      "step": 10555
    },
    {
      "entropy": 5.542890167236328,
      "epoch": 9.072195960464116,
      "grad_norm": 1.4609375,
      "learning_rate": 6.1177270294646e-05,
      "loss": 4.7789,
      "mean_token_accuracy": 0.23930480927228928,
      "num_tokens": 19193791.0,
      "step": 10560
    },
    {
      "entropy": 5.530588150024414,
      "epoch": 9.076493339063171,
      "grad_norm": 1.4609375,
      "learning_rate": 6.107401362682401e-05,
      "loss": 4.7989,
      "mean_token_accuracy": 0.2353051245212555,
      "num_tokens": 19202595.0,
      "step": 10565
    },
    {
      "entropy": 5.524761152267456,
      "epoch": 9.080790717662227,
      "grad_norm": 1.4140625,
      "learning_rate": 6.097122408785076e-05,
      "loss": 4.7841,
      "mean_token_accuracy": 0.23663593232631683,
      "num_tokens": 19211994.0,
      "step": 10570
    },
    {
      "entropy": 5.547695446014404,
      "epoch": 9.08508809626128,
      "grad_norm": 1.3984375,
      "learning_rate": 6.086890190217752e-05,
      "loss": 4.724,
      "mean_token_accuracy": 0.24338321685791015,
      "num_tokens": 19220588.0,
      "step": 10575
    },
    {
      "entropy": 5.5205755710601805,
      "epoch": 9.089385474860336,
      "grad_norm": 1.484375,
      "learning_rate": 6.076704729323507e-05,
      "loss": 4.6812,
      "mean_token_accuracy": 0.24725836366415024,
      "num_tokens": 19229679.0,
      "step": 10580
    },
    {
      "entropy": 5.546969413757324,
      "epoch": 9.09368285345939,
      "grad_norm": 1.3984375,
      "learning_rate": 6.0665660483433173e-05,
      "loss": 4.8034,
      "mean_token_accuracy": 0.2318977028131485,
      "num_tokens": 19239269.0,
      "step": 10585
    },
    {
      "entropy": 5.560802888870239,
      "epoch": 9.097980232058445,
      "grad_norm": 1.46875,
      "learning_rate": 6.05647416941601e-05,
      "loss": 4.7919,
      "mean_token_accuracy": 0.2426741361618042,
      "num_tokens": 19249046.0,
      "step": 10590
    },
    {
      "entropy": 5.524609899520874,
      "epoch": 9.102277610657499,
      "grad_norm": 1.4921875,
      "learning_rate": 6.046429114578212e-05,
      "loss": 4.674,
      "mean_token_accuracy": 0.23944690972566604,
      "num_tokens": 19257398.0,
      "step": 10595
    },
    {
      "entropy": 5.594950199127197,
      "epoch": 9.106574989256554,
      "grad_norm": 1.4921875,
      "learning_rate": 6.0364309057643084e-05,
      "loss": 4.7621,
      "mean_token_accuracy": 0.2294335812330246,
      "num_tokens": 19266619.0,
      "step": 10600
    },
    {
      "entropy": 5.587691688537598,
      "epoch": 9.110872367855608,
      "grad_norm": 1.46875,
      "learning_rate": 6.0264795648063904e-05,
      "loss": 4.7904,
      "mean_token_accuracy": 0.2357139065861702,
      "num_tokens": 19275837.0,
      "step": 10605
    },
    {
      "entropy": 5.555129718780518,
      "epoch": 9.115169746454663,
      "grad_norm": 1.3203125,
      "learning_rate": 6.0165751134342155e-05,
      "loss": 4.698,
      "mean_token_accuracy": 0.2403426080942154,
      "num_tokens": 19284745.0,
      "step": 10610
    },
    {
      "entropy": 5.527823162078858,
      "epoch": 9.119467125053717,
      "grad_norm": 1.3046875,
      "learning_rate": 6.006717573275138e-05,
      "loss": 4.6933,
      "mean_token_accuracy": 0.2427384376525879,
      "num_tokens": 19293956.0,
      "step": 10615
    },
    {
      "entropy": 5.534629058837891,
      "epoch": 9.123764503652772,
      "grad_norm": 1.2421875,
      "learning_rate": 5.996906965854093e-05,
      "loss": 4.7934,
      "mean_token_accuracy": 0.23455743342638016,
      "num_tokens": 19303629.0,
      "step": 10620
    },
    {
      "entropy": 5.591663837432861,
      "epoch": 9.128061882251826,
      "grad_norm": 1.5234375,
      "learning_rate": 5.987143312593522e-05,
      "loss": 4.8316,
      "mean_token_accuracy": 0.23300620764493943,
      "num_tokens": 19312458.0,
      "step": 10625
    },
    {
      "entropy": 5.521102237701416,
      "epoch": 9.132359260850881,
      "grad_norm": 1.4453125,
      "learning_rate": 5.977426634813338e-05,
      "loss": 4.7143,
      "mean_token_accuracy": 0.2471512332558632,
      "num_tokens": 19321428.0,
      "step": 10630
    },
    {
      "entropy": 5.621559953689575,
      "epoch": 9.136656639449935,
      "grad_norm": 1.484375,
      "learning_rate": 5.9677569537308866e-05,
      "loss": 4.878,
      "mean_token_accuracy": 0.22608232945203782,
      "num_tokens": 19331176.0,
      "step": 10635
    },
    {
      "entropy": 5.567727375030517,
      "epoch": 9.14095401804899,
      "grad_norm": 1.421875,
      "learning_rate": 5.958134290460888e-05,
      "loss": 4.7976,
      "mean_token_accuracy": 0.23378018885850907,
      "num_tokens": 19340401.0,
      "step": 10640
    },
    {
      "entropy": 5.49088191986084,
      "epoch": 9.145251396648044,
      "grad_norm": 1.25,
      "learning_rate": 5.94855866601539e-05,
      "loss": 4.6959,
      "mean_token_accuracy": 0.2410607010126114,
      "num_tokens": 19350700.0,
      "step": 10645
    },
    {
      "entropy": 5.578736591339111,
      "epoch": 9.1495487752471,
      "grad_norm": 1.4765625,
      "learning_rate": 5.939030101303724e-05,
      "loss": 4.781,
      "mean_token_accuracy": 0.22958478033542634,
      "num_tokens": 19360040.0,
      "step": 10650
    },
    {
      "entropy": 5.56622200012207,
      "epoch": 9.153846153846153,
      "grad_norm": 1.5546875,
      "learning_rate": 5.929548617132472e-05,
      "loss": 4.7091,
      "mean_token_accuracy": 0.24165599048137665,
      "num_tokens": 19368546.0,
      "step": 10655
    },
    {
      "entropy": 5.483640956878662,
      "epoch": 9.158143532445209,
      "grad_norm": 1.453125,
      "learning_rate": 5.920114234205407e-05,
      "loss": 4.6691,
      "mean_token_accuracy": 0.24358928054571152,
      "num_tokens": 19377137.0,
      "step": 10660
    },
    {
      "entropy": 5.611877679824829,
      "epoch": 9.162440911044262,
      "grad_norm": 1.4453125,
      "learning_rate": 5.910726973123451e-05,
      "loss": 4.8526,
      "mean_token_accuracy": 0.22434186935424805,
      "num_tokens": 19385781.0,
      "step": 10665
    },
    {
      "entropy": 5.538211917877197,
      "epoch": 9.166738289643318,
      "grad_norm": 1.4765625,
      "learning_rate": 5.901386854384622e-05,
      "loss": 4.7748,
      "mean_token_accuracy": 0.24201036542654036,
      "num_tokens": 19395186.0,
      "step": 10670
    },
    {
      "entropy": 5.583196592330933,
      "epoch": 9.171035668242371,
      "grad_norm": 1.5,
      "learning_rate": 5.892093898384017e-05,
      "loss": 4.7366,
      "mean_token_accuracy": 0.23687657117843627,
      "num_tokens": 19404140.0,
      "step": 10675
    },
    {
      "entropy": 5.559548377990723,
      "epoch": 9.175333046841427,
      "grad_norm": 1.484375,
      "learning_rate": 5.8828481254137276e-05,
      "loss": 4.7594,
      "mean_token_accuracy": 0.2367544636130333,
      "num_tokens": 19414481.0,
      "step": 10680
    },
    {
      "entropy": 5.447652816772461,
      "epoch": 9.17963042544048,
      "grad_norm": 1.5546875,
      "learning_rate": 5.873649555662836e-05,
      "loss": 4.606,
      "mean_token_accuracy": 0.24810488224029542,
      "num_tokens": 19423259.0,
      "step": 10685
    },
    {
      "entropy": 5.553141069412232,
      "epoch": 9.183927804039536,
      "grad_norm": 1.484375,
      "learning_rate": 5.8644982092173335e-05,
      "loss": 4.7377,
      "mean_token_accuracy": 0.24042945057153703,
      "num_tokens": 19432011.0,
      "step": 10690
    },
    {
      "entropy": 5.539479351043701,
      "epoch": 9.18822518263859,
      "grad_norm": 1.4375,
      "learning_rate": 5.85539410606011e-05,
      "loss": 4.7256,
      "mean_token_accuracy": 0.2451218143105507,
      "num_tokens": 19441221.0,
      "step": 10695
    },
    {
      "entropy": 5.502447271347046,
      "epoch": 9.192522561237645,
      "grad_norm": 1.4921875,
      "learning_rate": 5.8463372660708836e-05,
      "loss": 4.6784,
      "mean_token_accuracy": 0.2455634817481041,
      "num_tokens": 19450014.0,
      "step": 10700
    },
    {
      "entropy": 5.556422424316406,
      "epoch": 9.196819939836699,
      "grad_norm": 1.40625,
      "learning_rate": 5.837327709026171e-05,
      "loss": 4.7915,
      "mean_token_accuracy": 0.23408205807209015,
      "num_tokens": 19459480.0,
      "step": 10705
    },
    {
      "entropy": 5.562296628952026,
      "epoch": 9.201117318435754,
      "grad_norm": 1.390625,
      "learning_rate": 5.8283654545992464e-05,
      "loss": 4.7789,
      "mean_token_accuracy": 0.23850567638874054,
      "num_tokens": 19469111.0,
      "step": 10710
    },
    {
      "entropy": 5.548806524276733,
      "epoch": 9.20541469703481,
      "grad_norm": 1.4765625,
      "learning_rate": 5.819450522360096e-05,
      "loss": 4.7344,
      "mean_token_accuracy": 0.23930932432413102,
      "num_tokens": 19477531.0,
      "step": 10715
    },
    {
      "entropy": 5.579662036895752,
      "epoch": 9.209712075633863,
      "grad_norm": 1.5234375,
      "learning_rate": 5.810582931775362e-05,
      "loss": 4.734,
      "mean_token_accuracy": 0.23548170775175095,
      "num_tokens": 19486041.0,
      "step": 10720
    },
    {
      "entropy": 5.554513311386108,
      "epoch": 9.214009454232919,
      "grad_norm": 1.4609375,
      "learning_rate": 5.801762702208317e-05,
      "loss": 4.7703,
      "mean_token_accuracy": 0.23784253895282745,
      "num_tokens": 19495514.0,
      "step": 10725
    },
    {
      "entropy": 5.486033010482788,
      "epoch": 9.218306832831972,
      "grad_norm": 1.53125,
      "learning_rate": 5.7929898529188215e-05,
      "loss": 4.7094,
      "mean_token_accuracy": 0.24143276065587999,
      "num_tokens": 19503879.0,
      "step": 10730
    },
    {
      "entropy": 5.507168817520141,
      "epoch": 9.222604211431028,
      "grad_norm": 1.3359375,
      "learning_rate": 5.784264403063272e-05,
      "loss": 4.679,
      "mean_token_accuracy": 0.24017557948827745,
      "num_tokens": 19513922.0,
      "step": 10735
    },
    {
      "entropy": 5.5334228515625,
      "epoch": 9.226901590030081,
      "grad_norm": 1.5390625,
      "learning_rate": 5.775586371694561e-05,
      "loss": 4.719,
      "mean_token_accuracy": 0.24083528220653533,
      "num_tokens": 19522555.0,
      "step": 10740
    },
    {
      "entropy": 5.530796527862549,
      "epoch": 9.231198968629137,
      "grad_norm": 1.3359375,
      "learning_rate": 5.7669557777620376e-05,
      "loss": 4.7259,
      "mean_token_accuracy": 0.24067745953798295,
      "num_tokens": 19531245.0,
      "step": 10745
    },
    {
      "entropy": 5.516829299926758,
      "epoch": 9.23549634722819,
      "grad_norm": 1.4609375,
      "learning_rate": 5.7583726401114756e-05,
      "loss": 4.6836,
      "mean_token_accuracy": 0.24842206984758378,
      "num_tokens": 19540258.0,
      "step": 10750
    },
    {
      "entropy": 5.5769330024719235,
      "epoch": 9.239793725827246,
      "grad_norm": 1.5234375,
      "learning_rate": 5.749836977485013e-05,
      "loss": 4.7511,
      "mean_token_accuracy": 0.2340255320072174,
      "num_tokens": 19550071.0,
      "step": 10755
    },
    {
      "entropy": 5.508723545074463,
      "epoch": 9.2440911044263,
      "grad_norm": 1.3515625,
      "learning_rate": 5.741348808521128e-05,
      "loss": 4.6691,
      "mean_token_accuracy": 0.24482452720403672,
      "num_tokens": 19558972.0,
      "step": 10760
    },
    {
      "entropy": 5.553051090240478,
      "epoch": 9.248388483025355,
      "grad_norm": 1.453125,
      "learning_rate": 5.7329081517545846e-05,
      "loss": 4.7748,
      "mean_token_accuracy": 0.23405635356903076,
      "num_tokens": 19568270.0,
      "step": 10765
    },
    {
      "entropy": 5.588024759292603,
      "epoch": 9.252685861624409,
      "grad_norm": 1.6484375,
      "learning_rate": 5.724515025616409e-05,
      "loss": 4.7435,
      "mean_token_accuracy": 0.2430342122912407,
      "num_tokens": 19576726.0,
      "step": 10770
    },
    {
      "entropy": 5.5403618812561035,
      "epoch": 9.256983240223464,
      "grad_norm": 1.375,
      "learning_rate": 5.716169448433832e-05,
      "loss": 4.7491,
      "mean_token_accuracy": 0.2323804259300232,
      "num_tokens": 19586706.0,
      "step": 10775
    },
    {
      "entropy": 5.551635217666626,
      "epoch": 9.261280618822518,
      "grad_norm": 1.4921875,
      "learning_rate": 5.707871438430255e-05,
      "loss": 4.7735,
      "mean_token_accuracy": 0.23547276854515076,
      "num_tokens": 19595370.0,
      "step": 10780
    },
    {
      "entropy": 5.55864109992981,
      "epoch": 9.265577997421573,
      "grad_norm": 1.5234375,
      "learning_rate": 5.699621013725218e-05,
      "loss": 4.7695,
      "mean_token_accuracy": 0.24265750795602797,
      "num_tokens": 19604134.0,
      "step": 10785
    },
    {
      "entropy": 5.484230804443359,
      "epoch": 9.269875376020627,
      "grad_norm": 1.421875,
      "learning_rate": 5.691418192334352e-05,
      "loss": 4.6909,
      "mean_token_accuracy": 0.24529312998056413,
      "num_tokens": 19612716.0,
      "step": 10790
    },
    {
      "entropy": 5.536768245697021,
      "epoch": 9.274172754619682,
      "grad_norm": 1.421875,
      "learning_rate": 5.683262992169341e-05,
      "loss": 4.7952,
      "mean_token_accuracy": 0.23541211187839509,
      "num_tokens": 19621850.0,
      "step": 10795
    },
    {
      "entropy": 5.527990388870239,
      "epoch": 9.278470133218736,
      "grad_norm": 1.515625,
      "learning_rate": 5.675155431037876e-05,
      "loss": 4.7376,
      "mean_token_accuracy": 0.23815283924341202,
      "num_tokens": 19630706.0,
      "step": 10800
    },
    {
      "entropy": 5.499881553649902,
      "epoch": 9.282767511817791,
      "grad_norm": 1.5078125,
      "learning_rate": 5.6670955266436365e-05,
      "loss": 4.7183,
      "mean_token_accuracy": 0.24916807562112808,
      "num_tokens": 19639364.0,
      "step": 10805
    },
    {
      "entropy": 5.586935234069824,
      "epoch": 9.287064890416845,
      "grad_norm": 1.4609375,
      "learning_rate": 5.659083296586229e-05,
      "loss": 4.6842,
      "mean_token_accuracy": 0.2437358945608139,
      "num_tokens": 19647816.0,
      "step": 10810
    },
    {
      "entropy": 5.586746644973755,
      "epoch": 9.2913622690159,
      "grad_norm": 1.34375,
      "learning_rate": 5.6511187583611663e-05,
      "loss": 4.7923,
      "mean_token_accuracy": 0.2331514835357666,
      "num_tokens": 19657136.0,
      "step": 10815
    },
    {
      "entropy": 5.582297992706299,
      "epoch": 9.295659647614954,
      "grad_norm": 1.5625,
      "learning_rate": 5.643201929359809e-05,
      "loss": 4.7481,
      "mean_token_accuracy": 0.23865289241075516,
      "num_tokens": 19665676.0,
      "step": 10820
    },
    {
      "entropy": 5.5720799446105955,
      "epoch": 9.29995702621401,
      "grad_norm": 1.5625,
      "learning_rate": 5.635332826869353e-05,
      "loss": 4.7126,
      "mean_token_accuracy": 0.24430434703826903,
      "num_tokens": 19674415.0,
      "step": 10825
    },
    {
      "entropy": 5.502606630325317,
      "epoch": 9.304254404813063,
      "grad_norm": 1.359375,
      "learning_rate": 5.6275114680727716e-05,
      "loss": 4.6971,
      "mean_token_accuracy": 0.2413142815232277,
      "num_tokens": 19684411.0,
      "step": 10830
    },
    {
      "entropy": 5.517111539840698,
      "epoch": 9.308551783412119,
      "grad_norm": 1.5234375,
      "learning_rate": 5.619737870048783e-05,
      "loss": 4.6844,
      "mean_token_accuracy": 0.2401023253798485,
      "num_tokens": 19693192.0,
      "step": 10835
    },
    {
      "entropy": 5.4301910400390625,
      "epoch": 9.312849162011172,
      "grad_norm": 1.484375,
      "learning_rate": 5.612012049771823e-05,
      "loss": 4.5982,
      "mean_token_accuracy": 0.24852366000413895,
      "num_tokens": 19701710.0,
      "step": 10840
    },
    {
      "entropy": 5.59416618347168,
      "epoch": 9.317146540610228,
      "grad_norm": 1.3984375,
      "learning_rate": 5.6043340241119924e-05,
      "loss": 4.8001,
      "mean_token_accuracy": 0.23299821466207504,
      "num_tokens": 19710909.0,
      "step": 10845
    },
    {
      "entropy": 5.561320972442627,
      "epoch": 9.321443919209282,
      "grad_norm": 1.4609375,
      "learning_rate": 5.596703809835033e-05,
      "loss": 4.7477,
      "mean_token_accuracy": 0.23609877973794938,
      "num_tokens": 19720634.0,
      "step": 10850
    },
    {
      "entropy": 5.4886026859283445,
      "epoch": 9.325741297808337,
      "grad_norm": 1.65625,
      "learning_rate": 5.589121423602277e-05,
      "loss": 4.6453,
      "mean_token_accuracy": 0.24567285627126695,
      "num_tokens": 19729763.0,
      "step": 10855
    },
    {
      "entropy": 5.574591732025146,
      "epoch": 9.33003867640739,
      "grad_norm": 1.4375,
      "learning_rate": 5.581586881970631e-05,
      "loss": 4.8244,
      "mean_token_accuracy": 0.2289576604962349,
      "num_tokens": 19739065.0,
      "step": 10860
    },
    {
      "entropy": 5.50982255935669,
      "epoch": 9.334336055006446,
      "grad_norm": 1.4140625,
      "learning_rate": 5.574100201392522e-05,
      "loss": 4.648,
      "mean_token_accuracy": 0.25283440500497817,
      "num_tokens": 19748519.0,
      "step": 10865
    },
    {
      "entropy": 5.568785524368286,
      "epoch": 9.338633433605501,
      "grad_norm": 1.4453125,
      "learning_rate": 5.5666613982158665e-05,
      "loss": 4.7237,
      "mean_token_accuracy": 0.24463753551244735,
      "num_tokens": 19757573.0,
      "step": 10870
    },
    {
      "entropy": 5.570475292205811,
      "epoch": 9.342930812204555,
      "grad_norm": 1.46875,
      "learning_rate": 5.559270488684036e-05,
      "loss": 4.8539,
      "mean_token_accuracy": 0.22743835896253586,
      "num_tokens": 19767534.0,
      "step": 10875
    },
    {
      "entropy": 5.605714893341064,
      "epoch": 9.34722819080361,
      "grad_norm": 1.3828125,
      "learning_rate": 5.551927488935826e-05,
      "loss": 4.8066,
      "mean_token_accuracy": 0.23367402851581573,
      "num_tokens": 19776929.0,
      "step": 10880
    },
    {
      "entropy": 5.574794292449951,
      "epoch": 9.351525569402664,
      "grad_norm": 1.46875,
      "learning_rate": 5.5446324150054086e-05,
      "loss": 4.7872,
      "mean_token_accuracy": 0.2343237966299057,
      "num_tokens": 19785772.0,
      "step": 10885
    },
    {
      "entropy": 5.544982481002807,
      "epoch": 9.35582294800172,
      "grad_norm": 1.28125,
      "learning_rate": 5.537385282822315e-05,
      "loss": 4.75,
      "mean_token_accuracy": 0.2428746283054352,
      "num_tokens": 19796073.0,
      "step": 10890
    },
    {
      "entropy": 5.553908967971802,
      "epoch": 9.360120326600773,
      "grad_norm": 1.6171875,
      "learning_rate": 5.53018610821138e-05,
      "loss": 4.7847,
      "mean_token_accuracy": 0.2373655453324318,
      "num_tokens": 19804511.0,
      "step": 10895
    },
    {
      "entropy": 5.570815658569336,
      "epoch": 9.364417705199829,
      "grad_norm": 1.4609375,
      "learning_rate": 5.523034906892728e-05,
      "loss": 4.7955,
      "mean_token_accuracy": 0.229778553545475,
      "num_tokens": 19813348.0,
      "step": 10900
    },
    {
      "entropy": 5.593582010269165,
      "epoch": 9.368715083798882,
      "grad_norm": 1.4453125,
      "learning_rate": 5.515931694481722e-05,
      "loss": 4.812,
      "mean_token_accuracy": 0.23042097091674804,
      "num_tokens": 19822390.0,
      "step": 10905
    },
    {
      "entropy": 5.603141593933105,
      "epoch": 9.373012462397938,
      "grad_norm": 1.4453125,
      "learning_rate": 5.508876486488936e-05,
      "loss": 4.8194,
      "mean_token_accuracy": 0.2299593523144722,
      "num_tokens": 19832423.0,
      "step": 10910
    },
    {
      "entropy": 5.550478839874268,
      "epoch": 9.377309840996991,
      "grad_norm": 1.5546875,
      "learning_rate": 5.501869298320128e-05,
      "loss": 4.7859,
      "mean_token_accuracy": 0.22471377700567247,
      "num_tokens": 19841271.0,
      "step": 10915
    },
    {
      "entropy": 5.563661766052246,
      "epoch": 9.381607219596047,
      "grad_norm": 1.546875,
      "learning_rate": 5.4949101452761995e-05,
      "loss": 4.7641,
      "mean_token_accuracy": 0.2375018134713173,
      "num_tokens": 19850648.0,
      "step": 10920
    },
    {
      "entropy": 5.534654378890991,
      "epoch": 9.3859045981951,
      "grad_norm": 1.3671875,
      "learning_rate": 5.4879990425531534e-05,
      "loss": 4.7045,
      "mean_token_accuracy": 0.247677543759346,
      "num_tokens": 19860102.0,
      "step": 10925
    },
    {
      "entropy": 5.562622356414795,
      "epoch": 9.390201976794156,
      "grad_norm": 1.28125,
      "learning_rate": 5.4811360052420754e-05,
      "loss": 4.7904,
      "mean_token_accuracy": 0.23720292150974273,
      "num_tokens": 19870008.0,
      "step": 10930
    },
    {
      "entropy": 5.517884731292725,
      "epoch": 9.39449935539321,
      "grad_norm": 1.3046875,
      "learning_rate": 5.4743210483290974e-05,
      "loss": 4.6946,
      "mean_token_accuracy": 0.24500344395637513,
      "num_tokens": 19879452.0,
      "step": 10935
    },
    {
      "entropy": 5.607058000564575,
      "epoch": 9.398796733992265,
      "grad_norm": 1.3515625,
      "learning_rate": 5.467554186695364e-05,
      "loss": 4.8197,
      "mean_token_accuracy": 0.23052308857440948,
      "num_tokens": 19888956.0,
      "step": 10940
    },
    {
      "entropy": 5.528801488876343,
      "epoch": 9.403094112591319,
      "grad_norm": 1.4296875,
      "learning_rate": 5.4608354351169944e-05,
      "loss": 4.752,
      "mean_token_accuracy": 0.2380228668451309,
      "num_tokens": 19897760.0,
      "step": 10945
    },
    {
      "entropy": 5.525875377655029,
      "epoch": 9.407391491190374,
      "grad_norm": 1.5078125,
      "learning_rate": 5.454164808265057e-05,
      "loss": 4.7546,
      "mean_token_accuracy": 0.23870436996221542,
      "num_tokens": 19905692.0,
      "step": 10950
    },
    {
      "entropy": 5.50925498008728,
      "epoch": 9.411688869789428,
      "grad_norm": 1.484375,
      "learning_rate": 5.447542320705532e-05,
      "loss": 4.6855,
      "mean_token_accuracy": 0.24661193042993546,
      "num_tokens": 19914633.0,
      "step": 10955
    },
    {
      "entropy": 5.5851202487945555,
      "epoch": 9.415986248388483,
      "grad_norm": 1.484375,
      "learning_rate": 5.440967986899289e-05,
      "loss": 4.7577,
      "mean_token_accuracy": 0.23358041048049927,
      "num_tokens": 19922749.0,
      "step": 10960
    },
    {
      "entropy": 5.555970668792725,
      "epoch": 9.420283626987537,
      "grad_norm": 1.484375,
      "learning_rate": 5.434441821202042e-05,
      "loss": 4.7291,
      "mean_token_accuracy": 0.243088997900486,
      "num_tokens": 19931274.0,
      "step": 10965
    },
    {
      "entropy": 5.561767435073852,
      "epoch": 9.424581005586592,
      "grad_norm": 1.421875,
      "learning_rate": 5.42796383786433e-05,
      "loss": 4.739,
      "mean_token_accuracy": 0.2367652416229248,
      "num_tokens": 19940052.0,
      "step": 10970
    },
    {
      "entropy": 5.543972492218018,
      "epoch": 9.428878384185646,
      "grad_norm": 1.40625,
      "learning_rate": 5.4215340510314805e-05,
      "loss": 4.7241,
      "mean_token_accuracy": 0.2396532729268074,
      "num_tokens": 19948890.0,
      "step": 10975
    },
    {
      "entropy": 5.542488622665405,
      "epoch": 9.433175762784701,
      "grad_norm": 1.3671875,
      "learning_rate": 5.41515247474358e-05,
      "loss": 4.7978,
      "mean_token_accuracy": 0.23807926774024962,
      "num_tokens": 19958158.0,
      "step": 10980
    },
    {
      "entropy": 5.587992143630982,
      "epoch": 9.437473141383755,
      "grad_norm": 1.375,
      "learning_rate": 5.4088191229354306e-05,
      "loss": 4.847,
      "mean_token_accuracy": 0.23087634593248368,
      "num_tokens": 19967432.0,
      "step": 10985
    },
    {
      "entropy": 5.522149085998535,
      "epoch": 9.44177051998281,
      "grad_norm": 1.4140625,
      "learning_rate": 5.402534009436552e-05,
      "loss": 4.7395,
      "mean_token_accuracy": 0.239899343252182,
      "num_tokens": 19977661.0,
      "step": 10990
    },
    {
      "entropy": 5.603775501251221,
      "epoch": 9.446067898581864,
      "grad_norm": 1.4765625,
      "learning_rate": 5.396297147971116e-05,
      "loss": 4.774,
      "mean_token_accuracy": 0.23459887355566025,
      "num_tokens": 19986359.0,
      "step": 10995
    },
    {
      "entropy": 5.5279768943786625,
      "epoch": 9.45036527718092,
      "grad_norm": 1.421875,
      "learning_rate": 5.390108552157935e-05,
      "loss": 4.6849,
      "mean_token_accuracy": 0.247470885515213,
      "num_tokens": 19994618.0,
      "step": 11000
    },
    {
      "epoch": 9.45036527718092,
      "eval_entropy": 5.443561657055004,
      "eval_loss": 5.928343772888184,
      "eval_mean_token_accuracy": 0.18021051776972977,
      "eval_num_tokens": 19994618.0,
      "eval_runtime": 2.6397,
      "eval_samples_per_second": 1344.455,
      "eval_steps_per_second": 168.199,
      "step": 11000
    },
    {
      "entropy": 5.569438886642456,
      "epoch": 9.454662655779973,
      "grad_norm": 1.4296875,
      "learning_rate": 5.383968235510427e-05,
      "loss": 4.7836,
      "mean_token_accuracy": 0.23856076449155808,
      "num_tokens": 20004050.0,
      "step": 11005
    },
    {
      "entropy": 5.54927978515625,
      "epoch": 9.458960034379029,
      "grad_norm": 1.4375,
      "learning_rate": 5.377876211436592e-05,
      "loss": 4.7357,
      "mean_token_accuracy": 0.24245603233575821,
      "num_tokens": 20013480.0,
      "step": 11010
    },
    {
      "entropy": 5.626982402801514,
      "epoch": 9.463257412978084,
      "grad_norm": 1.578125,
      "learning_rate": 5.371832493238973e-05,
      "loss": 4.8181,
      "mean_token_accuracy": 0.23222310096025467,
      "num_tokens": 20022133.0,
      "step": 11015
    },
    {
      "entropy": 5.500302076339722,
      "epoch": 9.467554791577138,
      "grad_norm": 1.5,
      "learning_rate": 5.365837094114639e-05,
      "loss": 4.7017,
      "mean_token_accuracy": 0.24035543948411942,
      "num_tokens": 20031462.0,
      "step": 11020
    },
    {
      "entropy": 5.60685625076294,
      "epoch": 9.471852170176193,
      "grad_norm": 1.53125,
      "learning_rate": 5.3598900271551396e-05,
      "loss": 4.8351,
      "mean_token_accuracy": 0.2273373618721962,
      "num_tokens": 20042055.0,
      "step": 11025
    },
    {
      "entropy": 5.552582740783691,
      "epoch": 9.476149548775247,
      "grad_norm": 1.28125,
      "learning_rate": 5.353991305346499e-05,
      "loss": 4.7242,
      "mean_token_accuracy": 0.23951998949050904,
      "num_tokens": 20051379.0,
      "step": 11030
    },
    {
      "entropy": 5.506077909469605,
      "epoch": 9.480446927374302,
      "grad_norm": 1.5,
      "learning_rate": 5.348140941569165e-05,
      "loss": 4.6659,
      "mean_token_accuracy": 0.24881837666034698,
      "num_tokens": 20060289.0,
      "step": 11035
    },
    {
      "entropy": 5.508519983291626,
      "epoch": 9.484744305973356,
      "grad_norm": 1.4296875,
      "learning_rate": 5.342338948597989e-05,
      "loss": 4.6968,
      "mean_token_accuracy": 0.23680495619773864,
      "num_tokens": 20070133.0,
      "step": 11040
    },
    {
      "entropy": 5.5720069885253904,
      "epoch": 9.489041684572411,
      "grad_norm": 1.484375,
      "learning_rate": 5.336585339102209e-05,
      "loss": 4.7199,
      "mean_token_accuracy": 0.23853890746831893,
      "num_tokens": 20078384.0,
      "step": 11045
    },
    {
      "entropy": 5.498673439025879,
      "epoch": 9.493339063171465,
      "grad_norm": 1.4765625,
      "learning_rate": 5.33088012564541e-05,
      "loss": 4.6772,
      "mean_token_accuracy": 0.2476629376411438,
      "num_tokens": 20087116.0,
      "step": 11050
    },
    {
      "entropy": 5.46187047958374,
      "epoch": 9.49763644177052,
      "grad_norm": 1.453125,
      "learning_rate": 5.3252233206854955e-05,
      "loss": 4.6903,
      "mean_token_accuracy": 0.24791076332330703,
      "num_tokens": 20096918.0,
      "step": 11055
    },
    {
      "entropy": 5.528538084030151,
      "epoch": 9.501933820369574,
      "grad_norm": 1.453125,
      "learning_rate": 5.3196149365746656e-05,
      "loss": 4.8271,
      "mean_token_accuracy": 0.22837469279766082,
      "num_tokens": 20107602.0,
      "step": 11060
    },
    {
      "entropy": 5.490460634231567,
      "epoch": 9.50623119896863,
      "grad_norm": 1.296875,
      "learning_rate": 5.31405498555939e-05,
      "loss": 4.7062,
      "mean_token_accuracy": 0.24608968198299408,
      "num_tokens": 20117732.0,
      "step": 11065
    },
    {
      "entropy": 5.627804183959961,
      "epoch": 9.510528577567683,
      "grad_norm": 1.3359375,
      "learning_rate": 5.308543479780384e-05,
      "loss": 4.8324,
      "mean_token_accuracy": 0.2293786123394966,
      "num_tokens": 20127765.0,
      "step": 11070
    },
    {
      "entropy": 5.573545598983765,
      "epoch": 9.514825956166739,
      "grad_norm": 1.5,
      "learning_rate": 5.303080431272567e-05,
      "loss": 4.8297,
      "mean_token_accuracy": 0.23137810081243515,
      "num_tokens": 20137135.0,
      "step": 11075
    },
    {
      "entropy": 5.538729810714722,
      "epoch": 9.519123334765792,
      "grad_norm": 1.4921875,
      "learning_rate": 5.297665851965055e-05,
      "loss": 4.7391,
      "mean_token_accuracy": 0.24402787089347838,
      "num_tokens": 20146156.0,
      "step": 11080
    },
    {
      "entropy": 5.597339344024658,
      "epoch": 9.523420713364848,
      "grad_norm": 1.59375,
      "learning_rate": 5.292299753681129e-05,
      "loss": 4.7505,
      "mean_token_accuracy": 0.24292106926441193,
      "num_tokens": 20154433.0,
      "step": 11085
    },
    {
      "entropy": 5.5522661209106445,
      "epoch": 9.527718091963902,
      "grad_norm": 1.3984375,
      "learning_rate": 5.286982148138196e-05,
      "loss": 4.7175,
      "mean_token_accuracy": 0.23762956261634827,
      "num_tokens": 20164209.0,
      "step": 11090
    },
    {
      "entropy": 5.5283619403839115,
      "epoch": 9.532015470562957,
      "grad_norm": 1.515625,
      "learning_rate": 5.281713046947787e-05,
      "loss": 4.7416,
      "mean_token_accuracy": 0.23554776310920716,
      "num_tokens": 20173116.0,
      "step": 11095
    },
    {
      "entropy": 5.581348609924317,
      "epoch": 9.53631284916201,
      "grad_norm": 1.515625,
      "learning_rate": 5.2764924616155116e-05,
      "loss": 4.8065,
      "mean_token_accuracy": 0.23168325424194336,
      "num_tokens": 20182933.0,
      "step": 11100
    },
    {
      "entropy": 5.535536575317383,
      "epoch": 9.540610227761066,
      "grad_norm": 1.5703125,
      "learning_rate": 5.271320403541038e-05,
      "loss": 4.7026,
      "mean_token_accuracy": 0.24211829155683517,
      "num_tokens": 20191899.0,
      "step": 11105
    },
    {
      "entropy": 5.556659936904907,
      "epoch": 9.54490760636012,
      "grad_norm": 1.453125,
      "learning_rate": 5.266196884018081e-05,
      "loss": 4.7518,
      "mean_token_accuracy": 0.23577206581830978,
      "num_tokens": 20201108.0,
      "step": 11110
    },
    {
      "entropy": 5.504491662979126,
      "epoch": 9.549204984959175,
      "grad_norm": 1.4765625,
      "learning_rate": 5.2611219142343494e-05,
      "loss": 4.7484,
      "mean_token_accuracy": 0.24216720461845398,
      "num_tokens": 20209137.0,
      "step": 11115
    },
    {
      "entropy": 5.50584979057312,
      "epoch": 9.553502363558229,
      "grad_norm": 1.3359375,
      "learning_rate": 5.2560955052715574e-05,
      "loss": 4.6619,
      "mean_token_accuracy": 0.24441113024950029,
      "num_tokens": 20217879.0,
      "step": 11120
    },
    {
      "entropy": 5.542237186431885,
      "epoch": 9.557799742157284,
      "grad_norm": 1.46875,
      "learning_rate": 5.2511176681053704e-05,
      "loss": 4.7594,
      "mean_token_accuracy": 0.23990135341882707,
      "num_tokens": 20226783.0,
      "step": 11125
    },
    {
      "entropy": 5.550592565536499,
      "epoch": 9.562097120756338,
      "grad_norm": 1.5078125,
      "learning_rate": 5.246188413605393e-05,
      "loss": 4.7425,
      "mean_token_accuracy": 0.24169102907180787,
      "num_tokens": 20235008.0,
      "step": 11130
    },
    {
      "entropy": 5.5253846645355225,
      "epoch": 9.566394499355393,
      "grad_norm": 1.5078125,
      "learning_rate": 5.241307752535149e-05,
      "loss": 4.6483,
      "mean_token_accuracy": 0.25130110830068586,
      "num_tokens": 20243157.0,
      "step": 11135
    },
    {
      "entropy": 5.560060596466064,
      "epoch": 9.570691877954447,
      "grad_norm": 1.390625,
      "learning_rate": 5.236475695552052e-05,
      "loss": 4.7289,
      "mean_token_accuracy": 0.24112499207258226,
      "num_tokens": 20251548.0,
      "step": 11140
    },
    {
      "entropy": 5.437600040435791,
      "epoch": 9.574989256553502,
      "grad_norm": 1.5390625,
      "learning_rate": 5.2316922532073796e-05,
      "loss": 4.6173,
      "mean_token_accuracy": 0.25002511292696,
      "num_tokens": 20260954.0,
      "step": 11145
    },
    {
      "entropy": 5.624392700195313,
      "epoch": 9.579286635152556,
      "grad_norm": 1.4296875,
      "learning_rate": 5.226957435946265e-05,
      "loss": 4.8293,
      "mean_token_accuracy": 0.22470595687627792,
      "num_tokens": 20269976.0,
      "step": 11150
    },
    {
      "entropy": 5.554257202148437,
      "epoch": 9.583584013751612,
      "grad_norm": 1.4375,
      "learning_rate": 5.2222712541076464e-05,
      "loss": 4.7946,
      "mean_token_accuracy": 0.23425038307905197,
      "num_tokens": 20279824.0,
      "step": 11155
    },
    {
      "entropy": 5.540955448150635,
      "epoch": 9.587881392350667,
      "grad_norm": 1.4375,
      "learning_rate": 5.217633717924282e-05,
      "loss": 4.7219,
      "mean_token_accuracy": 0.25006516724824907,
      "num_tokens": 20289284.0,
      "step": 11160
    },
    {
      "entropy": 5.572676515579223,
      "epoch": 9.59217877094972,
      "grad_norm": 1.3125,
      "learning_rate": 5.213044837522689e-05,
      "loss": 4.6807,
      "mean_token_accuracy": 0.25359575897455217,
      "num_tokens": 20297905.0,
      "step": 11165
    },
    {
      "entropy": 5.512086963653564,
      "epoch": 9.596476149548776,
      "grad_norm": 1.5703125,
      "learning_rate": 5.208504622923154e-05,
      "loss": 4.7027,
      "mean_token_accuracy": 0.24958239495754242,
      "num_tokens": 20307219.0,
      "step": 11170
    },
    {
      "entropy": 5.547673749923706,
      "epoch": 9.60077352814783,
      "grad_norm": 1.4375,
      "learning_rate": 5.204013084039687e-05,
      "loss": 4.754,
      "mean_token_accuracy": 0.23772649317979813,
      "num_tokens": 20316369.0,
      "step": 11175
    },
    {
      "entropy": 5.506791448593139,
      "epoch": 9.605070906746885,
      "grad_norm": 1.4765625,
      "learning_rate": 5.199570230680017e-05,
      "loss": 4.7338,
      "mean_token_accuracy": 0.24074728190898895,
      "num_tokens": 20325764.0,
      "step": 11180
    },
    {
      "entropy": 5.626205492019653,
      "epoch": 9.609368285345939,
      "grad_norm": 1.484375,
      "learning_rate": 5.19517607254556e-05,
      "loss": 4.8329,
      "mean_token_accuracy": 0.22183477282524108,
      "num_tokens": 20334671.0,
      "step": 11185
    },
    {
      "entropy": 5.522506809234619,
      "epoch": 9.613665663944994,
      "grad_norm": 1.5390625,
      "learning_rate": 5.190830619231397e-05,
      "loss": 4.7246,
      "mean_token_accuracy": 0.24368579983711242,
      "num_tokens": 20343585.0,
      "step": 11190
    },
    {
      "entropy": 5.53656120300293,
      "epoch": 9.617963042544048,
      "grad_norm": 1.484375,
      "learning_rate": 5.186533880226263e-05,
      "loss": 4.6832,
      "mean_token_accuracy": 0.2463853880763054,
      "num_tokens": 20352569.0,
      "step": 11195
    },
    {
      "entropy": 5.584325551986694,
      "epoch": 9.622260421143103,
      "grad_norm": 1.5546875,
      "learning_rate": 5.1822858649125197e-05,
      "loss": 4.7783,
      "mean_token_accuracy": 0.23171369433403016,
      "num_tokens": 20361479.0,
      "step": 11200
    },
    {
      "entropy": 5.6076744556427,
      "epoch": 9.626557799742157,
      "grad_norm": 1.578125,
      "learning_rate": 5.178086582566134e-05,
      "loss": 4.7873,
      "mean_token_accuracy": 0.23102419525384904,
      "num_tokens": 20370286.0,
      "step": 11205
    },
    {
      "entropy": 5.538750076293946,
      "epoch": 9.630855178341212,
      "grad_norm": 1.46875,
      "learning_rate": 5.1739360423566596e-05,
      "loss": 4.7674,
      "mean_token_accuracy": 0.23591473400592805,
      "num_tokens": 20379253.0,
      "step": 11210
    },
    {
      "entropy": 5.518057918548584,
      "epoch": 9.635152556940266,
      "grad_norm": 1.453125,
      "learning_rate": 5.16983425334722e-05,
      "loss": 4.7153,
      "mean_token_accuracy": 0.24006852358579636,
      "num_tokens": 20387796.0,
      "step": 11215
    },
    {
      "entropy": 5.511827611923218,
      "epoch": 9.639449935539322,
      "grad_norm": 1.484375,
      "learning_rate": 5.1657812244944796e-05,
      "loss": 4.6537,
      "mean_token_accuracy": 0.24130041152238846,
      "num_tokens": 20396209.0,
      "step": 11220
    },
    {
      "entropy": 5.613779830932617,
      "epoch": 9.643747314138375,
      "grad_norm": 1.4921875,
      "learning_rate": 5.1617769646486344e-05,
      "loss": 4.8282,
      "mean_token_accuracy": 0.23585615754127504,
      "num_tokens": 20405102.0,
      "step": 11225
    },
    {
      "entropy": 5.583826637268066,
      "epoch": 9.64804469273743,
      "grad_norm": 1.3828125,
      "learning_rate": 5.157821482553389e-05,
      "loss": 4.8626,
      "mean_token_accuracy": 0.22848095148801803,
      "num_tokens": 20415126.0,
      "step": 11230
    },
    {
      "entropy": 5.580589628219604,
      "epoch": 9.652342071336484,
      "grad_norm": 1.5,
      "learning_rate": 5.153914786845932e-05,
      "loss": 4.799,
      "mean_token_accuracy": 0.23437997102737426,
      "num_tokens": 20424166.0,
      "step": 11235
    },
    {
      "entropy": 5.489545631408691,
      "epoch": 9.65663944993554,
      "grad_norm": 1.6015625,
      "learning_rate": 5.1500568860569285e-05,
      "loss": 4.6164,
      "mean_token_accuracy": 0.2578002229332924,
      "num_tokens": 20431984.0,
      "step": 11240
    },
    {
      "entropy": 5.491452598571778,
      "epoch": 9.660936828534593,
      "grad_norm": 1.4921875,
      "learning_rate": 5.1462477886104904e-05,
      "loss": 4.681,
      "mean_token_accuracy": 0.24954406768083573,
      "num_tokens": 20440923.0,
      "step": 11245
    },
    {
      "entropy": 5.562239503860473,
      "epoch": 9.665234207133649,
      "grad_norm": 1.5234375,
      "learning_rate": 5.1424875028241625e-05,
      "loss": 4.6961,
      "mean_token_accuracy": 0.2422264337539673,
      "num_tokens": 20449525.0,
      "step": 11250
    },
    {
      "entropy": 5.547700357437134,
      "epoch": 9.669531585732702,
      "grad_norm": 1.34375,
      "learning_rate": 5.138776036908911e-05,
      "loss": 4.7718,
      "mean_token_accuracy": 0.23322735726833344,
      "num_tokens": 20459193.0,
      "step": 11255
    },
    {
      "entropy": 5.537365674972534,
      "epoch": 9.673828964331758,
      "grad_norm": 1.6640625,
      "learning_rate": 5.135113398969091e-05,
      "loss": 4.7258,
      "mean_token_accuracy": 0.23440928012132645,
      "num_tokens": 20467564.0,
      "step": 11260
    },
    {
      "entropy": 5.626916456222534,
      "epoch": 9.678126342930812,
      "grad_norm": 1.421875,
      "learning_rate": 5.131499597002437e-05,
      "loss": 4.8461,
      "mean_token_accuracy": 0.2367245838046074,
      "num_tokens": 20476573.0,
      "step": 11265
    },
    {
      "entropy": 5.543498516082764,
      "epoch": 9.682423721529867,
      "grad_norm": 1.3984375,
      "learning_rate": 5.12793463890005e-05,
      "loss": 4.7844,
      "mean_token_accuracy": 0.237060983479023,
      "num_tokens": 20486788.0,
      "step": 11270
    },
    {
      "entropy": 5.590712547302246,
      "epoch": 9.68672110012892,
      "grad_norm": 1.34375,
      "learning_rate": 5.124418532446376e-05,
      "loss": 4.8091,
      "mean_token_accuracy": 0.2270214393734932,
      "num_tokens": 20495982.0,
      "step": 11275
    },
    {
      "entropy": 5.591723012924194,
      "epoch": 9.691018478727976,
      "grad_norm": 1.3828125,
      "learning_rate": 5.120951285319187e-05,
      "loss": 4.8184,
      "mean_token_accuracy": 0.22862080931663514,
      "num_tokens": 20504678.0,
      "step": 11280
    },
    {
      "entropy": 5.531742334365845,
      "epoch": 9.69531585732703,
      "grad_norm": 1.515625,
      "learning_rate": 5.1175329050895584e-05,
      "loss": 4.6385,
      "mean_token_accuracy": 0.24624322205781937,
      "num_tokens": 20513362.0,
      "step": 11285
    },
    {
      "entropy": 5.539655017852783,
      "epoch": 9.699613235926085,
      "grad_norm": 1.3984375,
      "learning_rate": 5.114163399221871e-05,
      "loss": 4.7929,
      "mean_token_accuracy": 0.23633970618247985,
      "num_tokens": 20522280.0,
      "step": 11290
    },
    {
      "entropy": 5.528562355041504,
      "epoch": 9.703910614525139,
      "grad_norm": 1.5390625,
      "learning_rate": 5.110842775073778e-05,
      "loss": 4.7185,
      "mean_token_accuracy": 0.24560501277446747,
      "num_tokens": 20531175.0,
      "step": 11295
    },
    {
      "entropy": 5.52716097831726,
      "epoch": 9.708207993124194,
      "grad_norm": 1.4140625,
      "learning_rate": 5.107571039896196e-05,
      "loss": 4.7612,
      "mean_token_accuracy": 0.2334544837474823,
      "num_tokens": 20540592.0,
      "step": 11300
    },
    {
      "entropy": 5.61093692779541,
      "epoch": 9.71250537172325,
      "grad_norm": 1.546875,
      "learning_rate": 5.1043482008332864e-05,
      "loss": 4.8516,
      "mean_token_accuracy": 0.22909058183431624,
      "num_tokens": 20549963.0,
      "step": 11305
    },
    {
      "entropy": 5.5275146007537845,
      "epoch": 9.716802750322303,
      "grad_norm": 1.3984375,
      "learning_rate": 5.1011742649224394e-05,
      "loss": 4.7597,
      "mean_token_accuracy": 0.2434886947274208,
      "num_tokens": 20558928.0,
      "step": 11310
    },
    {
      "entropy": 5.526972007751465,
      "epoch": 9.721100128921359,
      "grad_norm": 1.5546875,
      "learning_rate": 5.098049239094267e-05,
      "loss": 4.6389,
      "mean_token_accuracy": 0.24909676909446715,
      "num_tokens": 20566483.0,
      "step": 11315
    },
    {
      "entropy": 5.581226873397827,
      "epoch": 9.725397507520412,
      "grad_norm": 1.484375,
      "learning_rate": 5.094973130172573e-05,
      "loss": 4.7548,
      "mean_token_accuracy": 0.23466918617486954,
      "num_tokens": 20575349.0,
      "step": 11320
    },
    {
      "entropy": 5.59463062286377,
      "epoch": 9.729694886119468,
      "grad_norm": 1.46875,
      "learning_rate": 5.09194594487435e-05,
      "loss": 4.7699,
      "mean_token_accuracy": 0.23163093030452728,
      "num_tokens": 20584235.0,
      "step": 11325
    },
    {
      "entropy": 5.49168815612793,
      "epoch": 9.733992264718522,
      "grad_norm": 1.546875,
      "learning_rate": 5.088967689809763e-05,
      "loss": 4.6754,
      "mean_token_accuracy": 0.24200426489114762,
      "num_tokens": 20593705.0,
      "step": 11330
    },
    {
      "entropy": 5.5771993637084964,
      "epoch": 9.738289643317577,
      "grad_norm": 1.25,
      "learning_rate": 5.086038371482128e-05,
      "loss": 4.7901,
      "mean_token_accuracy": 0.23622879087924958,
      "num_tokens": 20603886.0,
      "step": 11335
    },
    {
      "entropy": 5.388237810134887,
      "epoch": 9.74258702191663,
      "grad_norm": 1.4453125,
      "learning_rate": 5.0831579962879074e-05,
      "loss": 4.6241,
      "mean_token_accuracy": 0.25767128467559813,
      "num_tokens": 20613181.0,
      "step": 11340
    },
    {
      "entropy": 5.574814939498902,
      "epoch": 9.746884400515686,
      "grad_norm": 1.4921875,
      "learning_rate": 5.080326570516686e-05,
      "loss": 4.7482,
      "mean_token_accuracy": 0.2339532494544983,
      "num_tokens": 20620938.0,
      "step": 11345
    },
    {
      "entropy": 5.594657754898071,
      "epoch": 9.75118177911474,
      "grad_norm": 1.40625,
      "learning_rate": 5.077544100351172e-05,
      "loss": 4.828,
      "mean_token_accuracy": 0.22980117201805114,
      "num_tokens": 20630992.0,
      "step": 11350
    },
    {
      "entropy": 5.613297796249389,
      "epoch": 9.755479157713795,
      "grad_norm": 1.53125,
      "learning_rate": 5.0748105918671616e-05,
      "loss": 4.8629,
      "mean_token_accuracy": 0.23168486803770066,
      "num_tokens": 20640543.0,
      "step": 11355
    },
    {
      "entropy": 5.587474203109741,
      "epoch": 9.759776536312849,
      "grad_norm": 1.40625,
      "learning_rate": 5.072126051033551e-05,
      "loss": 4.7484,
      "mean_token_accuracy": 0.23580139577388765,
      "num_tokens": 20649814.0,
      "step": 11360
    },
    {
      "entropy": 5.581478548049927,
      "epoch": 9.764073914911904,
      "grad_norm": 1.4296875,
      "learning_rate": 5.069490483712298e-05,
      "loss": 4.791,
      "mean_token_accuracy": 0.23648376315832137,
      "num_tokens": 20658215.0,
      "step": 11365
    },
    {
      "entropy": 5.575561571121216,
      "epoch": 9.768371293510958,
      "grad_norm": 1.4296875,
      "learning_rate": 5.066903895658433e-05,
      "loss": 4.7595,
      "mean_token_accuracy": 0.237857486307621,
      "num_tokens": 20667412.0,
      "step": 11370
    },
    {
      "entropy": 5.518033885955811,
      "epoch": 9.772668672110013,
      "grad_norm": 1.546875,
      "learning_rate": 5.064366292520028e-05,
      "loss": 4.7616,
      "mean_token_accuracy": 0.2352076694369316,
      "num_tokens": 20676366.0,
      "step": 11375
    },
    {
      "entropy": 5.609922790527344,
      "epoch": 9.776966050709067,
      "grad_norm": 1.3046875,
      "learning_rate": 5.061877679838192e-05,
      "loss": 4.8579,
      "mean_token_accuracy": 0.21730295717716216,
      "num_tokens": 20686872.0,
      "step": 11380
    },
    {
      "entropy": 5.572014713287354,
      "epoch": 9.781263429308122,
      "grad_norm": 1.5,
      "learning_rate": 5.059438063047066e-05,
      "loss": 4.6947,
      "mean_token_accuracy": 0.2450913593173027,
      "num_tokens": 20695162.0,
      "step": 11385
    },
    {
      "entropy": 5.572461843490601,
      "epoch": 9.785560807907176,
      "grad_norm": 1.5078125,
      "learning_rate": 5.057047447473796e-05,
      "loss": 4.7422,
      "mean_token_accuracy": 0.23934031128883362,
      "num_tokens": 20703897.0,
      "step": 11390
    },
    {
      "entropy": 5.505301189422608,
      "epoch": 9.789858186506232,
      "grad_norm": 1.390625,
      "learning_rate": 5.054705838338529e-05,
      "loss": 4.7117,
      "mean_token_accuracy": 0.24422628730535506,
      "num_tokens": 20712115.0,
      "step": 11395
    },
    {
      "entropy": 5.5154540061950685,
      "epoch": 9.794155565105285,
      "grad_norm": 1.390625,
      "learning_rate": 5.052413240754404e-05,
      "loss": 4.6965,
      "mean_token_accuracy": 0.24811999648809432,
      "num_tokens": 20721073.0,
      "step": 11400
    },
    {
      "entropy": 5.527085781097412,
      "epoch": 9.79845294370434,
      "grad_norm": 1.3046875,
      "learning_rate": 5.0501696597275376e-05,
      "loss": 4.7272,
      "mean_token_accuracy": 0.23672485053539277,
      "num_tokens": 20731370.0,
      "step": 11405
    },
    {
      "entropy": 5.563812637329102,
      "epoch": 9.802750322303394,
      "grad_norm": 1.46875,
      "learning_rate": 5.047975100157018e-05,
      "loss": 4.7929,
      "mean_token_accuracy": 0.2345522254705429,
      "num_tokens": 20740744.0,
      "step": 11410
    },
    {
      "entropy": 5.552879047393799,
      "epoch": 9.80704770090245,
      "grad_norm": 1.3515625,
      "learning_rate": 5.045829566834879e-05,
      "loss": 4.7578,
      "mean_token_accuracy": 0.24406700432300568,
      "num_tokens": 20749838.0,
      "step": 11415
    },
    {
      "entropy": 5.52188663482666,
      "epoch": 9.811345079501503,
      "grad_norm": 1.40625,
      "learning_rate": 5.043733064446113e-05,
      "loss": 4.6661,
      "mean_token_accuracy": 0.25192087292671206,
      "num_tokens": 20759689.0,
      "step": 11420
    },
    {
      "entropy": 5.519135904312134,
      "epoch": 9.815642458100559,
      "grad_norm": 1.5234375,
      "learning_rate": 5.041685597568641e-05,
      "loss": 4.7144,
      "mean_token_accuracy": 0.24467841386795045,
      "num_tokens": 20769234.0,
      "step": 11425
    },
    {
      "entropy": 5.540931367874146,
      "epoch": 9.819939836699612,
      "grad_norm": 1.484375,
      "learning_rate": 5.039687170673315e-05,
      "loss": 4.7908,
      "mean_token_accuracy": 0.23664245307445525,
      "num_tokens": 20778794.0,
      "step": 11430
    },
    {
      "entropy": 5.553414011001587,
      "epoch": 9.824237215298668,
      "grad_norm": 1.4296875,
      "learning_rate": 5.037737788123895e-05,
      "loss": 4.7417,
      "mean_token_accuracy": 0.24065371751785278,
      "num_tokens": 20788327.0,
      "step": 11435
    },
    {
      "entropy": 5.58979229927063,
      "epoch": 9.828534593897722,
      "grad_norm": 1.375,
      "learning_rate": 5.03583745417706e-05,
      "loss": 4.7618,
      "mean_token_accuracy": 0.23460249304771424,
      "num_tokens": 20797161.0,
      "step": 11440
    },
    {
      "entropy": 5.657240629196167,
      "epoch": 9.832831972496777,
      "grad_norm": 1.375,
      "learning_rate": 5.033986172982375e-05,
      "loss": 4.8438,
      "mean_token_accuracy": 0.23100274354219436,
      "num_tokens": 20806264.0,
      "step": 11445
    },
    {
      "entropy": 5.5797782897949215,
      "epoch": 9.837129351095832,
      "grad_norm": 1.40625,
      "learning_rate": 5.0321839485823014e-05,
      "loss": 4.8075,
      "mean_token_accuracy": 0.2275300994515419,
      "num_tokens": 20815569.0,
      "step": 11450
    },
    {
      "entropy": 5.615515422821045,
      "epoch": 9.841426729694886,
      "grad_norm": 1.4296875,
      "learning_rate": 5.030430784912177e-05,
      "loss": 4.8769,
      "mean_token_accuracy": 0.2271632432937622,
      "num_tokens": 20825234.0,
      "step": 11455
    },
    {
      "entropy": 5.488658237457275,
      "epoch": 9.845724108293942,
      "grad_norm": 1.4921875,
      "learning_rate": 5.0287266858002054e-05,
      "loss": 4.7266,
      "mean_token_accuracy": 0.24024616777896882,
      "num_tokens": 20834601.0,
      "step": 11460
    },
    {
      "entropy": 5.574060869216919,
      "epoch": 9.850021486892995,
      "grad_norm": 1.3203125,
      "learning_rate": 5.027071654967465e-05,
      "loss": 4.7467,
      "mean_token_accuracy": 0.2403283268213272,
      "num_tokens": 20843686.0,
      "step": 11465
    },
    {
      "entropy": 5.5605486869812015,
      "epoch": 9.85431886549205,
      "grad_norm": 1.4375,
      "learning_rate": 5.025465696027875e-05,
      "loss": 4.7655,
      "mean_token_accuracy": 0.24174125641584396,
      "num_tokens": 20852009.0,
      "step": 11470
    },
    {
      "entropy": 5.543330717086792,
      "epoch": 9.858616244091104,
      "grad_norm": 1.4375,
      "learning_rate": 5.023908812488211e-05,
      "loss": 4.7116,
      "mean_token_accuracy": 0.24110664129257203,
      "num_tokens": 20861241.0,
      "step": 11475
    },
    {
      "entropy": 5.565051937103272,
      "epoch": 9.86291362269016,
      "grad_norm": 1.421875,
      "learning_rate": 5.022401007748087e-05,
      "loss": 4.7804,
      "mean_token_accuracy": 0.23628752082586288,
      "num_tokens": 20869774.0,
      "step": 11480
    },
    {
      "entropy": 5.5596190929412845,
      "epoch": 9.867211001289213,
      "grad_norm": 1.484375,
      "learning_rate": 5.0209422850999414e-05,
      "loss": 4.7906,
      "mean_token_accuracy": 0.23690288364887238,
      "num_tokens": 20878844.0,
      "step": 11485
    },
    {
      "entropy": 5.53383617401123,
      "epoch": 9.871508379888269,
      "grad_norm": 1.40625,
      "learning_rate": 5.019532647729046e-05,
      "loss": 4.7886,
      "mean_token_accuracy": 0.2342674031853676,
      "num_tokens": 20888054.0,
      "step": 11490
    },
    {
      "entropy": 5.579972553253174,
      "epoch": 9.875805758487322,
      "grad_norm": 1.2734375,
      "learning_rate": 5.0181720987134815e-05,
      "loss": 4.753,
      "mean_token_accuracy": 0.23917244374752045,
      "num_tokens": 20897752.0,
      "step": 11495
    },
    {
      "entropy": 5.450816297531128,
      "epoch": 9.880103137086378,
      "grad_norm": 1.5625,
      "learning_rate": 5.016860641024143e-05,
      "loss": 4.6561,
      "mean_token_accuracy": 0.2509209454059601,
      "num_tokens": 20906186.0,
      "step": 11500
    },
    {
      "epoch": 9.880103137086378,
      "eval_entropy": 5.432724556944391,
      "eval_loss": 5.929876804351807,
      "eval_mean_token_accuracy": 0.18007116308351895,
      "eval_num_tokens": 20906186.0,
      "eval_runtime": 2.8456,
      "eval_samples_per_second": 1247.186,
      "eval_steps_per_second": 156.03,
      "step": 11500
    }
  ],
  "logging_steps": 5,
  "max_steps": 11630,
  "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": 3.1721243602176e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}