{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.243374797187669,
  "eval_steps": 1000,
  "global_step": 3000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 6.745217990875244,
      "epoch": 0.005408328826392645,
      "grad_norm": 1.1875,
      "learning_rate": 2e-06,
      "loss": 6.4399,
      "mean_token_accuracy": 0.12539481073617936,
      "num_tokens": 9564.0,
      "step": 5
    },
    {
      "entropy": 6.7341703414917,
      "epoch": 0.01081665765278529,
      "grad_norm": 1.1484375,
      "learning_rate": 4.5e-06,
      "loss": 6.4966,
      "mean_token_accuracy": 0.12570957243442535,
      "num_tokens": 20265.0,
      "step": 10
    },
    {
      "entropy": 6.810567569732666,
      "epoch": 0.016224986479177934,
      "grad_norm": 1.171875,
      "learning_rate": 7e-06,
      "loss": 6.6044,
      "mean_token_accuracy": 0.12070711851119995,
      "num_tokens": 30733.0,
      "step": 15
    },
    {
      "entropy": 6.691426467895508,
      "epoch": 0.02163331530557058,
      "grad_norm": 1.1015625,
      "learning_rate": 9.5e-06,
      "loss": 6.5463,
      "mean_token_accuracy": 0.1250626839697361,
      "num_tokens": 41633.0,
      "step": 20
    },
    {
      "entropy": 6.713900947570801,
      "epoch": 0.02704164413196322,
      "grad_norm": 1.1484375,
      "learning_rate": 1.2e-05,
      "loss": 6.6055,
      "mean_token_accuracy": 0.12766451835632325,
      "num_tokens": 53776.0,
      "step": 25
    },
    {
      "entropy": 6.794030284881591,
      "epoch": 0.03244997295835587,
      "grad_norm": 1.1796875,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 6.55,
      "mean_token_accuracy": 0.13146025240421294,
      "num_tokens": 64936.0,
      "step": 30
    },
    {
      "entropy": 6.735936450958252,
      "epoch": 0.03785830178474851,
      "grad_norm": 1.0859375,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 6.5551,
      "mean_token_accuracy": 0.12577222511172295,
      "num_tokens": 76664.0,
      "step": 35
    },
    {
      "entropy": 6.720685720443726,
      "epoch": 0.04326663061114116,
      "grad_norm": 1.1171875,
      "learning_rate": 1.95e-05,
      "loss": 6.4589,
      "mean_token_accuracy": 0.12645927220582961,
      "num_tokens": 87188.0,
      "step": 40
    },
    {
      "entropy": 6.770437097549438,
      "epoch": 0.048674959437533805,
      "grad_norm": 1.0859375,
      "learning_rate": 2.2e-05,
      "loss": 6.6097,
      "mean_token_accuracy": 0.1324235811829567,
      "num_tokens": 99337.0,
      "step": 45
    },
    {
      "entropy": 6.7956713199615475,
      "epoch": 0.05408328826392644,
      "grad_norm": 1.0625,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 6.5528,
      "mean_token_accuracy": 0.1259615793824196,
      "num_tokens": 110870.0,
      "step": 50
    },
    {
      "entropy": 6.821929407119751,
      "epoch": 0.05949161709031909,
      "grad_norm": 1.0625,
      "learning_rate": 2.7e-05,
      "loss": 6.4581,
      "mean_token_accuracy": 0.13586550429463387,
      "num_tokens": 122183.0,
      "step": 55
    },
    {
      "entropy": 6.810019540786743,
      "epoch": 0.06489994591671173,
      "grad_norm": 1.078125,
      "learning_rate": 2.95e-05,
      "loss": 6.471,
      "mean_token_accuracy": 0.134011822193861,
      "num_tokens": 133676.0,
      "step": 60
    },
    {
      "entropy": 6.7516721248626705,
      "epoch": 0.07030827474310439,
      "grad_norm": 1.046875,
      "learning_rate": 3.2e-05,
      "loss": 6.5101,
      "mean_token_accuracy": 0.13094406202435493,
      "num_tokens": 144821.0,
      "step": 65
    },
    {
      "entropy": 6.73709626197815,
      "epoch": 0.07571660356949703,
      "grad_norm": 0.98828125,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 6.4945,
      "mean_token_accuracy": 0.13041364997625352,
      "num_tokens": 156873.0,
      "step": 70
    },
    {
      "entropy": 6.649084663391113,
      "epoch": 0.08112493239588967,
      "grad_norm": 1.0546875,
      "learning_rate": 3.7e-05,
      "loss": 6.4459,
      "mean_token_accuracy": 0.12853534668684005,
      "num_tokens": 168794.0,
      "step": 75
    },
    {
      "entropy": 6.839602088928222,
      "epoch": 0.08653326122228232,
      "grad_norm": 1.0078125,
      "learning_rate": 3.95e-05,
      "loss": 6.603,
      "mean_token_accuracy": 0.12388623431324959,
      "num_tokens": 180756.0,
      "step": 80
    },
    {
      "entropy": 6.882611703872681,
      "epoch": 0.09194159004867496,
      "grad_norm": 1.125,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 6.5729,
      "mean_token_accuracy": 0.12938066497445105,
      "num_tokens": 191127.0,
      "step": 85
    },
    {
      "entropy": 6.704812479019165,
      "epoch": 0.09734991887506761,
      "grad_norm": 0.90625,
      "learning_rate": 4.45e-05,
      "loss": 6.455,
      "mean_token_accuracy": 0.13043807074427605,
      "num_tokens": 203352.0,
      "step": 90
    },
    {
      "entropy": 6.752549362182617,
      "epoch": 0.10275824770146025,
      "grad_norm": 1.2265625,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 6.4423,
      "mean_token_accuracy": 0.1375568687915802,
      "num_tokens": 213810.0,
      "step": 95
    },
    {
      "entropy": 6.83305549621582,
      "epoch": 0.10816657652785289,
      "grad_norm": 1.015625,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 6.5476,
      "mean_token_accuracy": 0.1357502222061157,
      "num_tokens": 225310.0,
      "step": 100
    },
    {
      "entropy": 6.757467412948609,
      "epoch": 0.11357490535424554,
      "grad_norm": 1.0234375,
      "learning_rate": 5.2e-05,
      "loss": 6.3829,
      "mean_token_accuracy": 0.13580593466758728,
      "num_tokens": 236676.0,
      "step": 105
    },
    {
      "entropy": 6.749653005599976,
      "epoch": 0.11898323418063818,
      "grad_norm": 1.1640625,
      "learning_rate": 5.45e-05,
      "loss": 6.4842,
      "mean_token_accuracy": 0.13619382083415985,
      "num_tokens": 247453.0,
      "step": 110
    },
    {
      "entropy": 6.7436058044433596,
      "epoch": 0.12439156300703083,
      "grad_norm": 1.2265625,
      "learning_rate": 5.7e-05,
      "loss": 6.4908,
      "mean_token_accuracy": 0.1296382687985897,
      "num_tokens": 258267.0,
      "step": 115
    },
    {
      "entropy": 6.699334526062012,
      "epoch": 0.12979989183342347,
      "grad_norm": 1.125,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 6.5306,
      "mean_token_accuracy": 0.1270687237381935,
      "num_tokens": 269839.0,
      "step": 120
    },
    {
      "entropy": 6.761990737915039,
      "epoch": 0.1352082206598161,
      "grad_norm": 1.0703125,
      "learning_rate": 6.2e-05,
      "loss": 6.5192,
      "mean_token_accuracy": 0.13135910406708717,
      "num_tokens": 281748.0,
      "step": 125
    },
    {
      "entropy": 6.810469007492065,
      "epoch": 0.14061654948620878,
      "grad_norm": 1.0703125,
      "learning_rate": 6.450000000000001e-05,
      "loss": 6.5302,
      "mean_token_accuracy": 0.12570069655776023,
      "num_tokens": 292592.0,
      "step": 130
    },
    {
      "entropy": 6.8416718482971195,
      "epoch": 0.1460248783126014,
      "grad_norm": 1.1953125,
      "learning_rate": 6.7e-05,
      "loss": 6.4401,
      "mean_token_accuracy": 0.13145631179213524,
      "num_tokens": 303122.0,
      "step": 135
    },
    {
      "entropy": 6.692539358139038,
      "epoch": 0.15143320713899405,
      "grad_norm": 1.1875,
      "learning_rate": 6.950000000000001e-05,
      "loss": 6.3644,
      "mean_token_accuracy": 0.13424549549818038,
      "num_tokens": 314728.0,
      "step": 140
    },
    {
      "entropy": 6.719494438171386,
      "epoch": 0.1568415359653867,
      "grad_norm": 1.1328125,
      "learning_rate": 7.2e-05,
      "loss": 6.5632,
      "mean_token_accuracy": 0.13301948606967925,
      "num_tokens": 325292.0,
      "step": 145
    },
    {
      "entropy": 6.830582857131958,
      "epoch": 0.16224986479177933,
      "grad_norm": 1.03125,
      "learning_rate": 7.45e-05,
      "loss": 6.5829,
      "mean_token_accuracy": 0.1278594434261322,
      "num_tokens": 336140.0,
      "step": 150
    },
    {
      "entropy": 6.741078948974609,
      "epoch": 0.167658193618172,
      "grad_norm": 1.1484375,
      "learning_rate": 7.7e-05,
      "loss": 6.4212,
      "mean_token_accuracy": 0.13609526231884955,
      "num_tokens": 346885.0,
      "step": 155
    },
    {
      "entropy": 6.667781114578247,
      "epoch": 0.17306652244456464,
      "grad_norm": 1.203125,
      "learning_rate": 7.950000000000001e-05,
      "loss": 6.4422,
      "mean_token_accuracy": 0.13269152715802193,
      "num_tokens": 356630.0,
      "step": 160
    },
    {
      "entropy": 6.826272296905517,
      "epoch": 0.17847485127095727,
      "grad_norm": 1.1015625,
      "learning_rate": 8.2e-05,
      "loss": 6.4434,
      "mean_token_accuracy": 0.13844147995114325,
      "num_tokens": 366693.0,
      "step": 165
    },
    {
      "entropy": 6.623198461532593,
      "epoch": 0.1838831800973499,
      "grad_norm": 1.0703125,
      "learning_rate": 8.450000000000001e-05,
      "loss": 6.3919,
      "mean_token_accuracy": 0.12981215193867685,
      "num_tokens": 377779.0,
      "step": 170
    },
    {
      "entropy": 6.907767629623413,
      "epoch": 0.18929150892374255,
      "grad_norm": 1.015625,
      "learning_rate": 8.7e-05,
      "loss": 6.5139,
      "mean_token_accuracy": 0.1325360544025898,
      "num_tokens": 388482.0,
      "step": 175
    },
    {
      "entropy": 6.72784013748169,
      "epoch": 0.19469983775013522,
      "grad_norm": 1.234375,
      "learning_rate": 8.95e-05,
      "loss": 6.577,
      "mean_token_accuracy": 0.13030516654253005,
      "num_tokens": 400110.0,
      "step": 180
    },
    {
      "entropy": 6.742452812194824,
      "epoch": 0.20010816657652786,
      "grad_norm": 1.109375,
      "learning_rate": 9.2e-05,
      "loss": 6.3125,
      "mean_token_accuracy": 0.13682465478777886,
      "num_tokens": 412535.0,
      "step": 185
    },
    {
      "entropy": 6.676760053634643,
      "epoch": 0.2055164954029205,
      "grad_norm": 1.1796875,
      "learning_rate": 9.45e-05,
      "loss": 6.4502,
      "mean_token_accuracy": 0.13751535415649413,
      "num_tokens": 424309.0,
      "step": 190
    },
    {
      "entropy": 6.74865140914917,
      "epoch": 0.21092482422931313,
      "grad_norm": 1.25,
      "learning_rate": 9.7e-05,
      "loss": 6.4565,
      "mean_token_accuracy": 0.13581018298864364,
      "num_tokens": 435592.0,
      "step": 195
    },
    {
      "entropy": 6.591330480575562,
      "epoch": 0.21633315305570577,
      "grad_norm": 1.1328125,
      "learning_rate": 9.95e-05,
      "loss": 6.3516,
      "mean_token_accuracy": 0.1346059188246727,
      "num_tokens": 446767.0,
      "step": 200
    },
    {
      "entropy": 6.724362230300903,
      "epoch": 0.22174148188209844,
      "grad_norm": 1.1875,
      "learning_rate": 0.000102,
      "loss": 6.3719,
      "mean_token_accuracy": 0.1348101980984211,
      "num_tokens": 457959.0,
      "step": 205
    },
    {
      "entropy": 6.5875874042510985,
      "epoch": 0.22714981070849108,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00010449999999999999,
      "loss": 6.3364,
      "mean_token_accuracy": 0.14194149523973465,
      "num_tokens": 468107.0,
      "step": 210
    },
    {
      "entropy": 6.896349143981934,
      "epoch": 0.23255813953488372,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000107,
      "loss": 6.5223,
      "mean_token_accuracy": 0.13508757278323175,
      "num_tokens": 478954.0,
      "step": 215
    },
    {
      "entropy": 6.597625112533569,
      "epoch": 0.23796646836127636,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001095,
      "loss": 6.436,
      "mean_token_accuracy": 0.12944695577025414,
      "num_tokens": 490297.0,
      "step": 220
    },
    {
      "entropy": 6.734339427947998,
      "epoch": 0.24337479718766902,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000112,
      "loss": 6.4306,
      "mean_token_accuracy": 0.1379121758043766,
      "num_tokens": 501597.0,
      "step": 225
    },
    {
      "entropy": 6.699416732788086,
      "epoch": 0.24878312601406166,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001145,
      "loss": 6.4431,
      "mean_token_accuracy": 0.13053352981805802,
      "num_tokens": 514182.0,
      "step": 230
    },
    {
      "entropy": 6.637861156463623,
      "epoch": 0.2541914548404543,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00011700000000000001,
      "loss": 6.3907,
      "mean_token_accuracy": 0.1391088232398033,
      "num_tokens": 525972.0,
      "step": 235
    },
    {
      "entropy": 6.706511783599853,
      "epoch": 0.25959978366684694,
      "grad_norm": 1.140625,
      "learning_rate": 0.00011949999999999999,
      "loss": 6.3833,
      "mean_token_accuracy": 0.1393005795776844,
      "num_tokens": 537445.0,
      "step": 240
    },
    {
      "entropy": 6.676162910461426,
      "epoch": 0.2650081124932396,
      "grad_norm": 1.078125,
      "learning_rate": 0.000122,
      "loss": 6.4325,
      "mean_token_accuracy": 0.13711966052651406,
      "num_tokens": 548599.0,
      "step": 245
    },
    {
      "entropy": 6.67521448135376,
      "epoch": 0.2704164413196322,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001245,
      "loss": 6.3094,
      "mean_token_accuracy": 0.1379949890077114,
      "num_tokens": 559222.0,
      "step": 250
    },
    {
      "entropy": 6.713015365600586,
      "epoch": 0.27582477014602486,
      "grad_norm": 1.3125,
      "learning_rate": 0.000127,
      "loss": 6.4632,
      "mean_token_accuracy": 0.12452752217650413,
      "num_tokens": 568949.0,
      "step": 255
    },
    {
      "entropy": 6.779383373260498,
      "epoch": 0.28123309897241755,
      "grad_norm": 1.0625,
      "learning_rate": 0.0001295,
      "loss": 6.4922,
      "mean_token_accuracy": 0.13144948631525039,
      "num_tokens": 580029.0,
      "step": 260
    },
    {
      "entropy": 6.726202201843262,
      "epoch": 0.2866414277988102,
      "grad_norm": 1.125,
      "learning_rate": 0.000132,
      "loss": 6.4077,
      "mean_token_accuracy": 0.1364477679133415,
      "num_tokens": 591633.0,
      "step": 265
    },
    {
      "entropy": 6.753827857971191,
      "epoch": 0.2920497566252028,
      "grad_norm": 1.25,
      "learning_rate": 0.00013450000000000002,
      "loss": 6.4197,
      "mean_token_accuracy": 0.1380621761083603,
      "num_tokens": 602511.0,
      "step": 270
    },
    {
      "entropy": 6.733678150177002,
      "epoch": 0.29745808545159547,
      "grad_norm": 1.265625,
      "learning_rate": 0.00013700000000000002,
      "loss": 6.4977,
      "mean_token_accuracy": 0.1339564248919487,
      "num_tokens": 613318.0,
      "step": 275
    },
    {
      "entropy": 6.607444524765015,
      "epoch": 0.3028664142779881,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001395,
      "loss": 6.3873,
      "mean_token_accuracy": 0.13400006145238877,
      "num_tokens": 624043.0,
      "step": 280
    },
    {
      "entropy": 6.67069182395935,
      "epoch": 0.30827474310438074,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014199999999999998,
      "loss": 6.3469,
      "mean_token_accuracy": 0.14132973700761794,
      "num_tokens": 634713.0,
      "step": 285
    },
    {
      "entropy": 6.528910064697266,
      "epoch": 0.3136830719307734,
      "grad_norm": 1.125,
      "learning_rate": 0.0001445,
      "loss": 6.2954,
      "mean_token_accuracy": 0.13748166486620902,
      "num_tokens": 645491.0,
      "step": 290
    },
    {
      "entropy": 6.753203821182251,
      "epoch": 0.319091400757166,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000147,
      "loss": 6.4336,
      "mean_token_accuracy": 0.13378573432564736,
      "num_tokens": 656832.0,
      "step": 295
    },
    {
      "entropy": 6.682645654678344,
      "epoch": 0.32449972958355866,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001495,
      "loss": 6.3984,
      "mean_token_accuracy": 0.13327668309211732,
      "num_tokens": 668499.0,
      "step": 300
    },
    {
      "entropy": 6.722540235519409,
      "epoch": 0.3299080584099513,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000152,
      "loss": 6.4665,
      "mean_token_accuracy": 0.1349737487733364,
      "num_tokens": 679962.0,
      "step": 305
    },
    {
      "entropy": 6.740982961654663,
      "epoch": 0.335316387236344,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00015450000000000001,
      "loss": 6.4479,
      "mean_token_accuracy": 0.13650912493467332,
      "num_tokens": 692170.0,
      "step": 310
    },
    {
      "entropy": 6.57421088218689,
      "epoch": 0.34072471606273663,
      "grad_norm": 1.1875,
      "learning_rate": 0.000157,
      "loss": 6.3711,
      "mean_token_accuracy": 0.13764476627111435,
      "num_tokens": 702886.0,
      "step": 315
    },
    {
      "entropy": 6.688414525985718,
      "epoch": 0.34613304488912927,
      "grad_norm": 1.25,
      "learning_rate": 0.0001595,
      "loss": 6.4313,
      "mean_token_accuracy": 0.13261876478791237,
      "num_tokens": 714257.0,
      "step": 320
    },
    {
      "entropy": 6.638928413391113,
      "epoch": 0.3515413737155219,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000162,
      "loss": 6.4197,
      "mean_token_accuracy": 0.13729432821273804,
      "num_tokens": 725751.0,
      "step": 325
    },
    {
      "entropy": 6.782710266113281,
      "epoch": 0.35694970254191455,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.4891,
      "mean_token_accuracy": 0.1316055290400982,
      "num_tokens": 737147.0,
      "step": 330
    },
    {
      "entropy": 6.616734218597412,
      "epoch": 0.3623580313683072,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.3442,
      "mean_token_accuracy": 0.14010964184999466,
      "num_tokens": 748439.0,
      "step": 335
    },
    {
      "entropy": 6.645461893081665,
      "epoch": 0.3677663601946998,
      "grad_norm": 1.171875,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.3558,
      "mean_token_accuracy": 0.1416398137807846,
      "num_tokens": 760380.0,
      "step": 340
    },
    {
      "entropy": 6.582087898254395,
      "epoch": 0.37317468902109246,
      "grad_norm": 1.203125,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.3419,
      "mean_token_accuracy": 0.13614206165075302,
      "num_tokens": 770827.0,
      "step": 345
    },
    {
      "entropy": 6.588772773742676,
      "epoch": 0.3785830178474851,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.371,
      "mean_token_accuracy": 0.130119176954031,
      "num_tokens": 781887.0,
      "step": 350
    },
    {
      "entropy": 6.6461952209472654,
      "epoch": 0.3839913466738778,
      "grad_norm": 1.296875,
      "learning_rate": 0.000177,
      "loss": 6.3707,
      "mean_token_accuracy": 0.12962670475244523,
      "num_tokens": 792676.0,
      "step": 355
    },
    {
      "entropy": 6.811271715164184,
      "epoch": 0.38939967550027044,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001795,
      "loss": 6.5791,
      "mean_token_accuracy": 0.13036322295665742,
      "num_tokens": 804148.0,
      "step": 360
    },
    {
      "entropy": 6.647534465789795,
      "epoch": 0.3948080043266631,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000182,
      "loss": 6.3464,
      "mean_token_accuracy": 0.13733255341649056,
      "num_tokens": 815835.0,
      "step": 365
    },
    {
      "entropy": 6.539557218551636,
      "epoch": 0.4002163331530557,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001845,
      "loss": 6.2902,
      "mean_token_accuracy": 0.14055102020502092,
      "num_tokens": 827975.0,
      "step": 370
    },
    {
      "entropy": 6.627246284484864,
      "epoch": 0.40562466197944835,
      "grad_norm": 1.25,
      "learning_rate": 0.000187,
      "loss": 6.3392,
      "mean_token_accuracy": 0.13682809248566627,
      "num_tokens": 838982.0,
      "step": 375
    },
    {
      "entropy": 6.624072790145874,
      "epoch": 0.411032990805841,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001895,
      "loss": 6.4402,
      "mean_token_accuracy": 0.13047717586159707,
      "num_tokens": 851377.0,
      "step": 380
    },
    {
      "entropy": 6.7495063781738285,
      "epoch": 0.41644131963223363,
      "grad_norm": 1.296875,
      "learning_rate": 0.000192,
      "loss": 6.4657,
      "mean_token_accuracy": 0.13667654022574424,
      "num_tokens": 862593.0,
      "step": 385
    },
    {
      "entropy": 6.551447343826294,
      "epoch": 0.42184964845862627,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001945,
      "loss": 6.3196,
      "mean_token_accuracy": 0.1354401208460331,
      "num_tokens": 873906.0,
      "step": 390
    },
    {
      "entropy": 6.712440156936646,
      "epoch": 0.4272579772850189,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.4657,
      "mean_token_accuracy": 0.1353951647877693,
      "num_tokens": 884477.0,
      "step": 395
    },
    {
      "entropy": 6.659290027618408,
      "epoch": 0.43266630611141155,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.3206,
      "mean_token_accuracy": 0.13656646832823754,
      "num_tokens": 896151.0,
      "step": 400
    },
    {
      "entropy": 6.6360023021698,
      "epoch": 0.43807463493780424,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000202,
      "loss": 6.4024,
      "mean_token_accuracy": 0.13927194178104402,
      "num_tokens": 907580.0,
      "step": 405
    },
    {
      "entropy": 6.59203953742981,
      "epoch": 0.4434829637641969,
      "grad_norm": 1.25,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.3489,
      "mean_token_accuracy": 0.13741599917411804,
      "num_tokens": 918141.0,
      "step": 410
    },
    {
      "entropy": 6.65588059425354,
      "epoch": 0.4488912925905895,
      "grad_norm": 1.203125,
      "learning_rate": 0.000207,
      "loss": 6.3899,
      "mean_token_accuracy": 0.14542332291603088,
      "num_tokens": 929868.0,
      "step": 415
    },
    {
      "entropy": 6.4867551803588865,
      "epoch": 0.45429962141698216,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002095,
      "loss": 6.2806,
      "mean_token_accuracy": 0.13965032547712325,
      "num_tokens": 940582.0,
      "step": 420
    },
    {
      "entropy": 6.6291666507720945,
      "epoch": 0.4597079502433748,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000212,
      "loss": 6.3463,
      "mean_token_accuracy": 0.13679536208510398,
      "num_tokens": 952233.0,
      "step": 425
    },
    {
      "entropy": 6.669621229171753,
      "epoch": 0.46511627906976744,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002145,
      "loss": 6.4395,
      "mean_token_accuracy": 0.13214596956968308,
      "num_tokens": 964252.0,
      "step": 430
    },
    {
      "entropy": 6.623089265823364,
      "epoch": 0.4705246078961601,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.3381,
      "mean_token_accuracy": 0.1373963512480259,
      "num_tokens": 974692.0,
      "step": 435
    },
    {
      "entropy": 6.575704860687256,
      "epoch": 0.4759329367225527,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002195,
      "loss": 6.354,
      "mean_token_accuracy": 0.13740625306963922,
      "num_tokens": 986019.0,
      "step": 440
    },
    {
      "entropy": 6.479989528656006,
      "epoch": 0.48134126554894535,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000222,
      "loss": 6.2588,
      "mean_token_accuracy": 0.1383764624595642,
      "num_tokens": 996701.0,
      "step": 445
    },
    {
      "entropy": 6.638360166549683,
      "epoch": 0.48674959437533805,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002245,
      "loss": 6.3562,
      "mean_token_accuracy": 0.1420356035232544,
      "num_tokens": 1008596.0,
      "step": 450
    },
    {
      "entropy": 6.603897285461426,
      "epoch": 0.4921579232017307,
      "grad_norm": 1.15625,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.3896,
      "mean_token_accuracy": 0.13321260511875152,
      "num_tokens": 1019263.0,
      "step": 455
    },
    {
      "entropy": 6.627950143814087,
      "epoch": 0.4975662520281233,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.3281,
      "mean_token_accuracy": 0.13852308169007302,
      "num_tokens": 1029460.0,
      "step": 460
    },
    {
      "entropy": 6.532049703598022,
      "epoch": 0.502974580854516,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.3947,
      "mean_token_accuracy": 0.13395455181598664,
      "num_tokens": 1041747.0,
      "step": 465
    },
    {
      "entropy": 6.580554580688476,
      "epoch": 0.5083829096809086,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.2302,
      "mean_token_accuracy": 0.13726723864674567,
      "num_tokens": 1051645.0,
      "step": 470
    },
    {
      "entropy": 6.394359159469604,
      "epoch": 0.5137912385073012,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000237,
      "loss": 6.1471,
      "mean_token_accuracy": 0.14182863682508468,
      "num_tokens": 1062358.0,
      "step": 475
    },
    {
      "entropy": 6.6230544090271,
      "epoch": 0.5191995673336939,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002395,
      "loss": 6.3291,
      "mean_token_accuracy": 0.13732218518853187,
      "num_tokens": 1074234.0,
      "step": 480
    },
    {
      "entropy": 6.400064849853516,
      "epoch": 0.5246078961600865,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000242,
      "loss": 6.257,
      "mean_token_accuracy": 0.1385449767112732,
      "num_tokens": 1086608.0,
      "step": 485
    },
    {
      "entropy": 6.640989637374878,
      "epoch": 0.5300162249864792,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002445,
      "loss": 6.3092,
      "mean_token_accuracy": 0.1422146663069725,
      "num_tokens": 1096878.0,
      "step": 490
    },
    {
      "entropy": 6.501981782913208,
      "epoch": 0.5354245538128718,
      "grad_norm": 1.125,
      "learning_rate": 0.000247,
      "loss": 6.3667,
      "mean_token_accuracy": 0.14159199818968773,
      "num_tokens": 1108089.0,
      "step": 495
    },
    {
      "entropy": 6.608476877212524,
      "epoch": 0.5408328826392644,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002495,
      "loss": 6.3983,
      "mean_token_accuracy": 0.13436152264475823,
      "num_tokens": 1120910.0,
      "step": 500
    },
    {
      "entropy": 6.558342504501343,
      "epoch": 0.5462412114656571,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000252,
      "loss": 6.2966,
      "mean_token_accuracy": 0.13886259347200394,
      "num_tokens": 1131685.0,
      "step": 505
    },
    {
      "entropy": 6.493104648590088,
      "epoch": 0.5516495402920497,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002545,
      "loss": 6.337,
      "mean_token_accuracy": 0.13808612152934074,
      "num_tokens": 1142273.0,
      "step": 510
    },
    {
      "entropy": 6.593916702270508,
      "epoch": 0.5570578691184424,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000257,
      "loss": 6.212,
      "mean_token_accuracy": 0.1403546467423439,
      "num_tokens": 1152818.0,
      "step": 515
    },
    {
      "entropy": 6.4184082508087155,
      "epoch": 0.5624661979448351,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002595,
      "loss": 6.2539,
      "mean_token_accuracy": 0.14156585782766343,
      "num_tokens": 1163398.0,
      "step": 520
    },
    {
      "entropy": 6.639800310134888,
      "epoch": 0.5678745267712277,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000262,
      "loss": 6.3113,
      "mean_token_accuracy": 0.14177494123578072,
      "num_tokens": 1174493.0,
      "step": 525
    },
    {
      "entropy": 6.449878311157226,
      "epoch": 0.5732828555976204,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.2964,
      "mean_token_accuracy": 0.14197195023298265,
      "num_tokens": 1185534.0,
      "step": 530
    },
    {
      "entropy": 6.524397659301758,
      "epoch": 0.578691184424013,
      "grad_norm": 1.46875,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.2393,
      "mean_token_accuracy": 0.14146455824375154,
      "num_tokens": 1196488.0,
      "step": 535
    },
    {
      "entropy": 6.451419448852539,
      "epoch": 0.5840995132504057,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.3155,
      "mean_token_accuracy": 0.139084542542696,
      "num_tokens": 1208116.0,
      "step": 540
    },
    {
      "entropy": 6.48581862449646,
      "epoch": 0.5895078420767983,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.3035,
      "mean_token_accuracy": 0.14370152205228806,
      "num_tokens": 1219571.0,
      "step": 545
    },
    {
      "entropy": 6.518029594421387,
      "epoch": 0.5949161709031909,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002745,
      "loss": 6.2869,
      "mean_token_accuracy": 0.13986791446805,
      "num_tokens": 1231049.0,
      "step": 550
    },
    {
      "entropy": 6.475363636016846,
      "epoch": 0.6003244997295836,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000277,
      "loss": 6.3097,
      "mean_token_accuracy": 0.1364640511572361,
      "num_tokens": 1242507.0,
      "step": 555
    },
    {
      "entropy": 6.44856014251709,
      "epoch": 0.6057328285559762,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002795,
      "loss": 6.2934,
      "mean_token_accuracy": 0.13697041645646096,
      "num_tokens": 1253607.0,
      "step": 560
    },
    {
      "entropy": 6.528074312210083,
      "epoch": 0.6111411573823688,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.3358,
      "mean_token_accuracy": 0.13758110031485557,
      "num_tokens": 1266604.0,
      "step": 565
    },
    {
      "entropy": 6.573901605606079,
      "epoch": 0.6165494862087615,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002845,
      "loss": 6.3815,
      "mean_token_accuracy": 0.13712944313883782,
      "num_tokens": 1276930.0,
      "step": 570
    },
    {
      "entropy": 6.598294305801391,
      "epoch": 0.6219578150351541,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000287,
      "loss": 6.3853,
      "mean_token_accuracy": 0.13915850892663,
      "num_tokens": 1288326.0,
      "step": 575
    },
    {
      "entropy": 6.450491666793823,
      "epoch": 0.6273661438615468,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002895,
      "loss": 6.3287,
      "mean_token_accuracy": 0.13975025117397308,
      "num_tokens": 1298950.0,
      "step": 580
    },
    {
      "entropy": 6.524539470672607,
      "epoch": 0.6327744726879394,
      "grad_norm": 1.1875,
      "learning_rate": 0.000292,
      "loss": 6.3231,
      "mean_token_accuracy": 0.13521830812096597,
      "num_tokens": 1311254.0,
      "step": 585
    },
    {
      "entropy": 6.515868997573852,
      "epoch": 0.638182801514332,
      "grad_norm": 1.25,
      "learning_rate": 0.0002945,
      "loss": 6.2362,
      "mean_token_accuracy": 0.13961164206266402,
      "num_tokens": 1322723.0,
      "step": 590
    },
    {
      "entropy": 6.4743040084838865,
      "epoch": 0.6435911303407247,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000297,
      "loss": 6.2488,
      "mean_token_accuracy": 0.14040838554501534,
      "num_tokens": 1333998.0,
      "step": 595
    },
    {
      "entropy": 6.309730291366577,
      "epoch": 0.6489994591671173,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002995,
      "loss": 6.1926,
      "mean_token_accuracy": 0.14279944226145744,
      "num_tokens": 1345113.0,
      "step": 600
    },
    {
      "entropy": 6.439900350570679,
      "epoch": 0.65440778799351,
      "grad_norm": 1.171875,
      "learning_rate": 0.000302,
      "loss": 6.2886,
      "mean_token_accuracy": 0.14252566993236543,
      "num_tokens": 1356850.0,
      "step": 605
    },
    {
      "entropy": 6.550169944763184,
      "epoch": 0.6598161168199026,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003045,
      "loss": 6.3744,
      "mean_token_accuracy": 0.1397650845348835,
      "num_tokens": 1369671.0,
      "step": 610
    },
    {
      "entropy": 6.427625703811645,
      "epoch": 0.6652244456462953,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000307,
      "loss": 6.2835,
      "mean_token_accuracy": 0.13882820978760718,
      "num_tokens": 1380298.0,
      "step": 615
    },
    {
      "entropy": 6.527406787872314,
      "epoch": 0.670632774472688,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003095,
      "loss": 6.2763,
      "mean_token_accuracy": 0.13872483968734742,
      "num_tokens": 1391617.0,
      "step": 620
    },
    {
      "entropy": 6.535118579864502,
      "epoch": 0.6760411032990806,
      "grad_norm": 1.375,
      "learning_rate": 0.000312,
      "loss": 6.3461,
      "mean_token_accuracy": 0.14173057228326796,
      "num_tokens": 1403509.0,
      "step": 625
    },
    {
      "entropy": 6.511589908599854,
      "epoch": 0.6814494321254733,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003145,
      "loss": 6.4039,
      "mean_token_accuracy": 0.13432303741574286,
      "num_tokens": 1415125.0,
      "step": 630
    },
    {
      "entropy": 6.580691337585449,
      "epoch": 0.6868577609518659,
      "grad_norm": 1.28125,
      "learning_rate": 0.000317,
      "loss": 6.402,
      "mean_token_accuracy": 0.13456533700227738,
      "num_tokens": 1426785.0,
      "step": 635
    },
    {
      "entropy": 6.480119514465332,
      "epoch": 0.6922660897782585,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003195,
      "loss": 6.3605,
      "mean_token_accuracy": 0.12881308048963547,
      "num_tokens": 1439013.0,
      "step": 640
    },
    {
      "entropy": 6.417783641815186,
      "epoch": 0.6976744186046512,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000322,
      "loss": 6.2202,
      "mean_token_accuracy": 0.14642192423343658,
      "num_tokens": 1449818.0,
      "step": 645
    },
    {
      "entropy": 6.413159942626953,
      "epoch": 0.7030827474310438,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.1358,
      "mean_token_accuracy": 0.1415444567799568,
      "num_tokens": 1460681.0,
      "step": 650
    },
    {
      "entropy": 6.338744592666626,
      "epoch": 0.7084910762574365,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.2476,
      "mean_token_accuracy": 0.13981930315494537,
      "num_tokens": 1471840.0,
      "step": 655
    },
    {
      "entropy": 6.466752004623413,
      "epoch": 0.7138994050838291,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.2436,
      "mean_token_accuracy": 0.14259599149227142,
      "num_tokens": 1482326.0,
      "step": 660
    },
    {
      "entropy": 6.328251123428345,
      "epoch": 0.7193077339102217,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.1715,
      "mean_token_accuracy": 0.14127768352627754,
      "num_tokens": 1493277.0,
      "step": 665
    },
    {
      "entropy": 6.4282067775726315,
      "epoch": 0.7247160627366144,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.2511,
      "mean_token_accuracy": 0.14475265368819237,
      "num_tokens": 1504563.0,
      "step": 670
    },
    {
      "entropy": 6.372472429275513,
      "epoch": 0.730124391563007,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000337,
      "loss": 6.2369,
      "mean_token_accuracy": 0.1476322367787361,
      "num_tokens": 1516311.0,
      "step": 675
    },
    {
      "entropy": 6.354723262786865,
      "epoch": 0.7355327203893997,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003395,
      "loss": 6.1563,
      "mean_token_accuracy": 0.14649124294519425,
      "num_tokens": 1527391.0,
      "step": 680
    },
    {
      "entropy": 6.439003276824951,
      "epoch": 0.7409410492157923,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000342,
      "loss": 6.2846,
      "mean_token_accuracy": 0.14316972196102143,
      "num_tokens": 1537755.0,
      "step": 685
    },
    {
      "entropy": 6.466528415679932,
      "epoch": 0.7463493780421849,
      "grad_norm": 1.125,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.3203,
      "mean_token_accuracy": 0.1379350833594799,
      "num_tokens": 1550525.0,
      "step": 690
    },
    {
      "entropy": 6.425952291488647,
      "epoch": 0.7517577068685776,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000347,
      "loss": 6.2344,
      "mean_token_accuracy": 0.14136113673448564,
      "num_tokens": 1561995.0,
      "step": 695
    },
    {
      "entropy": 6.384309434890747,
      "epoch": 0.7571660356949702,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003495,
      "loss": 6.2363,
      "mean_token_accuracy": 0.13751797899603843,
      "num_tokens": 1573137.0,
      "step": 700
    },
    {
      "entropy": 6.332474660873413,
      "epoch": 0.7625743645213628,
      "grad_norm": 1.234375,
      "learning_rate": 0.000352,
      "loss": 6.2516,
      "mean_token_accuracy": 0.1449413001537323,
      "num_tokens": 1584938.0,
      "step": 705
    },
    {
      "entropy": 6.549186563491821,
      "epoch": 0.7679826933477556,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003545,
      "loss": 6.308,
      "mean_token_accuracy": 0.1355679027736187,
      "num_tokens": 1596306.0,
      "step": 710
    },
    {
      "entropy": 6.34313998222351,
      "epoch": 0.7733910221741482,
      "grad_norm": 1.15625,
      "learning_rate": 0.000357,
      "loss": 6.2934,
      "mean_token_accuracy": 0.14217483699321748,
      "num_tokens": 1608112.0,
      "step": 715
    },
    {
      "entropy": 6.5022155284881595,
      "epoch": 0.7787993510005409,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003595,
      "loss": 6.2833,
      "mean_token_accuracy": 0.14302016869187356,
      "num_tokens": 1618635.0,
      "step": 720
    },
    {
      "entropy": 6.277605819702148,
      "epoch": 0.7842076798269335,
      "grad_norm": 1.171875,
      "learning_rate": 0.000362,
      "loss": 6.1878,
      "mean_token_accuracy": 0.14160637855529784,
      "num_tokens": 1629476.0,
      "step": 725
    },
    {
      "entropy": 6.544418144226074,
      "epoch": 0.7896160086533262,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003645,
      "loss": 6.2851,
      "mean_token_accuracy": 0.14023412466049195,
      "num_tokens": 1641305.0,
      "step": 730
    },
    {
      "entropy": 6.323581790924072,
      "epoch": 0.7950243374797188,
      "grad_norm": 1.046875,
      "learning_rate": 0.000367,
      "loss": 6.228,
      "mean_token_accuracy": 0.1453120455145836,
      "num_tokens": 1653424.0,
      "step": 735
    },
    {
      "entropy": 6.338683032989502,
      "epoch": 0.8004326663061114,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003695,
      "loss": 6.2738,
      "mean_token_accuracy": 0.13886456340551376,
      "num_tokens": 1664222.0,
      "step": 740
    },
    {
      "entropy": 6.45950026512146,
      "epoch": 0.8058409951325041,
      "grad_norm": 1.21875,
      "learning_rate": 0.000372,
      "loss": 6.3182,
      "mean_token_accuracy": 0.13927288502454757,
      "num_tokens": 1676829.0,
      "step": 745
    },
    {
      "entropy": 6.389509057998657,
      "epoch": 0.8112493239588967,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003745,
      "loss": 6.2171,
      "mean_token_accuracy": 0.13743837997317315,
      "num_tokens": 1688534.0,
      "step": 750
    },
    {
      "entropy": 6.2615196228027346,
      "epoch": 0.8166576527852893,
      "grad_norm": 1.171875,
      "learning_rate": 0.000377,
      "loss": 6.1995,
      "mean_token_accuracy": 0.14325060099363326,
      "num_tokens": 1699051.0,
      "step": 755
    },
    {
      "entropy": 6.446281814575196,
      "epoch": 0.822065981611682,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003795,
      "loss": 6.2276,
      "mean_token_accuracy": 0.13753046318888665,
      "num_tokens": 1709849.0,
      "step": 760
    },
    {
      "entropy": 6.236609411239624,
      "epoch": 0.8274743104380746,
      "grad_norm": 1.109375,
      "learning_rate": 0.000382,
      "loss": 6.0689,
      "mean_token_accuracy": 0.15064243003726005,
      "num_tokens": 1721956.0,
      "step": 765
    },
    {
      "entropy": 6.327112913131714,
      "epoch": 0.8328826392644673,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003845,
      "loss": 6.0921,
      "mean_token_accuracy": 0.15326208472251893,
      "num_tokens": 1733302.0,
      "step": 770
    },
    {
      "entropy": 6.198913812637329,
      "epoch": 0.8382909680908599,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.041,
      "mean_token_accuracy": 0.15577499121427535,
      "num_tokens": 1744835.0,
      "step": 775
    },
    {
      "entropy": 6.3341676712036135,
      "epoch": 0.8436992969172525,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.1797,
      "mean_token_accuracy": 0.14762855991721152,
      "num_tokens": 1755174.0,
      "step": 780
    },
    {
      "entropy": 6.381376791000366,
      "epoch": 0.8491076257436452,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.3989,
      "mean_token_accuracy": 0.13875442296266555,
      "num_tokens": 1767572.0,
      "step": 785
    },
    {
      "entropy": 6.218066310882568,
      "epoch": 0.8545159545700378,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.1405,
      "mean_token_accuracy": 0.1449118934571743,
      "num_tokens": 1777537.0,
      "step": 790
    },
    {
      "entropy": 6.425797414779663,
      "epoch": 0.8599242833964305,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.1712,
      "mean_token_accuracy": 0.14846592620015145,
      "num_tokens": 1789731.0,
      "step": 795
    },
    {
      "entropy": 6.419421625137329,
      "epoch": 0.8653326122228231,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003995,
      "loss": 6.3029,
      "mean_token_accuracy": 0.14081404209136963,
      "num_tokens": 1800754.0,
      "step": 800
    },
    {
      "entropy": 6.272062110900879,
      "epoch": 0.8707409410492158,
      "grad_norm": 1.28125,
      "learning_rate": 0.000402,
      "loss": 6.2734,
      "mean_token_accuracy": 0.1466397300362587,
      "num_tokens": 1811496.0,
      "step": 805
    },
    {
      "entropy": 6.3589723110198975,
      "epoch": 0.8761492698756085,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004045,
      "loss": 6.184,
      "mean_token_accuracy": 0.1434461772441864,
      "num_tokens": 1822733.0,
      "step": 810
    },
    {
      "entropy": 6.342323112487793,
      "epoch": 0.8815575987020011,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.1623,
      "mean_token_accuracy": 0.14631599932909012,
      "num_tokens": 1834121.0,
      "step": 815
    },
    {
      "entropy": 6.243607139587402,
      "epoch": 0.8869659275283938,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004095,
      "loss": 6.0895,
      "mean_token_accuracy": 0.1486702710390091,
      "num_tokens": 1844674.0,
      "step": 820
    },
    {
      "entropy": 6.2104573249816895,
      "epoch": 0.8923742563547864,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000412,
      "loss": 6.0988,
      "mean_token_accuracy": 0.1491689234972,
      "num_tokens": 1856692.0,
      "step": 825
    },
    {
      "entropy": 6.453680181503296,
      "epoch": 0.897782585181179,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004145,
      "loss": 6.3707,
      "mean_token_accuracy": 0.1408935658633709,
      "num_tokens": 1867888.0,
      "step": 830
    },
    {
      "entropy": 6.105977869033813,
      "epoch": 0.9031909140075717,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000417,
      "loss": 6.043,
      "mean_token_accuracy": 0.15483584851026536,
      "num_tokens": 1877869.0,
      "step": 835
    },
    {
      "entropy": 6.322730207443238,
      "epoch": 0.9085992428339643,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004195,
      "loss": 6.1548,
      "mean_token_accuracy": 0.1412374958395958,
      "num_tokens": 1887975.0,
      "step": 840
    },
    {
      "entropy": 6.305921459197998,
      "epoch": 0.914007571660357,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000422,
      "loss": 6.1018,
      "mean_token_accuracy": 0.14518715515732766,
      "num_tokens": 1898289.0,
      "step": 845
    },
    {
      "entropy": 6.211407423019409,
      "epoch": 0.9194159004867496,
      "grad_norm": 1.25,
      "learning_rate": 0.0004245,
      "loss": 6.2213,
      "mean_token_accuracy": 0.15009732991456987,
      "num_tokens": 1910038.0,
      "step": 850
    },
    {
      "entropy": 6.209096813201905,
      "epoch": 0.9248242293131422,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000427,
      "loss": 6.0914,
      "mean_token_accuracy": 0.15314256697893142,
      "num_tokens": 1920774.0,
      "step": 855
    },
    {
      "entropy": 6.293406915664673,
      "epoch": 0.9302325581395349,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004295,
      "loss": 6.1251,
      "mean_token_accuracy": 0.15404994785785675,
      "num_tokens": 1932085.0,
      "step": 860
    },
    {
      "entropy": 6.232555866241455,
      "epoch": 0.9356408869659275,
      "grad_norm": 1.109375,
      "learning_rate": 0.000432,
      "loss": 6.1331,
      "mean_token_accuracy": 0.15166230872273445,
      "num_tokens": 1943949.0,
      "step": 865
    },
    {
      "entropy": 6.28867917060852,
      "epoch": 0.9410492157923201,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004345,
      "loss": 6.1927,
      "mean_token_accuracy": 0.14988541305065156,
      "num_tokens": 1956051.0,
      "step": 870
    },
    {
      "entropy": 6.260223913192749,
      "epoch": 0.9464575446187128,
      "grad_norm": 1.140625,
      "learning_rate": 0.000437,
      "loss": 6.1868,
      "mean_token_accuracy": 0.14389916732907296,
      "num_tokens": 1967804.0,
      "step": 875
    },
    {
      "entropy": 6.208997774124145,
      "epoch": 0.9518658734451054,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004395,
      "loss": 6.2995,
      "mean_token_accuracy": 0.13391195982694626,
      "num_tokens": 1979222.0,
      "step": 880
    },
    {
      "entropy": 6.381609487533569,
      "epoch": 0.9572742022714981,
      "grad_norm": 1.140625,
      "learning_rate": 0.000442,
      "loss": 6.2248,
      "mean_token_accuracy": 0.14088413566350938,
      "num_tokens": 1988998.0,
      "step": 885
    },
    {
      "entropy": 6.245026969909668,
      "epoch": 0.9626825310978907,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004445,
      "loss": 6.1451,
      "mean_token_accuracy": 0.14165820479393004,
      "num_tokens": 2000539.0,
      "step": 890
    },
    {
      "entropy": 6.321556282043457,
      "epoch": 0.9680908599242833,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000447,
      "loss": 6.2385,
      "mean_token_accuracy": 0.14436768293380736,
      "num_tokens": 2012321.0,
      "step": 895
    },
    {
      "entropy": 6.173920106887818,
      "epoch": 0.9734991887506761,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.1231,
      "mean_token_accuracy": 0.14439835995435715,
      "num_tokens": 2023330.0,
      "step": 900
    },
    {
      "entropy": 6.284840297698975,
      "epoch": 0.9789075175770687,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.1095,
      "mean_token_accuracy": 0.13921767249703407,
      "num_tokens": 2034655.0,
      "step": 905
    },
    {
      "entropy": 6.110538482666016,
      "epoch": 0.9843158464034614,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.1513,
      "mean_token_accuracy": 0.1441517509520054,
      "num_tokens": 2045785.0,
      "step": 910
    },
    {
      "entropy": 6.445047616958618,
      "epoch": 0.989724175229854,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.2424,
      "mean_token_accuracy": 0.14532435238361358,
      "num_tokens": 2056038.0,
      "step": 915
    },
    {
      "entropy": 6.129134559631348,
      "epoch": 0.9951325040562466,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.1023,
      "mean_token_accuracy": 0.15468567311763765,
      "num_tokens": 2066803.0,
      "step": 920
    },
    {
      "entropy": 6.241815196143256,
      "epoch": 1.0,
      "grad_norm": 1.90625,
      "learning_rate": 0.000462,
      "loss": 6.2124,
      "mean_token_accuracy": 0.14124820878108343,
      "num_tokens": 2076889.0,
      "step": 925
    },
    {
      "entropy": 6.257176446914673,
      "epoch": 1.0054083288263926,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004645,
      "loss": 5.9752,
      "mean_token_accuracy": 0.14727601259946824,
      "num_tokens": 2089019.0,
      "step": 930
    },
    {
      "entropy": 6.028801822662354,
      "epoch": 1.0108166576527853,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000467,
      "loss": 5.8447,
      "mean_token_accuracy": 0.16022417545318604,
      "num_tokens": 2099849.0,
      "step": 935
    },
    {
      "entropy": 6.177911376953125,
      "epoch": 1.016224986479178,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004695,
      "loss": 6.0094,
      "mean_token_accuracy": 0.1548767253756523,
      "num_tokens": 2110671.0,
      "step": 940
    },
    {
      "entropy": 6.136840963363648,
      "epoch": 1.0216333153055706,
      "grad_norm": 1.125,
      "learning_rate": 0.000472,
      "loss": 6.0524,
      "mean_token_accuracy": 0.1497804716229439,
      "num_tokens": 2122357.0,
      "step": 945
    },
    {
      "entropy": 6.268298530578614,
      "epoch": 1.0270416441319632,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004745,
      "loss": 5.9944,
      "mean_token_accuracy": 0.14714996218681337,
      "num_tokens": 2133878.0,
      "step": 950
    },
    {
      "entropy": 6.203853130340576,
      "epoch": 1.0324499729583558,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000477,
      "loss": 6.0033,
      "mean_token_accuracy": 0.15483788400888443,
      "num_tokens": 2145565.0,
      "step": 955
    },
    {
      "entropy": 6.0750096321105955,
      "epoch": 1.0378583017847485,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004795,
      "loss": 5.9273,
      "mean_token_accuracy": 0.14823657721281053,
      "num_tokens": 2157393.0,
      "step": 960
    },
    {
      "entropy": 6.056691598892212,
      "epoch": 1.043266630611141,
      "grad_norm": 1.03125,
      "learning_rate": 0.000482,
      "loss": 5.7809,
      "mean_token_accuracy": 0.1595247745513916,
      "num_tokens": 2169318.0,
      "step": 965
    },
    {
      "entropy": 6.050301027297974,
      "epoch": 1.0486749594375337,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004845,
      "loss": 5.8674,
      "mean_token_accuracy": 0.15364039540290833,
      "num_tokens": 2179727.0,
      "step": 970
    },
    {
      "entropy": 6.027638483047485,
      "epoch": 1.0540832882639264,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000487,
      "loss": 5.9021,
      "mean_token_accuracy": 0.1545193985104561,
      "num_tokens": 2190119.0,
      "step": 975
    },
    {
      "entropy": 6.070952320098877,
      "epoch": 1.059491617090319,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004895,
      "loss": 5.9234,
      "mean_token_accuracy": 0.15558279752731324,
      "num_tokens": 2201077.0,
      "step": 980
    },
    {
      "entropy": 6.101561832427978,
      "epoch": 1.0648999459167117,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000492,
      "loss": 5.9922,
      "mean_token_accuracy": 0.15358568280935286,
      "num_tokens": 2211572.0,
      "step": 985
    },
    {
      "entropy": 6.181616640090942,
      "epoch": 1.0703082747431043,
      "grad_norm": 1.0,
      "learning_rate": 0.0004945,
      "loss": 6.0144,
      "mean_token_accuracy": 0.1533959075808525,
      "num_tokens": 2223372.0,
      "step": 990
    },
    {
      "entropy": 5.968746328353882,
      "epoch": 1.075716603569497,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000497,
      "loss": 5.886,
      "mean_token_accuracy": 0.154812690615654,
      "num_tokens": 2234946.0,
      "step": 995
    },
    {
      "entropy": 6.149488687515259,
      "epoch": 1.0811249323958896,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004995,
      "loss": 5.969,
      "mean_token_accuracy": 0.14452582895755767,
      "num_tokens": 2246608.0,
      "step": 1000
    },
    {
      "epoch": 1.0811249323958896,
      "eval_entropy": 5.931731963157654,
      "eval_loss": 6.293067455291748,
      "eval_mean_token_accuracy": 0.14824318431995132,
      "eval_num_tokens": 2246608.0,
      "eval_runtime": 2.567,
      "eval_samples_per_second": 1369.326,
      "eval_steps_per_second": 171.409,
      "step": 1000
    },
    {
      "entropy": 6.084950351715088,
      "epoch": 1.0865332612222822,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999997383518051,
      "loss": 6.0811,
      "mean_token_accuracy": 0.14338656812906264,
      "num_tokens": 2258473.0,
      "step": 1005
    },
    {
      "entropy": 6.1302765846252445,
      "epoch": 1.0919415900486749,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999986754070562,
      "loss": 5.9806,
      "mean_token_accuracy": 0.15734953135252,
      "num_tokens": 2269235.0,
      "step": 1010
    },
    {
      "entropy": 6.098620748519897,
      "epoch": 1.0973499188750675,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999967948166009,
      "loss": 5.9067,
      "mean_token_accuracy": 0.15507067888975143,
      "num_tokens": 2280369.0,
      "step": 1015
    },
    {
      "entropy": 5.965772438049316,
      "epoch": 1.1027582477014604,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004999940965872734,
      "loss": 5.8955,
      "mean_token_accuracy": 0.16817984730005264,
      "num_tokens": 2289701.0,
      "step": 1020
    },
    {
      "entropy": 6.086247968673706,
      "epoch": 1.1081665765278528,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999905807288788,
      "loss": 5.9501,
      "mean_token_accuracy": 0.15449455082416536,
      "num_tokens": 2300820.0,
      "step": 1025
    },
    {
      "entropy": 6.2378397464752195,
      "epoch": 1.1135749053542456,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499986247254194,
      "loss": 6.0194,
      "mean_token_accuracy": 0.14141876846551896,
      "num_tokens": 2311710.0,
      "step": 1030
    },
    {
      "entropy": 6.080769062042236,
      "epoch": 1.1189832341806383,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999810961789667,
      "loss": 6.0102,
      "mean_token_accuracy": 0.15261175036430358,
      "num_tokens": 2324472.0,
      "step": 1035
    },
    {
      "entropy": 6.025978469848633,
      "epoch": 1.124391563007031,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499975127521916,
      "loss": 5.8953,
      "mean_token_accuracy": 0.15188806802034377,
      "num_tokens": 2336016.0,
      "step": 1040
    },
    {
      "entropy": 6.085336542129516,
      "epoch": 1.1297998918334236,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999683413047319,
      "loss": 5.9726,
      "mean_token_accuracy": 0.15661023557186127,
      "num_tokens": 2346433.0,
      "step": 1045
    },
    {
      "entropy": 6.114025735855103,
      "epoch": 1.1352082206598162,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999607375520759,
      "loss": 5.9346,
      "mean_token_accuracy": 0.16011944636702538,
      "num_tokens": 2357171.0,
      "step": 1050
    },
    {
      "entropy": 5.95264163017273,
      "epoch": 1.1406165494862088,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999523162915794,
      "loss": 5.9868,
      "mean_token_accuracy": 0.1551058605313301,
      "num_tokens": 2367918.0,
      "step": 1055
    },
    {
      "entropy": 6.105375480651856,
      "epoch": 1.1460248783126015,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999430775538459,
      "loss": 5.8611,
      "mean_token_accuracy": 0.1594705879688263,
      "num_tokens": 2378588.0,
      "step": 1060
    },
    {
      "entropy": 5.9998544216156,
      "epoch": 1.151433207138994,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999330213724487,
      "loss": 5.9136,
      "mean_token_accuracy": 0.15627419352531433,
      "num_tokens": 2390339.0,
      "step": 1065
    },
    {
      "entropy": 6.131874227523804,
      "epoch": 1.1568415359653867,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999221477839319,
      "loss": 6.0764,
      "mean_token_accuracy": 0.14559624195098878,
      "num_tokens": 2401855.0,
      "step": 1070
    },
    {
      "entropy": 6.033861970901489,
      "epoch": 1.1622498647917794,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999104568278104,
      "loss": 5.9031,
      "mean_token_accuracy": 0.15754418671131135,
      "num_tokens": 2411902.0,
      "step": 1075
    },
    {
      "entropy": 6.130727386474609,
      "epoch": 1.167658193618172,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499897948546569,
      "loss": 6.003,
      "mean_token_accuracy": 0.15505098104476928,
      "num_tokens": 2423275.0,
      "step": 1080
    },
    {
      "entropy": 5.969374370574951,
      "epoch": 1.1730665224445647,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998846229856629,
      "loss": 5.9628,
      "mean_token_accuracy": 0.15772853493690492,
      "num_tokens": 2434347.0,
      "step": 1085
    },
    {
      "entropy": 6.132894992828369,
      "epoch": 1.1784748512709573,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998704801935171,
      "loss": 6.0229,
      "mean_token_accuracy": 0.1508957713842392,
      "num_tokens": 2446486.0,
      "step": 1090
    },
    {
      "entropy": 5.953886795043945,
      "epoch": 1.18388318009735,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998555202215267,
      "loss": 5.886,
      "mean_token_accuracy": 0.15241267830133437,
      "num_tokens": 2458864.0,
      "step": 1095
    },
    {
      "entropy": 6.058213758468628,
      "epoch": 1.1892915089237426,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998397431240561,
      "loss": 5.8926,
      "mean_token_accuracy": 0.15482796281576156,
      "num_tokens": 2471225.0,
      "step": 1100
    },
    {
      "entropy": 5.991312265396118,
      "epoch": 1.1946998377501352,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998231489584395,
      "loss": 5.9081,
      "mean_token_accuracy": 0.15361883193254472,
      "num_tokens": 2481914.0,
      "step": 1105
    },
    {
      "entropy": 6.007622146606446,
      "epoch": 1.2001081665765279,
      "grad_norm": 1.09375,
      "learning_rate": 0.00049980573778498,
      "loss": 5.9317,
      "mean_token_accuracy": 0.1619701385498047,
      "num_tokens": 2493391.0,
      "step": 1110
    },
    {
      "entropy": 5.986660623550415,
      "epoch": 1.2055164954029205,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997875096669498,
      "loss": 5.8213,
      "mean_token_accuracy": 0.15980194956064225,
      "num_tokens": 2504576.0,
      "step": 1115
    },
    {
      "entropy": 5.949508237838745,
      "epoch": 1.2109248242293131,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997684646705901,
      "loss": 5.8822,
      "mean_token_accuracy": 0.15818627998232843,
      "num_tokens": 2516364.0,
      "step": 1120
    },
    {
      "entropy": 6.021165227890014,
      "epoch": 1.2163331530557058,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997486028651103,
      "loss": 5.949,
      "mean_token_accuracy": 0.15565033853054047,
      "num_tokens": 2526296.0,
      "step": 1125
    },
    {
      "entropy": 5.929991960525513,
      "epoch": 1.2217414818820984,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997279243226884,
      "loss": 5.8364,
      "mean_token_accuracy": 0.16595554798841478,
      "num_tokens": 2537558.0,
      "step": 1130
    },
    {
      "entropy": 6.006711053848266,
      "epoch": 1.227149810708491,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997064291184701,
      "loss": 5.9242,
      "mean_token_accuracy": 0.15124313682317733,
      "num_tokens": 2549109.0,
      "step": 1135
    },
    {
      "entropy": 6.145928907394409,
      "epoch": 1.2325581395348837,
      "grad_norm": 1.25,
      "learning_rate": 0.0004996841173305691,
      "loss": 6.0694,
      "mean_token_accuracy": 0.15124114975333214,
      "num_tokens": 2560019.0,
      "step": 1140
    },
    {
      "entropy": 5.955758571624756,
      "epoch": 1.2379664683612763,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996609890400665,
      "loss": 5.9005,
      "mean_token_accuracy": 0.15468142032623292,
      "num_tokens": 2571455.0,
      "step": 1145
    },
    {
      "entropy": 5.977803087234497,
      "epoch": 1.243374797187669,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996370443310107,
      "loss": 5.8316,
      "mean_token_accuracy": 0.1626493364572525,
      "num_tokens": 2582827.0,
      "step": 1150
    },
    {
      "entropy": 5.962737560272217,
      "epoch": 1.2487831260140616,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996122832904165,
      "loss": 5.9224,
      "mean_token_accuracy": 0.15286874324083327,
      "num_tokens": 2594980.0,
      "step": 1155
    },
    {
      "entropy": 6.072302770614624,
      "epoch": 1.2541914548404542,
      "grad_norm": 1.125,
      "learning_rate": 0.0004995867060082658,
      "loss": 5.9808,
      "mean_token_accuracy": 0.15695970505475998,
      "num_tokens": 2605697.0,
      "step": 1160
    },
    {
      "entropy": 6.013327169418335,
      "epoch": 1.2595997836668469,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995603125775065,
      "loss": 5.9401,
      "mean_token_accuracy": 0.1553099974989891,
      "num_tokens": 2617515.0,
      "step": 1165
    },
    {
      "entropy": 5.9621072769165036,
      "epoch": 1.2650081124932395,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004995331030940523,
      "loss": 5.9188,
      "mean_token_accuracy": 0.1621383920311928,
      "num_tokens": 2628293.0,
      "step": 1170
    },
    {
      "entropy": 5.955429220199585,
      "epoch": 1.2704164413196322,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995050776567826,
      "loss": 5.8808,
      "mean_token_accuracy": 0.14956496506929398,
      "num_tokens": 2639949.0,
      "step": 1175
    },
    {
      "entropy": 6.067929220199585,
      "epoch": 1.2758247701460248,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994762363675421,
      "loss": 6.0005,
      "mean_token_accuracy": 0.15124383419752122,
      "num_tokens": 2651186.0,
      "step": 1180
    },
    {
      "entropy": 6.039377021789551,
      "epoch": 1.2812330989724177,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994465793311396,
      "loss": 5.9522,
      "mean_token_accuracy": 0.14760056138038635,
      "num_tokens": 2662452.0,
      "step": 1185
    },
    {
      "entropy": 5.962746095657349,
      "epoch": 1.28664142779881,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004994161066553493,
      "loss": 5.889,
      "mean_token_accuracy": 0.15765392333269118,
      "num_tokens": 2673117.0,
      "step": 1190
    },
    {
      "entropy": 6.005440950393677,
      "epoch": 1.292049756625203,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004993848184509086,
      "loss": 5.8946,
      "mean_token_accuracy": 0.15160483941435815,
      "num_tokens": 2685634.0,
      "step": 1195
    },
    {
      "entropy": 6.03985481262207,
      "epoch": 1.2974580854515954,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004993527148315193,
      "loss": 5.7672,
      "mean_token_accuracy": 0.15762777477502823,
      "num_tokens": 2695890.0,
      "step": 1200
    },
    {
      "entropy": 5.909327554702759,
      "epoch": 1.3028664142779882,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993197959138458,
      "loss": 5.9199,
      "mean_token_accuracy": 0.16145165413618087,
      "num_tokens": 2707169.0,
      "step": 1205
    },
    {
      "entropy": 5.963200855255127,
      "epoch": 1.3082747431043806,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004992860618175156,
      "loss": 5.8768,
      "mean_token_accuracy": 0.1663581445813179,
      "num_tokens": 2718118.0,
      "step": 1210
    },
    {
      "entropy": 5.981848621368409,
      "epoch": 1.3136830719307735,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004992515126651184,
      "loss": 5.8184,
      "mean_token_accuracy": 0.1567670851945877,
      "num_tokens": 2728351.0,
      "step": 1215
    },
    {
      "entropy": 5.87762303352356,
      "epoch": 1.319091400757166,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004992161485822059,
      "loss": 5.8697,
      "mean_token_accuracy": 0.16071014404296874,
      "num_tokens": 2739084.0,
      "step": 1220
    },
    {
      "entropy": 5.946296977996826,
      "epoch": 1.3244997295835588,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004991799696972913,
      "loss": 5.8888,
      "mean_token_accuracy": 0.15838139951229097,
      "num_tokens": 2750715.0,
      "step": 1225
    },
    {
      "entropy": 5.898415899276733,
      "epoch": 1.3299080584099512,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004991429761418488,
      "loss": 5.8427,
      "mean_token_accuracy": 0.16084625273942948,
      "num_tokens": 2761483.0,
      "step": 1230
    },
    {
      "entropy": 6.078216409683227,
      "epoch": 1.335316387236344,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499105168050313,
      "loss": 5.9922,
      "mean_token_accuracy": 0.14723649471998215,
      "num_tokens": 2774053.0,
      "step": 1235
    },
    {
      "entropy": 5.825717401504517,
      "epoch": 1.3407247160627367,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004990665455600787,
      "loss": 5.751,
      "mean_token_accuracy": 0.16435801833868027,
      "num_tokens": 2784928.0,
      "step": 1240
    },
    {
      "entropy": 6.0135369300842285,
      "epoch": 1.3461330448891293,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004990271088114999,
      "loss": 5.8566,
      "mean_token_accuracy": 0.1591189458966255,
      "num_tokens": 2795839.0,
      "step": 1245
    },
    {
      "entropy": 5.885397672653198,
      "epoch": 1.351541373715522,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004989868579478901,
      "loss": 5.7649,
      "mean_token_accuracy": 0.1652156174182892,
      "num_tokens": 2807624.0,
      "step": 1250
    },
    {
      "entropy": 5.896845531463623,
      "epoch": 1.3569497025419146,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004989457931155209,
      "loss": 5.8639,
      "mean_token_accuracy": 0.15702212899923323,
      "num_tokens": 2819740.0,
      "step": 1255
    },
    {
      "entropy": 5.957349491119385,
      "epoch": 1.3623580313683072,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004989039144636219,
      "loss": 5.9088,
      "mean_token_accuracy": 0.16128307580947876,
      "num_tokens": 2830467.0,
      "step": 1260
    },
    {
      "entropy": 6.03181004524231,
      "epoch": 1.3677663601946999,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004988612221443804,
      "loss": 6.0047,
      "mean_token_accuracy": 0.15610331892967225,
      "num_tokens": 2841978.0,
      "step": 1265
    },
    {
      "entropy": 5.989248514175415,
      "epoch": 1.3731746890210925,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004988177163129403,
      "loss": 5.899,
      "mean_token_accuracy": 0.15807392299175263,
      "num_tokens": 2853760.0,
      "step": 1270
    },
    {
      "entropy": 5.99000825881958,
      "epoch": 1.3785830178474852,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004987733971274018,
      "loss": 5.9337,
      "mean_token_accuracy": 0.15205859392881393,
      "num_tokens": 2865734.0,
      "step": 1275
    },
    {
      "entropy": 6.0495459079742435,
      "epoch": 1.3839913466738778,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004987282647488211,
      "loss": 5.9673,
      "mean_token_accuracy": 0.15523016154766084,
      "num_tokens": 2877872.0,
      "step": 1280
    },
    {
      "entropy": 5.842286205291748,
      "epoch": 1.3893996755002704,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004986823193412093,
      "loss": 5.8175,
      "mean_token_accuracy": 0.15781135559082032,
      "num_tokens": 2888544.0,
      "step": 1285
    },
    {
      "entropy": 5.940902042388916,
      "epoch": 1.394808004326663,
      "grad_norm": 1.09375,
      "learning_rate": 0.000498635561071532,
      "loss": 5.8806,
      "mean_token_accuracy": 0.1593789353966713,
      "num_tokens": 2899635.0,
      "step": 1290
    },
    {
      "entropy": 6.046967172622681,
      "epoch": 1.4002163331530557,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000498587990109709,
      "loss": 5.8925,
      "mean_token_accuracy": 0.15153441727161407,
      "num_tokens": 2911514.0,
      "step": 1295
    },
    {
      "entropy": 5.935166358947754,
      "epoch": 1.4056246619794484,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004985396066286133,
      "loss": 5.9422,
      "mean_token_accuracy": 0.1631091997027397,
      "num_tokens": 2922894.0,
      "step": 1300
    },
    {
      "entropy": 5.971725797653198,
      "epoch": 1.411032990805841,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004984904108040706,
      "loss": 5.9614,
      "mean_token_accuracy": 0.15326455384492874,
      "num_tokens": 2934563.0,
      "step": 1305
    },
    {
      "entropy": 5.957499170303345,
      "epoch": 1.4164413196322336,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004984404028148585,
      "loss": 5.8413,
      "mean_token_accuracy": 0.15767469108104706,
      "num_tokens": 2946101.0,
      "step": 1310
    },
    {
      "entropy": 5.75806155204773,
      "epoch": 1.4218496484586263,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004983895828427062,
      "loss": 5.8233,
      "mean_token_accuracy": 0.15837134867906572,
      "num_tokens": 2957110.0,
      "step": 1315
    },
    {
      "entropy": 6.118568181991577,
      "epoch": 1.427257977285019,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004983379510722935,
      "loss": 5.8945,
      "mean_token_accuracy": 0.16046863794326782,
      "num_tokens": 2968110.0,
      "step": 1320
    },
    {
      "entropy": 5.905016231536865,
      "epoch": 1.4326663061114115,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004982855076912506,
      "loss": 5.8924,
      "mean_token_accuracy": 0.1579983189702034,
      "num_tokens": 2979176.0,
      "step": 1325
    },
    {
      "entropy": 5.90566782951355,
      "epoch": 1.4380746349378042,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004982322528901568,
      "loss": 5.8431,
      "mean_token_accuracy": 0.16347377896308898,
      "num_tokens": 2990613.0,
      "step": 1330
    },
    {
      "entropy": 6.004480457305908,
      "epoch": 1.4434829637641968,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00049817818686254,
      "loss": 5.8586,
      "mean_token_accuracy": 0.1688141405582428,
      "num_tokens": 3002469.0,
      "step": 1335
    },
    {
      "entropy": 5.884701061248779,
      "epoch": 1.4488912925905895,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004981233098048765,
      "loss": 5.8874,
      "mean_token_accuracy": 0.1575832858681679,
      "num_tokens": 3013458.0,
      "step": 1340
    },
    {
      "entropy": 5.908089876174927,
      "epoch": 1.454299621416982,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004980676219165894,
      "loss": 5.7819,
      "mean_token_accuracy": 0.16221198439598083,
      "num_tokens": 3023435.0,
      "step": 1345
    },
    {
      "entropy": 5.851230573654175,
      "epoch": 1.4597079502433747,
      "grad_norm": 1.015625,
      "learning_rate": 0.000498011123400049,
      "loss": 5.7953,
      "mean_token_accuracy": 0.16473986059427262,
      "num_tokens": 3035124.0,
      "step": 1350
    },
    {
      "entropy": 5.867345571517944,
      "epoch": 1.4651162790697674,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004979538144605707,
      "loss": 5.7913,
      "mean_token_accuracy": 0.16175651252269746,
      "num_tokens": 3046978.0,
      "step": 1355
    },
    {
      "entropy": 5.878073263168335,
      "epoch": 1.47052460789616,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004978956953064154,
      "loss": 5.7709,
      "mean_token_accuracy": 0.16673534065485002,
      "num_tokens": 3057915.0,
      "step": 1360
    },
    {
      "entropy": 5.918510389328003,
      "epoch": 1.4759329367225527,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004978367661487883,
      "loss": 5.9289,
      "mean_token_accuracy": 0.15895457044243813,
      "num_tokens": 3069600.0,
      "step": 1365
    },
    {
      "entropy": 5.804135656356811,
      "epoch": 1.4813412655489453,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004977770272018379,
      "loss": 5.7364,
      "mean_token_accuracy": 0.16775057166814805,
      "num_tokens": 3081057.0,
      "step": 1370
    },
    {
      "entropy": 5.953877735137939,
      "epoch": 1.486749594375338,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000497716478682656,
      "loss": 5.8342,
      "mean_token_accuracy": 0.15982251316308976,
      "num_tokens": 3092967.0,
      "step": 1375
    },
    {
      "entropy": 5.905659866333008,
      "epoch": 1.4921579232017308,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004976551208112759,
      "loss": 5.8947,
      "mean_token_accuracy": 0.15754662826657295,
      "num_tokens": 3104728.0,
      "step": 1380
    },
    {
      "entropy": 5.974631023406983,
      "epoch": 1.4975662520281232,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004975929538106721,
      "loss": 5.8642,
      "mean_token_accuracy": 0.15649076998233796,
      "num_tokens": 3115946.0,
      "step": 1385
    },
    {
      "entropy": 5.901013946533203,
      "epoch": 1.502974580854516,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004975299779067598,
      "loss": 5.8748,
      "mean_token_accuracy": 0.16528355479240417,
      "num_tokens": 3126880.0,
      "step": 1390
    },
    {
      "entropy": 5.802594184875488,
      "epoch": 1.5083829096809085,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004974661933283936,
      "loss": 5.8,
      "mean_token_accuracy": 0.16498437374830247,
      "num_tokens": 3137820.0,
      "step": 1395
    },
    {
      "entropy": 5.792008113861084,
      "epoch": 1.5137912385073014,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004974016003073667,
      "loss": 5.7453,
      "mean_token_accuracy": 0.16289967000484468,
      "num_tokens": 3149620.0,
      "step": 1400
    },
    {
      "entropy": 5.800507736206055,
      "epoch": 1.5191995673336938,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004973361990784104,
      "loss": 5.737,
      "mean_token_accuracy": 0.16973724961280823,
      "num_tokens": 3160205.0,
      "step": 1405
    },
    {
      "entropy": 5.9592376232147215,
      "epoch": 1.5246078961600866,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004972699898791927,
      "loss": 5.9833,
      "mean_token_accuracy": 0.16072821915149688,
      "num_tokens": 3171749.0,
      "step": 1410
    },
    {
      "entropy": 5.9620068073272705,
      "epoch": 1.530016224986479,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004972029729503182,
      "loss": 5.9394,
      "mean_token_accuracy": 0.1557013750076294,
      "num_tokens": 3183303.0,
      "step": 1415
    },
    {
      "entropy": 6.018831586837768,
      "epoch": 1.535424553812872,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004971351485353265,
      "loss": 5.9681,
      "mean_token_accuracy": 0.1568077340722084,
      "num_tokens": 3194784.0,
      "step": 1420
    },
    {
      "entropy": 5.939594316482544,
      "epoch": 1.5408328826392643,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004970665168806917,
      "loss": 5.7919,
      "mean_token_accuracy": 0.1602090999484062,
      "num_tokens": 3205732.0,
      "step": 1425
    },
    {
      "entropy": 5.892387580871582,
      "epoch": 1.5462412114656572,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004969970782358214,
      "loss": 5.8812,
      "mean_token_accuracy": 0.15621185600757598,
      "num_tokens": 3216927.0,
      "step": 1430
    },
    {
      "entropy": 5.9382976531982425,
      "epoch": 1.5516495402920496,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004969268328530557,
      "loss": 5.8656,
      "mean_token_accuracy": 0.16953830569982528,
      "num_tokens": 3227464.0,
      "step": 1435
    },
    {
      "entropy": 5.889416885375977,
      "epoch": 1.5570578691184425,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004968557809876666,
      "loss": 5.892,
      "mean_token_accuracy": 0.1629958361387253,
      "num_tokens": 3239357.0,
      "step": 1440
    },
    {
      "entropy": 5.872169113159179,
      "epoch": 1.562466197944835,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004967839228978568,
      "loss": 5.6692,
      "mean_token_accuracy": 0.16706007570028306,
      "num_tokens": 3250893.0,
      "step": 1445
    },
    {
      "entropy": 5.700922346115112,
      "epoch": 1.5678745267712277,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004967112588447585,
      "loss": 5.7193,
      "mean_token_accuracy": 0.17445658445358275,
      "num_tokens": 3262081.0,
      "step": 1450
    },
    {
      "entropy": 5.831270933151245,
      "epoch": 1.5732828555976204,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004966377890924332,
      "loss": 5.8279,
      "mean_token_accuracy": 0.17075785547494887,
      "num_tokens": 3273059.0,
      "step": 1455
    },
    {
      "entropy": 5.984502220153809,
      "epoch": 1.578691184424013,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004965635139078704,
      "loss": 5.8425,
      "mean_token_accuracy": 0.1614866703748703,
      "num_tokens": 3283901.0,
      "step": 1460
    },
    {
      "entropy": 5.857378387451172,
      "epoch": 1.5840995132504057,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004964884335609857,
      "loss": 5.9535,
      "mean_token_accuracy": 0.15531671792268753,
      "num_tokens": 3295687.0,
      "step": 1465
    },
    {
      "entropy": 5.899958944320678,
      "epoch": 1.5895078420767983,
      "grad_norm": 1.125,
      "learning_rate": 0.0004964125483246218,
      "loss": 5.781,
      "mean_token_accuracy": 0.16107384413480758,
      "num_tokens": 3305600.0,
      "step": 1470
    },
    {
      "entropy": 5.8675659656524655,
      "epoch": 1.594916170903191,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004963358584745457,
      "loss": 5.8908,
      "mean_token_accuracy": 0.15747047811746598,
      "num_tokens": 3316337.0,
      "step": 1475
    },
    {
      "entropy": 5.928437662124634,
      "epoch": 1.6003244997295836,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004962583642894482,
      "loss": 5.9135,
      "mean_token_accuracy": 0.16109205186367034,
      "num_tokens": 3327214.0,
      "step": 1480
    },
    {
      "entropy": 5.892904615402221,
      "epoch": 1.6057328285559762,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004961800660509437,
      "loss": 5.8,
      "mean_token_accuracy": 0.15795575678348542,
      "num_tokens": 3338663.0,
      "step": 1485
    },
    {
      "entropy": 5.8454772472381595,
      "epoch": 1.6111411573823688,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000496100964043568,
      "loss": 5.73,
      "mean_token_accuracy": 0.16322666704654692,
      "num_tokens": 3351278.0,
      "step": 1490
    },
    {
      "entropy": 5.931005859375,
      "epoch": 1.6165494862087615,
      "grad_norm": 1.125,
      "learning_rate": 0.000496021058554778,
      "loss": 5.8335,
      "mean_token_accuracy": 0.16325978934764862,
      "num_tokens": 3363144.0,
      "step": 1495
    },
    {
      "entropy": 5.68018856048584,
      "epoch": 1.6219578150351541,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004959403498749505,
      "loss": 5.761,
      "mean_token_accuracy": 0.16344596147537233,
      "num_tokens": 3373565.0,
      "step": 1500
    },
    {
      "entropy": 5.926155233383179,
      "epoch": 1.6273661438615468,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004958588382973809,
      "loss": 5.8746,
      "mean_token_accuracy": 0.16343246549367904,
      "num_tokens": 3384486.0,
      "step": 1505
    },
    {
      "entropy": 5.84269151687622,
      "epoch": 1.6327744726879394,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004957765241182826,
      "loss": 5.8588,
      "mean_token_accuracy": 0.15702430605888368,
      "num_tokens": 3395808.0,
      "step": 1510
    },
    {
      "entropy": 5.915395021438599,
      "epoch": 1.638182801514332,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004956934076367854,
      "loss": 5.7329,
      "mean_token_accuracy": 0.16698796004056932,
      "num_tokens": 3406515.0,
      "step": 1515
    },
    {
      "entropy": 5.708258962631225,
      "epoch": 1.6435911303407247,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000495609489154935,
      "loss": 5.8782,
      "mean_token_accuracy": 0.15216176956892014,
      "num_tokens": 3416507.0,
      "step": 1520
    },
    {
      "entropy": 6.006617736816406,
      "epoch": 1.6489994591671173,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004955247689776911,
      "loss": 5.8751,
      "mean_token_accuracy": 0.16319621950387955,
      "num_tokens": 3428195.0,
      "step": 1525
    },
    {
      "entropy": 5.8480085849761965,
      "epoch": 1.65440778799351,
      "grad_norm": 1.078125,
      "learning_rate": 0.000495439247412927,
      "loss": 5.8367,
      "mean_token_accuracy": 0.15666660219430922,
      "num_tokens": 3439500.0,
      "step": 1530
    },
    {
      "entropy": 5.699786853790283,
      "epoch": 1.6598161168199026,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004953529247714285,
      "loss": 5.6818,
      "mean_token_accuracy": 0.16468058079481124,
      "num_tokens": 3451042.0,
      "step": 1535
    },
    {
      "entropy": 5.882201433181763,
      "epoch": 1.6652244456462952,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004952658013668922,
      "loss": 5.9044,
      "mean_token_accuracy": 0.16032216250896453,
      "num_tokens": 3462286.0,
      "step": 1540
    },
    {
      "entropy": 5.886834716796875,
      "epoch": 1.670632774472688,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004951778775159246,
      "loss": 5.794,
      "mean_token_accuracy": 0.1650930419564247,
      "num_tokens": 3473309.0,
      "step": 1545
    },
    {
      "entropy": 5.836314010620117,
      "epoch": 1.6760411032990805,
      "grad_norm": 1.0,
      "learning_rate": 0.0004950891535380414,
      "loss": 5.8509,
      "mean_token_accuracy": 0.16708696037530898,
      "num_tokens": 3485733.0,
      "step": 1550
    },
    {
      "entropy": 5.806645298004151,
      "epoch": 1.6814494321254734,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004949996297556656,
      "loss": 5.6799,
      "mean_token_accuracy": 0.17222748547792435,
      "num_tokens": 3495585.0,
      "step": 1555
    },
    {
      "entropy": 5.823585128784179,
      "epoch": 1.6868577609518658,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004949093064941268,
      "loss": 5.919,
      "mean_token_accuracy": 0.1641972318291664,
      "num_tokens": 3507973.0,
      "step": 1560
    },
    {
      "entropy": 5.878605794906616,
      "epoch": 1.6922660897782587,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004948181840816601,
      "loss": 5.8549,
      "mean_token_accuracy": 0.15803509652614595,
      "num_tokens": 3518276.0,
      "step": 1565
    },
    {
      "entropy": 5.7111487865448,
      "epoch": 1.697674418604651,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004947262628494045,
      "loss": 5.5794,
      "mean_token_accuracy": 0.16848823577165603,
      "num_tokens": 3529786.0,
      "step": 1570
    },
    {
      "entropy": 5.778642225265503,
      "epoch": 1.703082747431044,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004946335431314018,
      "loss": 5.6825,
      "mean_token_accuracy": 0.17200134098529815,
      "num_tokens": 3540884.0,
      "step": 1575
    },
    {
      "entropy": 5.810184478759766,
      "epoch": 1.7084910762574363,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004945400252645959,
      "loss": 5.8803,
      "mean_token_accuracy": 0.16829971969127655,
      "num_tokens": 3553246.0,
      "step": 1580
    },
    {
      "entropy": 5.857252788543701,
      "epoch": 1.7138994050838292,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004944457095888308,
      "loss": 5.7839,
      "mean_token_accuracy": 0.16421065777540206,
      "num_tokens": 3563626.0,
      "step": 1585
    },
    {
      "entropy": 5.899531316757202,
      "epoch": 1.7193077339102216,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00049435059644685,
      "loss": 5.7919,
      "mean_token_accuracy": 0.16754550635814666,
      "num_tokens": 3574221.0,
      "step": 1590
    },
    {
      "entropy": 5.7845847606658936,
      "epoch": 1.7247160627366145,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004942546861842947,
      "loss": 5.8729,
      "mean_token_accuracy": 0.1651529476046562,
      "num_tokens": 3585111.0,
      "step": 1595
    },
    {
      "entropy": 5.907976865768433,
      "epoch": 1.730124391563007,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004941579791497032,
      "loss": 5.8713,
      "mean_token_accuracy": 0.1592499554157257,
      "num_tokens": 3595614.0,
      "step": 1600
    },
    {
      "entropy": 5.924918222427368,
      "epoch": 1.7355327203893998,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004940604756945091,
      "loss": 5.8245,
      "mean_token_accuracy": 0.15840010195970536,
      "num_tokens": 3606280.0,
      "step": 1605
    },
    {
      "entropy": 5.925652408599854,
      "epoch": 1.7409410492157922,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004939621761730399,
      "loss": 6.0454,
      "mean_token_accuracy": 0.15688003301620485,
      "num_tokens": 3618212.0,
      "step": 1610
    },
    {
      "entropy": 5.800121688842774,
      "epoch": 1.746349378042185,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004938630809425166,
      "loss": 5.7711,
      "mean_token_accuracy": 0.16920888274908066,
      "num_tokens": 3629869.0,
      "step": 1615
    },
    {
      "entropy": 5.8043921947479244,
      "epoch": 1.7517577068685775,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004937631903630513,
      "loss": 5.7201,
      "mean_token_accuracy": 0.16540371924638747,
      "num_tokens": 3640522.0,
      "step": 1620
    },
    {
      "entropy": 5.8789771556854244,
      "epoch": 1.7571660356949703,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004936625047976468,
      "loss": 5.9076,
      "mean_token_accuracy": 0.15947771072387695,
      "num_tokens": 3652407.0,
      "step": 1625
    },
    {
      "entropy": 5.766597938537598,
      "epoch": 1.7625743645213627,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004935610246121944,
      "loss": 5.6778,
      "mean_token_accuracy": 0.16362657248973847,
      "num_tokens": 3663256.0,
      "step": 1630
    },
    {
      "entropy": 5.953878927230835,
      "epoch": 1.7679826933477556,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004934587501754736,
      "loss": 5.9086,
      "mean_token_accuracy": 0.15510267913341522,
      "num_tokens": 3675029.0,
      "step": 1635
    },
    {
      "entropy": 5.794557094573975,
      "epoch": 1.7733910221741482,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004933556818591498,
      "loss": 5.6824,
      "mean_token_accuracy": 0.1662204995751381,
      "num_tokens": 3685565.0,
      "step": 1640
    },
    {
      "entropy": 5.892733764648438,
      "epoch": 1.7787993510005409,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004932518200377736,
      "loss": 5.973,
      "mean_token_accuracy": 0.15337257981300353,
      "num_tokens": 3697189.0,
      "step": 1645
    },
    {
      "entropy": 5.782190942764283,
      "epoch": 1.7842076798269335,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004931471650887789,
      "loss": 5.7113,
      "mean_token_accuracy": 0.16868968158960343,
      "num_tokens": 3708212.0,
      "step": 1650
    },
    {
      "entropy": 5.789211988449097,
      "epoch": 1.7896160086533262,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004930417173924822,
      "loss": 5.8201,
      "mean_token_accuracy": 0.16540355384349822,
      "num_tokens": 3720282.0,
      "step": 1655
    },
    {
      "entropy": 5.806392335891724,
      "epoch": 1.7950243374797188,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004929354773320806,
      "loss": 5.7492,
      "mean_token_accuracy": 0.17258062809705735,
      "num_tokens": 3731466.0,
      "step": 1660
    },
    {
      "entropy": 5.8481114387512205,
      "epoch": 1.8004326663061114,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004928284452936506,
      "loss": 5.7884,
      "mean_token_accuracy": 0.15631649047136306,
      "num_tokens": 3742875.0,
      "step": 1665
    },
    {
      "entropy": 5.81857213973999,
      "epoch": 1.805840995132504,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004927206216661473,
      "loss": 5.7782,
      "mean_token_accuracy": 0.16983126550912858,
      "num_tokens": 3754127.0,
      "step": 1670
    },
    {
      "entropy": 5.814593744277954,
      "epoch": 1.8112493239588967,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004926120068414015,
      "loss": 5.7363,
      "mean_token_accuracy": 0.1717712476849556,
      "num_tokens": 3766287.0,
      "step": 1675
    },
    {
      "entropy": 5.832261562347412,
      "epoch": 1.8166576527852893,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004925026012141201,
      "loss": 5.7533,
      "mean_token_accuracy": 0.16840974539518355,
      "num_tokens": 3777701.0,
      "step": 1680
    },
    {
      "entropy": 5.688795137405395,
      "epoch": 1.822065981611682,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004923924051818832,
      "loss": 5.77,
      "mean_token_accuracy": 0.16398487389087676,
      "num_tokens": 3789545.0,
      "step": 1685
    },
    {
      "entropy": 5.87932186126709,
      "epoch": 1.8274743104380746,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004922814191451435,
      "loss": 5.7877,
      "mean_token_accuracy": 0.1675479829311371,
      "num_tokens": 3799937.0,
      "step": 1690
    },
    {
      "entropy": 5.675029563903808,
      "epoch": 1.8328826392644673,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004921696435072245,
      "loss": 5.7024,
      "mean_token_accuracy": 0.16483644843101503,
      "num_tokens": 3811066.0,
      "step": 1695
    },
    {
      "entropy": 5.856202793121338,
      "epoch": 1.83829096809086,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004920570786743191,
      "loss": 5.8093,
      "mean_token_accuracy": 0.16380842924118041,
      "num_tokens": 3823409.0,
      "step": 1700
    },
    {
      "entropy": 5.8508154392242435,
      "epoch": 1.8436992969172525,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004919437250554882,
      "loss": 5.7945,
      "mean_token_accuracy": 0.1685192406177521,
      "num_tokens": 3835375.0,
      "step": 1705
    },
    {
      "entropy": 5.687268495559692,
      "epoch": 1.8491076257436452,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004918295830626592,
      "loss": 5.6199,
      "mean_token_accuracy": 0.17425741553306578,
      "num_tokens": 3845807.0,
      "step": 1710
    },
    {
      "entropy": 5.788436841964722,
      "epoch": 1.8545159545700378,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004917146531106242,
      "loss": 5.6934,
      "mean_token_accuracy": 0.16580729633569719,
      "num_tokens": 3857391.0,
      "step": 1715
    },
    {
      "entropy": 5.771870946884155,
      "epoch": 1.8599242833964305,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004915989356170389,
      "loss": 5.7494,
      "mean_token_accuracy": 0.17230032980442048,
      "num_tokens": 3869101.0,
      "step": 1720
    },
    {
      "entropy": 5.770860576629639,
      "epoch": 1.865332612222823,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004914824310024213,
      "loss": 5.7552,
      "mean_token_accuracy": 0.17646077275276184,
      "num_tokens": 3880516.0,
      "step": 1725
    },
    {
      "entropy": 5.802186679840088,
      "epoch": 1.870740941049216,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004913651396901489,
      "loss": 5.7375,
      "mean_token_accuracy": 0.17009686082601547,
      "num_tokens": 3891869.0,
      "step": 1730
    },
    {
      "entropy": 5.660962438583374,
      "epoch": 1.8761492698756084,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000491247062106459,
      "loss": 5.6714,
      "mean_token_accuracy": 0.17097828686237335,
      "num_tokens": 3902621.0,
      "step": 1735
    },
    {
      "entropy": 5.697124576568603,
      "epoch": 1.8815575987020012,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004911281986804455,
      "loss": 5.6599,
      "mean_token_accuracy": 0.1695478856563568,
      "num_tokens": 3913052.0,
      "step": 1740
    },
    {
      "entropy": 5.864525413513183,
      "epoch": 1.8869659275283936,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004910085498440587,
      "loss": 5.8739,
      "mean_token_accuracy": 0.16169072985649108,
      "num_tokens": 3924727.0,
      "step": 1745
    },
    {
      "entropy": 5.882341718673706,
      "epoch": 1.8923742563547865,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004908881160321025,
      "loss": 5.8919,
      "mean_token_accuracy": 0.15799911320209503,
      "num_tokens": 3938053.0,
      "step": 1750
    },
    {
      "entropy": 5.737301301956177,
      "epoch": 1.897782585181179,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004907668976822338,
      "loss": 5.7302,
      "mean_token_accuracy": 0.16925743073225022,
      "num_tokens": 3949265.0,
      "step": 1755
    },
    {
      "entropy": 5.788043117523193,
      "epoch": 1.9031909140075718,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004906448952349603,
      "loss": 5.7029,
      "mean_token_accuracy": 0.16863157749176025,
      "num_tokens": 3958808.0,
      "step": 1760
    },
    {
      "entropy": 5.826327991485596,
      "epoch": 1.9085992428339642,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004905221091336392,
      "loss": 5.6841,
      "mean_token_accuracy": 0.16804007589817047,
      "num_tokens": 3970593.0,
      "step": 1765
    },
    {
      "entropy": 5.780668926239014,
      "epoch": 1.914007571660357,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004903985398244753,
      "loss": 5.7403,
      "mean_token_accuracy": 0.16854489743709564,
      "num_tokens": 3981552.0,
      "step": 1770
    },
    {
      "entropy": 5.681759595870972,
      "epoch": 1.9194159004867495,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004902741877565201,
      "loss": 5.6348,
      "mean_token_accuracy": 0.17166891396045686,
      "num_tokens": 3991932.0,
      "step": 1775
    },
    {
      "entropy": 5.894567060470581,
      "epoch": 1.9248242293131423,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000490149053381669,
      "loss": 5.8364,
      "mean_token_accuracy": 0.1588580071926117,
      "num_tokens": 4002234.0,
      "step": 1780
    },
    {
      "entropy": 5.816265201568603,
      "epoch": 1.9302325581395348,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004900231371546608,
      "loss": 5.669,
      "mean_token_accuracy": 0.1731552764773369,
      "num_tokens": 4011800.0,
      "step": 1785
    },
    {
      "entropy": 5.7900268077850345,
      "epoch": 1.9356408869659276,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004898964395330753,
      "loss": 5.8816,
      "mean_token_accuracy": 0.1661091312766075,
      "num_tokens": 4022503.0,
      "step": 1790
    },
    {
      "entropy": 5.808759117126465,
      "epoch": 1.94104921579232,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004897689609773318,
      "loss": 5.7556,
      "mean_token_accuracy": 0.16821117252111434,
      "num_tokens": 4032310.0,
      "step": 1795
    },
    {
      "entropy": 5.823995923995971,
      "epoch": 1.946457544618713,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004896407019506879,
      "loss": 5.7947,
      "mean_token_accuracy": 0.16529604643583298,
      "num_tokens": 4043290.0,
      "step": 1800
    },
    {
      "entropy": 5.788540172576904,
      "epoch": 1.9518658734451053,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004895116629192371,
      "loss": 5.8935,
      "mean_token_accuracy": 0.15940507650375366,
      "num_tokens": 4054550.0,
      "step": 1805
    },
    {
      "entropy": 5.859425354003906,
      "epoch": 1.9572742022714982,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004893818443519074,
      "loss": 5.6579,
      "mean_token_accuracy": 0.17276726216077803,
      "num_tokens": 4064740.0,
      "step": 1810
    },
    {
      "entropy": 5.77848072052002,
      "epoch": 1.9626825310978906,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048925124672046,
      "loss": 5.8104,
      "mean_token_accuracy": 0.16045169085264205,
      "num_tokens": 4075466.0,
      "step": 1815
    },
    {
      "entropy": 5.7506016254425045,
      "epoch": 1.9680908599242835,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004891198704994868,
      "loss": 5.7746,
      "mean_token_accuracy": 0.16419024467468263,
      "num_tokens": 4085794.0,
      "step": 1820
    },
    {
      "entropy": 5.774007272720337,
      "epoch": 1.973499188750676,
      "grad_norm": 1.0,
      "learning_rate": 0.0004889877161664096,
      "loss": 5.776,
      "mean_token_accuracy": 0.16607008278369903,
      "num_tokens": 4097587.0,
      "step": 1825
    },
    {
      "entropy": 5.8042127132415775,
      "epoch": 1.9789075175770687,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004888547842014771,
      "loss": 5.7397,
      "mean_token_accuracy": 0.16423825323581695,
      "num_tokens": 4110194.0,
      "step": 1830
    },
    {
      "entropy": 5.843971633911133,
      "epoch": 1.9843158464034614,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004887210750877648,
      "loss": 5.788,
      "mean_token_accuracy": 0.16245697140693666,
      "num_tokens": 4121967.0,
      "step": 1835
    },
    {
      "entropy": 5.6416247367858885,
      "epoch": 1.989724175229854,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004885865893111716,
      "loss": 5.6734,
      "mean_token_accuracy": 0.17218596786260604,
      "num_tokens": 4131457.0,
      "step": 1840
    },
    {
      "entropy": 5.8589592456817625,
      "epoch": 1.9951325040562466,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004884513273604195,
      "loss": 5.795,
      "mean_token_accuracy": 0.1732928141951561,
      "num_tokens": 4143820.0,
      "step": 1845
    },
    {
      "entropy": 5.841017987993029,
      "epoch": 2.0,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004883152897270502,
      "loss": 5.8574,
      "mean_token_accuracy": 0.163099843594763,
      "num_tokens": 4153778.0,
      "step": 1850
    },
    {
      "entropy": 5.737104320526123,
      "epoch": 2.005408328826393,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004881784769054253,
      "loss": 5.4329,
      "mean_token_accuracy": 0.1771388202905655,
      "num_tokens": 4166682.0,
      "step": 1855
    },
    {
      "entropy": 5.7277398109436035,
      "epoch": 2.0108166576527853,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004880408893927226,
      "loss": 5.4478,
      "mean_token_accuracy": 0.17328148931264878,
      "num_tokens": 4178621.0,
      "step": 1860
    },
    {
      "entropy": 5.772768115997314,
      "epoch": 2.016224986479178,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004879025276889355,
      "loss": 5.3683,
      "mean_token_accuracy": 0.18826774805784224,
      "num_tokens": 4189956.0,
      "step": 1865
    },
    {
      "entropy": 5.555465173721314,
      "epoch": 2.0216333153055706,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048776339229687077,
      "loss": 5.3996,
      "mean_token_accuracy": 0.17681935131549836,
      "num_tokens": 4202547.0,
      "step": 1870
    },
    {
      "entropy": 5.801024436950684,
      "epoch": 2.0270416441319634,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048762348372214683,
      "loss": 5.3898,
      "mean_token_accuracy": 0.18697095662355423,
      "num_tokens": 4212416.0,
      "step": 1875
    },
    {
      "entropy": 5.594231700897216,
      "epoch": 2.032449972958356,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048748280247319165,
      "loss": 5.3488,
      "mean_token_accuracy": 0.17242017537355422,
      "num_tokens": 4223710.0,
      "step": 1880
    },
    {
      "entropy": 5.726772785186768,
      "epoch": 2.0378583017847487,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004873413490612412,
      "loss": 5.4429,
      "mean_token_accuracy": 0.17841096520423888,
      "num_tokens": 4235245.0,
      "step": 1885
    },
    {
      "entropy": 5.665938758850098,
      "epoch": 2.043266630611141,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048719912400033757,
      "loss": 5.3764,
      "mean_token_accuracy": 0.18664392828941345,
      "num_tokens": 4247113.0,
      "step": 1890
    },
    {
      "entropy": 5.736206531524658,
      "epoch": 2.048674959437534,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000487056127807327,
      "loss": 5.4366,
      "mean_token_accuracy": 0.1796937808394432,
      "num_tokens": 4259413.0,
      "step": 1895
    },
    {
      "entropy": 5.617617559432984,
      "epoch": 2.0540832882639264,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048691236100185795,
      "loss": 5.3371,
      "mean_token_accuracy": 0.1845716044306755,
      "num_tokens": 4269413.0,
      "step": 1900
    },
    {
      "entropy": 5.596398448944091,
      "epoch": 2.0594916170903192,
      "grad_norm": 1.125,
      "learning_rate": 0.0004867678241063792,
      "loss": 5.3148,
      "mean_token_accuracy": 0.19401925355195998,
      "num_tokens": 4279453.0,
      "step": 1905
    },
    {
      "entropy": 5.688255643844604,
      "epoch": 2.0648999459167117,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048662251764613833,
      "loss": 5.4226,
      "mean_token_accuracy": 0.18174491822719574,
      "num_tokens": 4290958.0,
      "step": 1910
    },
    {
      "entropy": 5.5896929740905765,
      "epoch": 2.0703082747431045,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004864764421491793,
      "loss": 5.3235,
      "mean_token_accuracy": 0.18415751457214355,
      "num_tokens": 4301867.0,
      "step": 1915
    },
    {
      "entropy": 5.628929805755615,
      "epoch": 2.075716603569497,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004863295981463408,
      "loss": 5.3725,
      "mean_token_accuracy": 0.17780343294143677,
      "num_tokens": 4314890.0,
      "step": 1920
    },
    {
      "entropy": 5.734420442581177,
      "epoch": 2.08112493239589,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004861819861712543,
      "loss": 5.5296,
      "mean_token_accuracy": 0.1750368744134903,
      "num_tokens": 4325368.0,
      "step": 1925
    },
    {
      "entropy": 5.699685955047608,
      "epoch": 2.086533261222282,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048603360676034203,
      "loss": 5.4862,
      "mean_token_accuracy": 0.17520413398742676,
      "num_tokens": 4337469.0,
      "step": 1930
    },
    {
      "entropy": 5.71870493888855,
      "epoch": 2.091941590048675,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048588446045281505,
      "loss": 5.411,
      "mean_token_accuracy": 0.18186349123716355,
      "num_tokens": 4349115.0,
      "step": 1935
    },
    {
      "entropy": 5.632260799407959,
      "epoch": 2.0973499188750675,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004857345477906714,
      "loss": 5.3729,
      "mean_token_accuracy": 0.18043224960565568,
      "num_tokens": 4360287.0,
      "step": 1940
    },
    {
      "entropy": 5.676126623153687,
      "epoch": 2.1027582477014604,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048558386931869417,
      "loss": 5.5284,
      "mean_token_accuracy": 0.17184396237134933,
      "num_tokens": 4371966.0,
      "step": 1945
    },
    {
      "entropy": 5.658498477935791,
      "epoch": 2.1081665765278528,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004854324255844492,
      "loss": 5.4212,
      "mean_token_accuracy": 0.18277220278978348,
      "num_tokens": 4382837.0,
      "step": 1950
    },
    {
      "entropy": 5.5803807258605955,
      "epoch": 2.1135749053542456,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048528021713828336,
      "loss": 5.385,
      "mean_token_accuracy": 0.18133683502674103,
      "num_tokens": 4393970.0,
      "step": 1955
    },
    {
      "entropy": 5.634076642990112,
      "epoch": 2.118983234180638,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004851272445333225,
      "loss": 5.3915,
      "mean_token_accuracy": 0.1862371161580086,
      "num_tokens": 4404450.0,
      "step": 1960
    },
    {
      "entropy": 5.712316226959229,
      "epoch": 2.124391563007031,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048497350832546937,
      "loss": 5.5442,
      "mean_token_accuracy": 0.1754376158118248,
      "num_tokens": 4417203.0,
      "step": 1965
    },
    {
      "entropy": 5.699717664718628,
      "epoch": 2.1297998918334233,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004848190090734018,
      "loss": 5.4613,
      "mean_token_accuracy": 0.17721105217933655,
      "num_tokens": 4429458.0,
      "step": 1970
    },
    {
      "entropy": 5.758607196807861,
      "epoch": 2.135208220659816,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048466374733857035,
      "loss": 5.5414,
      "mean_token_accuracy": 0.16932629495859147,
      "num_tokens": 4441445.0,
      "step": 1975
    },
    {
      "entropy": 5.592848825454712,
      "epoch": 2.1406165494862086,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004845077236851966,
      "loss": 5.425,
      "mean_token_accuracy": 0.17384567111730576,
      "num_tokens": 4453152.0,
      "step": 1980
    },
    {
      "entropy": 5.719186878204345,
      "epoch": 2.1460248783126015,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048435093868027076,
      "loss": 5.3971,
      "mean_token_accuracy": 0.18340317606925965,
      "num_tokens": 4464478.0,
      "step": 1985
    },
    {
      "entropy": 5.505934143066407,
      "epoch": 2.151433207138994,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048419339289354995,
      "loss": 5.4474,
      "mean_token_accuracy": 0.17782567739486693,
      "num_tokens": 4476253.0,
      "step": 1990
    },
    {
      "entropy": 5.706979608535766,
      "epoch": 2.1568415359653867,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000484035086897556,
      "loss": 5.4232,
      "mean_token_accuracy": 0.18010570853948593,
      "num_tokens": 4488209.0,
      "step": 1995
    },
    {
      "entropy": 5.565637159347534,
      "epoch": 2.162249864791779,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004838760212675732,
      "loss": 5.365,
      "mean_token_accuracy": 0.18658868819475175,
      "num_tokens": 4499937.0,
      "step": 2000
    },
    {
      "epoch": 2.162249864791779,
      "eval_entropy": 5.472852083769712,
      "eval_loss": 6.006684303283691,
      "eval_mean_token_accuracy": 0.16748105433176863,
      "eval_num_tokens": 4499937.0,
      "eval_runtime": 2.6965,
      "eval_samples_per_second": 1303.524,
      "eval_steps_per_second": 163.172,
      "step": 2000
    },
    {
      "entropy": 5.576435613632202,
      "epoch": 2.167658193618172,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004837161965816464,
      "loss": 5.4196,
      "mean_token_accuracy": 0.16989136040210723,
      "num_tokens": 4510567.0,
      "step": 2005
    },
    {
      "entropy": 5.540714836120605,
      "epoch": 2.1730665224445644,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048355561342057886,
      "loss": 5.351,
      "mean_token_accuracy": 0.18173110485076904,
      "num_tokens": 4522695.0,
      "step": 2010
    },
    {
      "entropy": 5.63709716796875,
      "epoch": 2.1784748512709573,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048339427236793024,
      "loss": 5.3435,
      "mean_token_accuracy": 0.18755321353673934,
      "num_tokens": 4534642.0,
      "step": 2015
    },
    {
      "entropy": 5.672750425338745,
      "epoch": 2.1838831800973497,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004832321740100143,
      "loss": 5.474,
      "mean_token_accuracy": 0.17547187954187393,
      "num_tokens": 4546021.0,
      "step": 2020
    },
    {
      "entropy": 5.571087646484375,
      "epoch": 2.1892915089237426,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048306931893589666,
      "loss": 5.3209,
      "mean_token_accuracy": 0.17968493103981018,
      "num_tokens": 4556814.0,
      "step": 2025
    },
    {
      "entropy": 5.581038284301758,
      "epoch": 2.194699837750135,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048290570773739327,
      "loss": 5.3348,
      "mean_token_accuracy": 0.18085583597421645,
      "num_tokens": 4568595.0,
      "step": 2030
    },
    {
      "entropy": 5.618092823028564,
      "epoch": 2.200108166576528,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048274134100906747,
      "loss": 5.4454,
      "mean_token_accuracy": 0.18696052879095076,
      "num_tokens": 4579536.0,
      "step": 2035
    },
    {
      "entropy": 5.591726541519165,
      "epoch": 2.2055164954029207,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048257621934822835,
      "loss": 5.3825,
      "mean_token_accuracy": 0.19023619443178177,
      "num_tokens": 4590497.0,
      "step": 2040
    },
    {
      "entropy": 5.658607339859008,
      "epoch": 2.210924824229313,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004824103433549285,
      "loss": 5.48,
      "mean_token_accuracy": 0.1759459152817726,
      "num_tokens": 4601699.0,
      "step": 2045
    },
    {
      "entropy": 5.565668439865112,
      "epoch": 2.2163331530557056,
      "grad_norm": 1.125,
      "learning_rate": 0.0004822437136319616,
      "loss": 5.3915,
      "mean_token_accuracy": 0.17944610118865967,
      "num_tokens": 4611857.0,
      "step": 2050
    },
    {
      "entropy": 5.6317449569702145,
      "epoch": 2.2217414818820984,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004820763307848604,
      "loss": 5.4326,
      "mean_token_accuracy": 0.18303177803754805,
      "num_tokens": 4623042.0,
      "step": 2055
    },
    {
      "entropy": 5.515517139434815,
      "epoch": 2.2271498107084913,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048190819542189456,
      "loss": 5.4309,
      "mean_token_accuracy": 0.1750242218375206,
      "num_tokens": 4634398.0,
      "step": 2060
    },
    {
      "entropy": 5.572394561767578,
      "epoch": 2.2325581395348837,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048173930815406847,
      "loss": 5.3289,
      "mean_token_accuracy": 0.1918608710169792,
      "num_tokens": 4645140.0,
      "step": 2065
    },
    {
      "entropy": 5.510449647903442,
      "epoch": 2.2379664683612766,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004815696695951187,
      "loss": 5.3579,
      "mean_token_accuracy": 0.17949569076299668,
      "num_tokens": 4657490.0,
      "step": 2070
    },
    {
      "entropy": 5.617029333114624,
      "epoch": 2.243374797187669,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004813992803615123,
      "loss": 5.3953,
      "mean_token_accuracy": 0.1817226678133011,
      "num_tokens": 4670109.0,
      "step": 2075
    },
    {
      "entropy": 5.497842073440552,
      "epoch": 2.248783126014062,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048122814107244414,
      "loss": 5.325,
      "mean_token_accuracy": 0.19015913605690002,
      "num_tokens": 4681498.0,
      "step": 2080
    },
    {
      "entropy": 5.674573040008545,
      "epoch": 2.2541914548404542,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048105625234983466,
      "loss": 5.5858,
      "mean_token_accuracy": 0.1682846337556839,
      "num_tokens": 4693351.0,
      "step": 2085
    },
    {
      "entropy": 5.653087711334228,
      "epoch": 2.259599783666847,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048088361481832807,
      "loss": 5.522,
      "mean_token_accuracy": 0.17714594304561615,
      "num_tokens": 4705088.0,
      "step": 2090
    },
    {
      "entropy": 5.638476133346558,
      "epoch": 2.2650081124932395,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048071022910528934,
      "loss": 5.4,
      "mean_token_accuracy": 0.18836630284786224,
      "num_tokens": 4716834.0,
      "step": 2095
    },
    {
      "entropy": 5.64174747467041,
      "epoch": 2.2704164413196324,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048053609584080274,
      "loss": 5.4491,
      "mean_token_accuracy": 0.17846577763557434,
      "num_tokens": 4728286.0,
      "step": 2100
    },
    {
      "entropy": 5.576927280426025,
      "epoch": 2.275824770146025,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048036121565766876,
      "loss": 5.3714,
      "mean_token_accuracy": 0.18769163638353348,
      "num_tokens": 4739296.0,
      "step": 2105
    },
    {
      "entropy": 5.574035024642944,
      "epoch": 2.2812330989724177,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004801855891914026,
      "loss": 5.3497,
      "mean_token_accuracy": 0.19533850848674775,
      "num_tokens": 4749526.0,
      "step": 2110
    },
    {
      "entropy": 5.5547194480896,
      "epoch": 2.28664142779881,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048000921708023117,
      "loss": 5.3854,
      "mean_token_accuracy": 0.18343230336904526,
      "num_tokens": 4760496.0,
      "step": 2115
    },
    {
      "entropy": 5.5593877792358395,
      "epoch": 2.292049756625203,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004798320999650911,
      "loss": 5.3665,
      "mean_token_accuracy": 0.1883173793554306,
      "num_tokens": 4771079.0,
      "step": 2120
    },
    {
      "entropy": 5.538218402862549,
      "epoch": 2.2974580854515954,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047965423848962636,
      "loss": 5.3897,
      "mean_token_accuracy": 0.19219416081905366,
      "num_tokens": 4781960.0,
      "step": 2125
    },
    {
      "entropy": 5.5187544345855715,
      "epoch": 2.302866414277988,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000479475633300186,
      "loss": 5.304,
      "mean_token_accuracy": 0.18261959850788118,
      "num_tokens": 4793174.0,
      "step": 2130
    },
    {
      "entropy": 5.629965209960938,
      "epoch": 2.3082747431043806,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047929628504582173,
      "loss": 5.4836,
      "mean_token_accuracy": 0.1778162494301796,
      "num_tokens": 4804344.0,
      "step": 2135
    },
    {
      "entropy": 5.6045232772827145,
      "epoch": 2.3136830719307735,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004791161943782856,
      "loss": 5.4379,
      "mean_token_accuracy": 0.18713171631097794,
      "num_tokens": 4815838.0,
      "step": 2140
    },
    {
      "entropy": 5.535661602020264,
      "epoch": 2.319091400757166,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047893536195202726,
      "loss": 5.3257,
      "mean_token_accuracy": 0.18756370097398758,
      "num_tokens": 4826153.0,
      "step": 2145
    },
    {
      "entropy": 5.4889808177948,
      "epoch": 2.3244997295835588,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004787537884241923,
      "loss": 5.4147,
      "mean_token_accuracy": 0.1823410928249359,
      "num_tokens": 4838647.0,
      "step": 2150
    },
    {
      "entropy": 5.69741678237915,
      "epoch": 2.329908058409951,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047857147445461943,
      "loss": 5.5028,
      "mean_token_accuracy": 0.174128720164299,
      "num_tokens": 4849320.0,
      "step": 2155
    },
    {
      "entropy": 5.574743413925171,
      "epoch": 2.335316387236344,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004783884207058379,
      "loss": 5.3685,
      "mean_token_accuracy": 0.1826545551419258,
      "num_tokens": 4860079.0,
      "step": 2160
    },
    {
      "entropy": 5.509880256652832,
      "epoch": 2.3407247160627365,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047820462784306555,
      "loss": 5.3293,
      "mean_token_accuracy": 0.1828920543193817,
      "num_tokens": 4871073.0,
      "step": 2165
    },
    {
      "entropy": 5.555741643905639,
      "epoch": 2.3461330448891293,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000478020096534206,
      "loss": 5.3453,
      "mean_token_accuracy": 0.18160604387521745,
      "num_tokens": 4881517.0,
      "step": 2170
    },
    {
      "entropy": 5.546141862869263,
      "epoch": 2.3515413737155217,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047783482744984644,
      "loss": 5.4308,
      "mean_token_accuracy": 0.1774826467037201,
      "num_tokens": 4892554.0,
      "step": 2175
    },
    {
      "entropy": 5.581753587722778,
      "epoch": 2.3569497025419146,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047764882126325527,
      "loss": 5.4502,
      "mean_token_accuracy": 0.17918796688318253,
      "num_tokens": 4903168.0,
      "step": 2180
    },
    {
      "entropy": 5.592440605163574,
      "epoch": 2.362358031368307,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004774620786503793,
      "loss": 5.4973,
      "mean_token_accuracy": 0.17680401653051375,
      "num_tokens": 4914213.0,
      "step": 2185
    },
    {
      "entropy": 5.670473623275757,
      "epoch": 2.3677663601947,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004772746002898417,
      "loss": 5.4517,
      "mean_token_accuracy": 0.18014305382966994,
      "num_tokens": 4926271.0,
      "step": 2190
    },
    {
      "entropy": 5.532607746124268,
      "epoch": 2.3731746890210923,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047708638686293903,
      "loss": 5.3515,
      "mean_token_accuracy": 0.19012952744960784,
      "num_tokens": 4938386.0,
      "step": 2195
    },
    {
      "entropy": 5.573096704483032,
      "epoch": 2.378583017847485,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047689743905363963,
      "loss": 5.3258,
      "mean_token_accuracy": 0.18657357543706893,
      "num_tokens": 4949920.0,
      "step": 2200
    },
    {
      "entropy": 5.493396759033203,
      "epoch": 2.383991346673878,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004767077575485801,
      "loss": 5.4508,
      "mean_token_accuracy": 0.17724529951810836,
      "num_tokens": 4961639.0,
      "step": 2205
    },
    {
      "entropy": 5.576059484481812,
      "epoch": 2.3893996755002704,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004765173430370635,
      "loss": 5.3795,
      "mean_token_accuracy": 0.1817702829837799,
      "num_tokens": 4972863.0,
      "step": 2210
    },
    {
      "entropy": 5.557861566543579,
      "epoch": 2.394808004326663,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004763261962110567,
      "loss": 5.3465,
      "mean_token_accuracy": 0.18829651176929474,
      "num_tokens": 4982767.0,
      "step": 2215
    },
    {
      "entropy": 5.482148265838623,
      "epoch": 2.4002163331530557,
      "grad_norm": 1.0,
      "learning_rate": 0.0004761343177651875,
      "loss": 5.3647,
      "mean_token_accuracy": 0.181600858271122,
      "num_tokens": 4993610.0,
      "step": 2220
    },
    {
      "entropy": 5.647186183929444,
      "epoch": 2.4056246619794486,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004759417083967428,
      "loss": 5.4748,
      "mean_token_accuracy": 0.18320268392562866,
      "num_tokens": 5004443.0,
      "step": 2225
    },
    {
      "entropy": 5.47214641571045,
      "epoch": 2.411032990805841,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047574836880566544,
      "loss": 5.322,
      "mean_token_accuracy": 0.18720456659793855,
      "num_tokens": 5014551.0,
      "step": 2230
    },
    {
      "entropy": 5.606646919250489,
      "epoch": 2.4164413196322334,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004755542996945519,
      "loss": 5.4073,
      "mean_token_accuracy": 0.18585125505924224,
      "num_tokens": 5025561.0,
      "step": 2235
    },
    {
      "entropy": 5.571566343307495,
      "epoch": 2.4218496484586263,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047535950176864976,
      "loss": 5.3324,
      "mean_token_accuracy": 0.19187741875648498,
      "num_tokens": 5036515.0,
      "step": 2240
    },
    {
      "entropy": 5.440071678161621,
      "epoch": 2.427257977285019,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047516397573585525,
      "loss": 5.3935,
      "mean_token_accuracy": 0.17414860427379608,
      "num_tokens": 5048228.0,
      "step": 2245
    },
    {
      "entropy": 5.583653306961059,
      "epoch": 2.4326663061114115,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047496772230671034,
      "loss": 5.4135,
      "mean_token_accuracy": 0.17929787188768387,
      "num_tokens": 5059574.0,
      "step": 2250
    },
    {
      "entropy": 5.500989532470703,
      "epoch": 2.4380746349378044,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004747707421944004,
      "loss": 5.345,
      "mean_token_accuracy": 0.1873723477125168,
      "num_tokens": 5070675.0,
      "step": 2255
    },
    {
      "entropy": 5.499913167953491,
      "epoch": 2.443482963764197,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004745730361147517,
      "loss": 5.3294,
      "mean_token_accuracy": 0.18599451929330826,
      "num_tokens": 5082620.0,
      "step": 2260
    },
    {
      "entropy": 5.598516607284546,
      "epoch": 2.4488912925905897,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004743746047862286,
      "loss": 5.3919,
      "mean_token_accuracy": 0.1780677855014801,
      "num_tokens": 5094120.0,
      "step": 2265
    },
    {
      "entropy": 5.5201959133148195,
      "epoch": 2.454299621416982,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004741754489299309,
      "loss": 5.3842,
      "mean_token_accuracy": 0.184432190656662,
      "num_tokens": 5104681.0,
      "step": 2270
    },
    {
      "entropy": 5.505268478393555,
      "epoch": 2.459707950243375,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047397556926959155,
      "loss": 5.3815,
      "mean_token_accuracy": 0.19052304178476334,
      "num_tokens": 5114901.0,
      "step": 2275
    },
    {
      "entropy": 5.483372545242309,
      "epoch": 2.4651162790697674,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047377496653157374,
      "loss": 5.3595,
      "mean_token_accuracy": 0.18044475913047792,
      "num_tokens": 5125947.0,
      "step": 2280
    },
    {
      "entropy": 5.68227915763855,
      "epoch": 2.4705246078961602,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047357364144486806,
      "loss": 5.4893,
      "mean_token_accuracy": 0.17698677629232407,
      "num_tokens": 5137905.0,
      "step": 2285
    },
    {
      "entropy": 5.558288478851319,
      "epoch": 2.4759329367225527,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047337159474109057,
      "loss": 5.4702,
      "mean_token_accuracy": 0.17549121230840684,
      "num_tokens": 5150266.0,
      "step": 2290
    },
    {
      "entropy": 5.514402580261231,
      "epoch": 2.4813412655489455,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047316882715447934,
      "loss": 5.4259,
      "mean_token_accuracy": 0.18137722760438918,
      "num_tokens": 5160909.0,
      "step": 2295
    },
    {
      "entropy": 5.480083179473877,
      "epoch": 2.486749594375338,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004729653394218922,
      "loss": 5.4046,
      "mean_token_accuracy": 0.18145928680896758,
      "num_tokens": 5173645.0,
      "step": 2300
    },
    {
      "entropy": 5.62817873954773,
      "epoch": 2.492157923201731,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047276113228280427,
      "loss": 5.5656,
      "mean_token_accuracy": 0.16851997673511504,
      "num_tokens": 5186269.0,
      "step": 2305
    },
    {
      "entropy": 5.606264257431031,
      "epoch": 2.497566252028123,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004725562064793045,
      "loss": 5.3757,
      "mean_token_accuracy": 0.19015525728464128,
      "num_tokens": 5197061.0,
      "step": 2310
    },
    {
      "entropy": 5.458608961105346,
      "epoch": 2.502974580854516,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047235056275609397,
      "loss": 5.3967,
      "mean_token_accuracy": 0.18642975986003876,
      "num_tokens": 5207182.0,
      "step": 2315
    },
    {
      "entropy": 5.570804214477539,
      "epoch": 2.5083829096809085,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004721442018604822,
      "loss": 5.4118,
      "mean_token_accuracy": 0.18205344974994658,
      "num_tokens": 5218381.0,
      "step": 2320
    },
    {
      "entropy": 5.588907957077026,
      "epoch": 2.5137912385073014,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004719371245423853,
      "loss": 5.2995,
      "mean_token_accuracy": 0.19421885162591934,
      "num_tokens": 5228504.0,
      "step": 2325
    },
    {
      "entropy": 5.47154221534729,
      "epoch": 2.5191995673336938,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047172933155432273,
      "loss": 5.4011,
      "mean_token_accuracy": 0.1852772742509842,
      "num_tokens": 5239145.0,
      "step": 2330
    },
    {
      "entropy": 5.434785079956055,
      "epoch": 2.5246078961600866,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047152082365141466,
      "loss": 5.2518,
      "mean_token_accuracy": 0.19392025023698806,
      "num_tokens": 5250697.0,
      "step": 2335
    },
    {
      "entropy": 5.573790073394775,
      "epoch": 2.530016224986479,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004713116015913794,
      "loss": 5.4118,
      "mean_token_accuracy": 0.18861660957336426,
      "num_tokens": 5262876.0,
      "step": 2340
    },
    {
      "entropy": 5.401359987258911,
      "epoch": 2.535424553812872,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047110166613453016,
      "loss": 5.3265,
      "mean_token_accuracy": 0.18738412708044053,
      "num_tokens": 5273813.0,
      "step": 2345
    },
    {
      "entropy": 5.584725189208984,
      "epoch": 2.5408328826392643,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047089101804377305,
      "loss": 5.5047,
      "mean_token_accuracy": 0.17414041757583618,
      "num_tokens": 5285053.0,
      "step": 2350
    },
    {
      "entropy": 5.498822641372681,
      "epoch": 2.546241211465657,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004706796580846037,
      "loss": 5.3338,
      "mean_token_accuracy": 0.19276315420866014,
      "num_tokens": 5295227.0,
      "step": 2355
    },
    {
      "entropy": 5.618280839920044,
      "epoch": 2.5516495402920496,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047046758702510465,
      "loss": 5.4182,
      "mean_token_accuracy": 0.17885240763425828,
      "num_tokens": 5306787.0,
      "step": 2360
    },
    {
      "entropy": 5.500233221054077,
      "epoch": 2.5570578691184425,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047025480563594264,
      "loss": 5.442,
      "mean_token_accuracy": 0.1758480668067932,
      "num_tokens": 5317809.0,
      "step": 2365
    },
    {
      "entropy": 5.597690486907959,
      "epoch": 2.5624661979448353,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047004131469036576,
      "loss": 5.4075,
      "mean_token_accuracy": 0.18495803326368332,
      "num_tokens": 5327738.0,
      "step": 2370
    },
    {
      "entropy": 5.4438732147216795,
      "epoch": 2.5678745267712277,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046982711496420057,
      "loss": 5.3914,
      "mean_token_accuracy": 0.18428546339273452,
      "num_tokens": 5339700.0,
      "step": 2375
    },
    {
      "entropy": 5.552294492721558,
      "epoch": 2.57328285559762,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046961220723584925,
      "loss": 5.4092,
      "mean_token_accuracy": 0.17879543006420134,
      "num_tokens": 5350934.0,
      "step": 2380
    },
    {
      "entropy": 5.463089847564698,
      "epoch": 2.578691184424013,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046939659228628725,
      "loss": 5.3289,
      "mean_token_accuracy": 0.19285490959882737,
      "num_tokens": 5361364.0,
      "step": 2385
    },
    {
      "entropy": 5.589974069595337,
      "epoch": 2.584099513250406,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046918027089905956,
      "loss": 5.5364,
      "mean_token_accuracy": 0.18631306886672974,
      "num_tokens": 5373376.0,
      "step": 2390
    },
    {
      "entropy": 5.527939748764038,
      "epoch": 2.5895078420767983,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046896324386027873,
      "loss": 5.419,
      "mean_token_accuracy": 0.1782801240682602,
      "num_tokens": 5384422.0,
      "step": 2395
    },
    {
      "entropy": 5.6619553565979,
      "epoch": 2.5949161709031907,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046874551195862145,
      "loss": 5.4323,
      "mean_token_accuracy": 0.18902574479579926,
      "num_tokens": 5395834.0,
      "step": 2400
    },
    {
      "entropy": 5.517275857925415,
      "epoch": 2.6003244997295836,
      "grad_norm": 1.0625,
      "learning_rate": 0.000468527075985326,
      "loss": 5.5484,
      "mean_token_accuracy": 0.17533761113882065,
      "num_tokens": 5407749.0,
      "step": 2405
    },
    {
      "entropy": 5.563359642028809,
      "epoch": 2.6057328285559764,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046830793673418924,
      "loss": 5.4235,
      "mean_token_accuracy": 0.18526053428649902,
      "num_tokens": 5418035.0,
      "step": 2410
    },
    {
      "entropy": 5.5195565700531,
      "epoch": 2.611141157382369,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046808809500156355,
      "loss": 5.3512,
      "mean_token_accuracy": 0.1831403434276581,
      "num_tokens": 5429027.0,
      "step": 2415
    },
    {
      "entropy": 5.454291677474975,
      "epoch": 2.6165494862087613,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004678675515863545,
      "loss": 5.3162,
      "mean_token_accuracy": 0.1908747971057892,
      "num_tokens": 5440312.0,
      "step": 2420
    },
    {
      "entropy": 5.524546337127686,
      "epoch": 2.621957815035154,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004676463072900173,
      "loss": 5.3406,
      "mean_token_accuracy": 0.1890124186873436,
      "num_tokens": 5451281.0,
      "step": 2425
    },
    {
      "entropy": 5.481093072891236,
      "epoch": 2.627366143861547,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046742436291655415,
      "loss": 5.3896,
      "mean_token_accuracy": 0.1857441246509552,
      "num_tokens": 5462958.0,
      "step": 2430
    },
    {
      "entropy": 5.560990715026856,
      "epoch": 2.6327744726879394,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046720171927251155,
      "loss": 5.3997,
      "mean_token_accuracy": 0.18795904368162156,
      "num_tokens": 5473364.0,
      "step": 2435
    },
    {
      "entropy": 5.525981903076172,
      "epoch": 2.638182801514332,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004669783771669769,
      "loss": 5.3372,
      "mean_token_accuracy": 0.186283440887928,
      "num_tokens": 5485141.0,
      "step": 2440
    },
    {
      "entropy": 5.501952028274536,
      "epoch": 2.6435911303407247,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000466754337411576,
      "loss": 5.4608,
      "mean_token_accuracy": 0.17689261436462403,
      "num_tokens": 5497802.0,
      "step": 2445
    },
    {
      "entropy": 5.52817325592041,
      "epoch": 2.6489994591671175,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046652960082046985,
      "loss": 5.3104,
      "mean_token_accuracy": 0.18938876837491989,
      "num_tokens": 5507735.0,
      "step": 2450
    },
    {
      "entropy": 5.457919502258301,
      "epoch": 2.65440778799351,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046630416821035164,
      "loss": 5.4141,
      "mean_token_accuracy": 0.18110267370939254,
      "num_tokens": 5519625.0,
      "step": 2455
    },
    {
      "entropy": 5.532788038253784,
      "epoch": 2.6598161168199024,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004660780404004441,
      "loss": 5.3051,
      "mean_token_accuracy": 0.18533146679401397,
      "num_tokens": 5531223.0,
      "step": 2460
    },
    {
      "entropy": 5.4125641822814945,
      "epoch": 2.6652244456462952,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046585121821249623,
      "loss": 5.3335,
      "mean_token_accuracy": 0.1871636375784874,
      "num_tokens": 5543250.0,
      "step": 2465
    },
    {
      "entropy": 5.4998725891113285,
      "epoch": 2.670632774472688,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046562370247078025,
      "loss": 5.3436,
      "mean_token_accuracy": 0.18235380053520203,
      "num_tokens": 5554817.0,
      "step": 2470
    },
    {
      "entropy": 5.529648303985596,
      "epoch": 2.6760411032990805,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000465395494002089,
      "loss": 5.4474,
      "mean_token_accuracy": 0.1817679688334465,
      "num_tokens": 5567716.0,
      "step": 2475
    },
    {
      "entropy": 5.515621042251587,
      "epoch": 2.6814494321254734,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004651665936357325,
      "loss": 5.4621,
      "mean_token_accuracy": 0.18281052708625795,
      "num_tokens": 5579357.0,
      "step": 2480
    },
    {
      "entropy": 5.485055685043335,
      "epoch": 2.686857760951866,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004649370022035354,
      "loss": 5.3676,
      "mean_token_accuracy": 0.19052984565496445,
      "num_tokens": 5590206.0,
      "step": 2485
    },
    {
      "entropy": 5.473781728744507,
      "epoch": 2.6922660897782587,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004647067205398332,
      "loss": 5.3773,
      "mean_token_accuracy": 0.1808069244027138,
      "num_tokens": 5601639.0,
      "step": 2490
    },
    {
      "entropy": 5.437832450866699,
      "epoch": 2.697674418604651,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046447574948147027,
      "loss": 5.3071,
      "mean_token_accuracy": 0.18197673112154006,
      "num_tokens": 5612915.0,
      "step": 2495
    },
    {
      "entropy": 5.569986343383789,
      "epoch": 2.703082747431044,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046424408986779587,
      "loss": 5.4836,
      "mean_token_accuracy": 0.17870566844940186,
      "num_tokens": 5623919.0,
      "step": 2500
    },
    {
      "entropy": 5.550902223587036,
      "epoch": 2.7084910762574363,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004640117425406615,
      "loss": 5.3881,
      "mean_token_accuracy": 0.18566381186246872,
      "num_tokens": 5635334.0,
      "step": 2505
    },
    {
      "entropy": 5.503871917724609,
      "epoch": 2.713899405083829,
      "grad_norm": 1.03125,
      "learning_rate": 0.000463778708344418,
      "loss": 5.3808,
      "mean_token_accuracy": 0.18955522179603576,
      "num_tokens": 5646053.0,
      "step": 2510
    },
    {
      "entropy": 5.507321166992187,
      "epoch": 2.7193077339102216,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046354498812591207,
      "loss": 5.3922,
      "mean_token_accuracy": 0.18845376521348953,
      "num_tokens": 5656445.0,
      "step": 2515
    },
    {
      "entropy": 5.598144817352295,
      "epoch": 2.7247160627366145,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046331058273448364,
      "loss": 5.3921,
      "mean_token_accuracy": 0.1822582095861435,
      "num_tokens": 5667324.0,
      "step": 2520
    },
    {
      "entropy": 5.447207260131836,
      "epoch": 2.730124391563007,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004630754930219624,
      "loss": 5.3905,
      "mean_token_accuracy": 0.18877604305744172,
      "num_tokens": 5677557.0,
      "step": 2525
    },
    {
      "entropy": 5.402798748016357,
      "epoch": 2.7355327203893998,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004628397198426648,
      "loss": 5.2329,
      "mean_token_accuracy": 0.1926985576748848,
      "num_tokens": 5688444.0,
      "step": 2530
    },
    {
      "entropy": 5.5604699611663815,
      "epoch": 2.740941049215792,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046260326405339126,
      "loss": 5.5044,
      "mean_token_accuracy": 0.17508567720651627,
      "num_tokens": 5699194.0,
      "step": 2535
    },
    {
      "entropy": 5.569923686981201,
      "epoch": 2.746349378042185,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004623661265134226,
      "loss": 5.4671,
      "mean_token_accuracy": 0.17520037144422532,
      "num_tokens": 5710831.0,
      "step": 2540
    },
    {
      "entropy": 5.584377336502075,
      "epoch": 2.7517577068685775,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004621283080845172,
      "loss": 5.347,
      "mean_token_accuracy": 0.19738965332508088,
      "num_tokens": 5721781.0,
      "step": 2545
    },
    {
      "entropy": 5.4828346252441404,
      "epoch": 2.7571660356949703,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046188980963090783,
      "loss": 5.3729,
      "mean_token_accuracy": 0.18409281522035598,
      "num_tokens": 5733303.0,
      "step": 2550
    },
    {
      "entropy": 5.446268129348755,
      "epoch": 2.7625743645213627,
      "grad_norm": 1.125,
      "learning_rate": 0.00046165063201929825,
      "loss": 5.3689,
      "mean_token_accuracy": 0.18689175695180893,
      "num_tokens": 5743978.0,
      "step": 2555
    },
    {
      "entropy": 5.46161675453186,
      "epoch": 2.7679826933477556,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046141077611886074,
      "loss": 5.3696,
      "mean_token_accuracy": 0.18578916043043137,
      "num_tokens": 5755465.0,
      "step": 2560
    },
    {
      "entropy": 5.4774675369262695,
      "epoch": 2.773391022174148,
      "grad_norm": 1.0,
      "learning_rate": 0.0004611702428012319,
      "loss": 5.3051,
      "mean_token_accuracy": 0.19155556112527847,
      "num_tokens": 5767132.0,
      "step": 2565
    },
    {
      "entropy": 5.469518041610717,
      "epoch": 2.778799351000541,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046092903294051057,
      "loss": 5.3605,
      "mean_token_accuracy": 0.19294181615114211,
      "num_tokens": 5777954.0,
      "step": 2570
    },
    {
      "entropy": 5.462495136260986,
      "epoch": 2.7842076798269337,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046068714741325385,
      "loss": 5.3367,
      "mean_token_accuracy": 0.1882034182548523,
      "num_tokens": 5789482.0,
      "step": 2575
    },
    {
      "entropy": 5.498905038833618,
      "epoch": 2.789616008653326,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004604445870984743,
      "loss": 5.3336,
      "mean_token_accuracy": 0.18698014914989472,
      "num_tokens": 5800624.0,
      "step": 2580
    },
    {
      "entropy": 5.48365159034729,
      "epoch": 2.7950243374797186,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004602013528776367,
      "loss": 5.3472,
      "mean_token_accuracy": 0.19222134351730347,
      "num_tokens": 5810402.0,
      "step": 2585
    },
    {
      "entropy": 5.55194730758667,
      "epoch": 2.8004326663061114,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004599574456346548,
      "loss": 5.4827,
      "mean_token_accuracy": 0.18214652687311172,
      "num_tokens": 5822596.0,
      "step": 2590
    },
    {
      "entropy": 5.495826387405396,
      "epoch": 2.8058409951325043,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045971286625588796,
      "loss": 5.3741,
      "mean_token_accuracy": 0.18433350175619126,
      "num_tokens": 5834201.0,
      "step": 2595
    },
    {
      "entropy": 5.492666625976563,
      "epoch": 2.8112493239588967,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004594676156301383,
      "loss": 5.3005,
      "mean_token_accuracy": 0.1921772375702858,
      "num_tokens": 5846440.0,
      "step": 2600
    },
    {
      "entropy": 5.438047790527344,
      "epoch": 2.816657652785289,
      "grad_norm": 1.0,
      "learning_rate": 0.00045922169464864725,
      "loss": 5.4765,
      "mean_token_accuracy": 0.1770583361387253,
      "num_tokens": 5858917.0,
      "step": 2605
    },
    {
      "entropy": 5.5323127746582035,
      "epoch": 2.822065981611682,
      "grad_norm": 1.09375,
      "learning_rate": 0.000458975104205092,
      "loss": 5.3066,
      "mean_token_accuracy": 0.19467224478721618,
      "num_tokens": 5869834.0,
      "step": 2610
    },
    {
      "entropy": 5.5151975631713865,
      "epoch": 2.827474310438075,
      "grad_norm": 0.9375,
      "learning_rate": 0.00045872784519558304,
      "loss": 5.4022,
      "mean_token_accuracy": 0.18480263352394105,
      "num_tokens": 5881895.0,
      "step": 2615
    },
    {
      "entropy": 5.438316297531128,
      "epoch": 2.8328826392644673,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004584799185186599,
      "loss": 5.3402,
      "mean_token_accuracy": 0.18830967247486113,
      "num_tokens": 5892929.0,
      "step": 2620
    },
    {
      "entropy": 5.462751626968384,
      "epoch": 2.8382909680908597,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045823132507528903,
      "loss": 5.3173,
      "mean_token_accuracy": 0.1983159974217415,
      "num_tokens": 5903347.0,
      "step": 2625
    },
    {
      "entropy": 5.605246448516846,
      "epoch": 2.8436992969172525,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004579820657688593,
      "loss": 5.4603,
      "mean_token_accuracy": 0.18683298975229262,
      "num_tokens": 5915025.0,
      "step": 2630
    },
    {
      "entropy": 5.528272914886474,
      "epoch": 2.8491076257436454,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004577321415051798,
      "loss": 5.3377,
      "mean_token_accuracy": 0.19072499722242356,
      "num_tokens": 5925953.0,
      "step": 2635
    },
    {
      "entropy": 5.530864620208741,
      "epoch": 2.854515954570038,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045748155319247564,
      "loss": 5.4823,
      "mean_token_accuracy": 0.1815452367067337,
      "num_tokens": 5938589.0,
      "step": 2640
    },
    {
      "entropy": 5.572798442840576,
      "epoch": 2.8599242833964302,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004572303017413856,
      "loss": 5.4122,
      "mean_token_accuracy": 0.17687317728996277,
      "num_tokens": 5949546.0,
      "step": 2645
    },
    {
      "entropy": 5.337789869308471,
      "epoch": 2.865332612222823,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004569783880649578,
      "loss": 5.3594,
      "mean_token_accuracy": 0.18260867148637772,
      "num_tokens": 5960395.0,
      "step": 2650
    },
    {
      "entropy": 5.551058387756347,
      "epoch": 2.870740941049216,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004567258130786474,
      "loss": 5.4024,
      "mean_token_accuracy": 0.18948372900485994,
      "num_tokens": 5969642.0,
      "step": 2655
    },
    {
      "entropy": 5.4566559314727785,
      "epoch": 2.8761492698756084,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045647257770031215,
      "loss": 5.3139,
      "mean_token_accuracy": 0.18019493967294692,
      "num_tokens": 5979001.0,
      "step": 2660
    },
    {
      "entropy": 5.621964836120606,
      "epoch": 2.8815575987020012,
      "grad_norm": 1.0,
      "learning_rate": 0.00045621868285021023,
      "loss": 5.5441,
      "mean_token_accuracy": 0.17837482690811157,
      "num_tokens": 5991511.0,
      "step": 2665
    },
    {
      "entropy": 5.444990110397339,
      "epoch": 2.8869659275283936,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045596412945099615,
      "loss": 5.2747,
      "mean_token_accuracy": 0.1955416664481163,
      "num_tokens": 6001902.0,
      "step": 2670
    },
    {
      "entropy": 5.40801887512207,
      "epoch": 2.8923742563547865,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045570891842771727,
      "loss": 5.3791,
      "mean_token_accuracy": 0.191572804749012,
      "num_tokens": 6012220.0,
      "step": 2675
    },
    {
      "entropy": 5.503115224838257,
      "epoch": 2.897782585181179,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045545305070781135,
      "loss": 5.3874,
      "mean_token_accuracy": 0.18666007816791536,
      "num_tokens": 6023795.0,
      "step": 2680
    },
    {
      "entropy": 5.490363645553589,
      "epoch": 2.903190914007572,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045519652722110206,
      "loss": 5.34,
      "mean_token_accuracy": 0.1851412460207939,
      "num_tokens": 6034335.0,
      "step": 2685
    },
    {
      "entropy": 5.5544792175292965,
      "epoch": 2.908599242833964,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045493934889979655,
      "loss": 5.4185,
      "mean_token_accuracy": 0.18491135239601136,
      "num_tokens": 6045435.0,
      "step": 2690
    },
    {
      "entropy": 5.475643587112427,
      "epoch": 2.914007571660357,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045468151667848116,
      "loss": 5.444,
      "mean_token_accuracy": 0.18254784345626832,
      "num_tokens": 6057130.0,
      "step": 2695
    },
    {
      "entropy": 5.632959985733033,
      "epoch": 2.9194159004867495,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000454423031494119,
      "loss": 5.4365,
      "mean_token_accuracy": 0.1868360236287117,
      "num_tokens": 6068382.0,
      "step": 2700
    },
    {
      "entropy": 5.513478803634643,
      "epoch": 2.9248242293131423,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004541638942860458,
      "loss": 5.4198,
      "mean_token_accuracy": 0.18837159723043442,
      "num_tokens": 6078958.0,
      "step": 2705
    },
    {
      "entropy": 5.526330518722534,
      "epoch": 2.9302325581395348,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004539041059959667,
      "loss": 5.4401,
      "mean_token_accuracy": 0.18759401738643647,
      "num_tokens": 6091054.0,
      "step": 2710
    },
    {
      "entropy": 5.538228273391724,
      "epoch": 2.9356408869659276,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045364366756795295,
      "loss": 5.3236,
      "mean_token_accuracy": 0.18440162241458893,
      "num_tokens": 6102371.0,
      "step": 2715
    },
    {
      "entropy": 5.423747777938843,
      "epoch": 2.94104921579232,
      "grad_norm": 1.125,
      "learning_rate": 0.00045338257994843845,
      "loss": 5.3537,
      "mean_token_accuracy": 0.19029029160737992,
      "num_tokens": 6113468.0,
      "step": 2720
    },
    {
      "entropy": 5.476720857620239,
      "epoch": 2.946457544618713,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004531208440862162,
      "loss": 5.4001,
      "mean_token_accuracy": 0.1901046320796013,
      "num_tokens": 6125205.0,
      "step": 2725
    },
    {
      "entropy": 5.478125047683716,
      "epoch": 2.9518658734451053,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045285846093243485,
      "loss": 5.3179,
      "mean_token_accuracy": 0.18634466975927352,
      "num_tokens": 6134831.0,
      "step": 2730
    },
    {
      "entropy": 5.560326480865479,
      "epoch": 2.957274202271498,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045259543144059557,
      "loss": 5.4177,
      "mean_token_accuracy": 0.18014927655458451,
      "num_tokens": 6145927.0,
      "step": 2735
    },
    {
      "entropy": 5.476049375534058,
      "epoch": 2.9626825310978906,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004523317565665479,
      "loss": 5.4213,
      "mean_token_accuracy": 0.18212687522172927,
      "num_tokens": 6156054.0,
      "step": 2740
    },
    {
      "entropy": 5.437103366851806,
      "epoch": 2.9680908599242835,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045206743726848716,
      "loss": 5.3387,
      "mean_token_accuracy": 0.18745884001255037,
      "num_tokens": 6166157.0,
      "step": 2745
    },
    {
      "entropy": 5.491754388809204,
      "epoch": 2.973499188750676,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004518024745069502,
      "loss": 5.366,
      "mean_token_accuracy": 0.18674905002117156,
      "num_tokens": 6178462.0,
      "step": 2750
    },
    {
      "entropy": 5.453402280807495,
      "epoch": 2.9789075175770687,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004515368692448123,
      "loss": 5.3232,
      "mean_token_accuracy": 0.18855093717575072,
      "num_tokens": 6189018.0,
      "step": 2755
    },
    {
      "entropy": 5.590845823287964,
      "epoch": 2.9843158464034616,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045127062244728353,
      "loss": 5.4735,
      "mean_token_accuracy": 0.18126190304756165,
      "num_tokens": 6199911.0,
      "step": 2760
    },
    {
      "entropy": 5.487865018844604,
      "epoch": 2.989724175229854,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004510037350819055,
      "loss": 5.3803,
      "mean_token_accuracy": 0.18124324083328247,
      "num_tokens": 6211822.0,
      "step": 2765
    },
    {
      "entropy": 5.42392144203186,
      "epoch": 2.9951325040562464,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045073620811854744,
      "loss": 5.2865,
      "mean_token_accuracy": 0.1973293974995613,
      "num_tokens": 6222365.0,
      "step": 2770
    },
    {
      "entropy": 5.446313963996039,
      "epoch": 3.0,
      "grad_norm": 2.078125,
      "learning_rate": 0.00045046804252940294,
      "loss": 5.3448,
      "mean_token_accuracy": 0.19321518474155003,
      "num_tokens": 6230667.0,
      "step": 2775
    },
    {
      "entropy": 5.500533962249756,
      "epoch": 3.005408328826393,
      "grad_norm": 1.0,
      "learning_rate": 0.00045019923928898644,
      "loss": 5.0612,
      "mean_token_accuracy": 0.19457762837409973,
      "num_tokens": 6241678.0,
      "step": 2780
    },
    {
      "entropy": 5.321163845062256,
      "epoch": 3.0108166576527853,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004499297993741294,
      "loss": 5.0075,
      "mean_token_accuracy": 0.19694564640522003,
      "num_tokens": 6253541.0,
      "step": 2785
    },
    {
      "entropy": 5.432270860671997,
      "epoch": 3.016224986479178,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004496597237639773,
      "loss": 5.0,
      "mean_token_accuracy": 0.19926079362630844,
      "num_tokens": 6265590.0,
      "step": 2790
    },
    {
      "entropy": 5.444063663482666,
      "epoch": 3.0216333153055706,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044938901343998536,
      "loss": 5.0456,
      "mean_token_accuracy": 0.20204200297594072,
      "num_tokens": 6276129.0,
      "step": 2795
    },
    {
      "entropy": 5.414129114151001,
      "epoch": 3.0270416441319634,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004491176693859156,
      "loss": 4.9712,
      "mean_token_accuracy": 0.2101418912410736,
      "num_tokens": 6286770.0,
      "step": 2800
    },
    {
      "entropy": 5.4351158618927,
      "epoch": 3.032449972958356,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004488456925878329,
      "loss": 5.0645,
      "mean_token_accuracy": 0.19975790232419968,
      "num_tokens": 6298276.0,
      "step": 2805
    },
    {
      "entropy": 5.424533796310425,
      "epoch": 3.0378583017847487,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044857308403410163,
      "loss": 5.0436,
      "mean_token_accuracy": 0.20333951115608215,
      "num_tokens": 6308856.0,
      "step": 2810
    },
    {
      "entropy": 5.381372022628784,
      "epoch": 3.043266630611141,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044829984471538183,
      "loss": 5.0321,
      "mean_token_accuracy": 0.2041509985923767,
      "num_tokens": 6320386.0,
      "step": 2815
    },
    {
      "entropy": 5.4244366645812985,
      "epoch": 3.048674959437534,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004480259756246258,
      "loss": 4.9969,
      "mean_token_accuracy": 0.19859097748994828,
      "num_tokens": 6332164.0,
      "step": 2820
    },
    {
      "entropy": 5.294694423675537,
      "epoch": 3.0540832882639264,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044775147775707465,
      "loss": 5.0078,
      "mean_token_accuracy": 0.20580752789974213,
      "num_tokens": 6343828.0,
      "step": 2825
    },
    {
      "entropy": 5.371400594711304,
      "epoch": 3.0594916170903192,
      "grad_norm": 1.078125,
      "learning_rate": 0.000447476352110254,
      "loss": 5.0844,
      "mean_token_accuracy": 0.19515798091888428,
      "num_tokens": 6355366.0,
      "step": 2830
    },
    {
      "entropy": 5.433107185363769,
      "epoch": 3.0648999459167117,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004472005996839712,
      "loss": 5.0519,
      "mean_token_accuracy": 0.21033187359571456,
      "num_tokens": 6365270.0,
      "step": 2835
    },
    {
      "entropy": 5.351367807388305,
      "epoch": 3.0703082747431045,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004469242214803112,
      "loss": 5.0275,
      "mean_token_accuracy": 0.21128651946783067,
      "num_tokens": 6376180.0,
      "step": 2840
    },
    {
      "entropy": 5.407835197448731,
      "epoch": 3.075716603569497,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004466472185036331,
      "loss": 4.9536,
      "mean_token_accuracy": 0.2117062970995903,
      "num_tokens": 6387068.0,
      "step": 2845
    },
    {
      "entropy": 5.414833641052246,
      "epoch": 3.08112493239589,
      "grad_norm": 1.125,
      "learning_rate": 0.0004463695917605663,
      "loss": 5.021,
      "mean_token_accuracy": 0.20732709020376205,
      "num_tokens": 6397540.0,
      "step": 2850
    },
    {
      "entropy": 5.340751600265503,
      "epoch": 3.086533261222282,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000446091342260007,
      "loss": 5.1168,
      "mean_token_accuracy": 0.19058479368686676,
      "num_tokens": 6410039.0,
      "step": 2855
    },
    {
      "entropy": 5.416444110870361,
      "epoch": 3.091941590048675,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004458124710131145,
      "loss": 4.9762,
      "mean_token_accuracy": 0.20852354317903518,
      "num_tokens": 6420964.0,
      "step": 2860
    },
    {
      "entropy": 5.380699825286865,
      "epoch": 3.0973499188750675,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044553297903330766,
      "loss": 5.1075,
      "mean_token_accuracy": 0.19606724083423616,
      "num_tokens": 6432752.0,
      "step": 2865
    },
    {
      "entropy": 5.475742959976197,
      "epoch": 3.1027582477014604,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044525286733626085,
      "loss": 5.0334,
      "mean_token_accuracy": 0.2051415890455246,
      "num_tokens": 6443899.0,
      "step": 2870
    },
    {
      "entropy": 5.429310083389282,
      "epoch": 3.1081665765278528,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004449721369399008,
      "loss": 5.1134,
      "mean_token_accuracy": 0.1922602593898773,
      "num_tokens": 6454133.0,
      "step": 2875
    },
    {
      "entropy": 5.352643203735352,
      "epoch": 3.1135749053542456,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044469078886440227,
      "loss": 5.0354,
      "mean_token_accuracy": 0.19823043793439865,
      "num_tokens": 6467574.0,
      "step": 2880
    },
    {
      "entropy": 5.444091510772705,
      "epoch": 3.118983234180638,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044440882413218484,
      "loss": 5.074,
      "mean_token_accuracy": 0.19586091190576554,
      "num_tokens": 6479399.0,
      "step": 2885
    },
    {
      "entropy": 5.340216112136841,
      "epoch": 3.124391563007031,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044412624376790905,
      "loss": 4.8886,
      "mean_token_accuracy": 0.21345536708831786,
      "num_tokens": 6489590.0,
      "step": 2890
    },
    {
      "entropy": 5.401822757720947,
      "epoch": 3.1297998918334233,
      "grad_norm": 1.125,
      "learning_rate": 0.0004438430487984726,
      "loss": 5.0282,
      "mean_token_accuracy": 0.20397695451974868,
      "num_tokens": 6500689.0,
      "step": 2895
    },
    {
      "entropy": 5.340320634841919,
      "epoch": 3.135208220659816,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004435592402530066,
      "loss": 5.0095,
      "mean_token_accuracy": 0.20189974159002305,
      "num_tokens": 6511190.0,
      "step": 2900
    },
    {
      "entropy": 5.328754806518555,
      "epoch": 3.1406165494862086,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000443274819162872,
      "loss": 4.9895,
      "mean_token_accuracy": 0.20199992805719375,
      "num_tokens": 6522288.0,
      "step": 2905
    },
    {
      "entropy": 5.396235084533691,
      "epoch": 3.1460248783126015,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004429897865616557,
      "loss": 5.0095,
      "mean_token_accuracy": 0.2032381221652031,
      "num_tokens": 6534391.0,
      "step": 2910
    },
    {
      "entropy": 5.3877345561981205,
      "epoch": 3.151433207138994,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004427041434851668,
      "loss": 4.9899,
      "mean_token_accuracy": 0.207269649207592,
      "num_tokens": 6546191.0,
      "step": 2915
    },
    {
      "entropy": 5.385879850387573,
      "epoch": 3.1568415359653867,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044241789097143293,
      "loss": 5.1539,
      "mean_token_accuracy": 0.18838376700878143,
      "num_tokens": 6558118.0,
      "step": 2920
    },
    {
      "entropy": 5.319383335113526,
      "epoch": 3.162249864791779,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044213103006069635,
      "loss": 5.0371,
      "mean_token_accuracy": 0.20676176995038986,
      "num_tokens": 6568636.0,
      "step": 2925
    },
    {
      "entropy": 5.447977113723755,
      "epoch": 3.167658193618172,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044184356179541035,
      "loss": 5.0624,
      "mean_token_accuracy": 0.20269704908132552,
      "num_tokens": 6579710.0,
      "step": 2930
    },
    {
      "entropy": 5.385798645019531,
      "epoch": 3.1730665224445644,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004415554872202352,
      "loss": 5.0467,
      "mean_token_accuracy": 0.20548543632030486,
      "num_tokens": 6590574.0,
      "step": 2935
    },
    {
      "entropy": 5.405917978286743,
      "epoch": 3.1784748512709573,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044126680738203466,
      "loss": 5.1356,
      "mean_token_accuracy": 0.19349176585674285,
      "num_tokens": 6601637.0,
      "step": 2940
    },
    {
      "entropy": 5.360622119903565,
      "epoch": 3.1838831800973497,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004409775233298718,
      "loss": 4.9065,
      "mean_token_accuracy": 0.21671828478574753,
      "num_tokens": 6611958.0,
      "step": 2945
    },
    {
      "entropy": 5.419522380828857,
      "epoch": 3.1892915089237426,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004406876361150056,
      "loss": 5.1699,
      "mean_token_accuracy": 0.1909557670354843,
      "num_tokens": 6623228.0,
      "step": 2950
    },
    {
      "entropy": 5.399460935592652,
      "epoch": 3.194699837750135,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004403971467908869,
      "loss": 5.0989,
      "mean_token_accuracy": 0.199129718542099,
      "num_tokens": 6633437.0,
      "step": 2955
    },
    {
      "entropy": 5.443213653564453,
      "epoch": 3.200108166576528,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004401060564131545,
      "loss": 5.0749,
      "mean_token_accuracy": 0.19572130888700484,
      "num_tokens": 6644605.0,
      "step": 2960
    },
    {
      "entropy": 5.34327507019043,
      "epoch": 3.2055164954029207,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004398143660396315,
      "loss": 5.1341,
      "mean_token_accuracy": 0.19737862199544906,
      "num_tokens": 6656457.0,
      "step": 2965
    },
    {
      "entropy": 5.454725503921509,
      "epoch": 3.210924824229313,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004395220767303215,
      "loss": 5.0676,
      "mean_token_accuracy": 0.20159655213356018,
      "num_tokens": 6667741.0,
      "step": 2970
    },
    {
      "entropy": 5.313960456848145,
      "epoch": 3.2163331530557056,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004392291895474045,
      "loss": 5.0565,
      "mean_token_accuracy": 0.20246104747056962,
      "num_tokens": 6679111.0,
      "step": 2975
    },
    {
      "entropy": 5.359478759765625,
      "epoch": 3.2217414818820984,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004389357055552331,
      "loss": 4.9841,
      "mean_token_accuracy": 0.2107232466340065,
      "num_tokens": 6689960.0,
      "step": 2980
    },
    {
      "entropy": 5.3306825160980225,
      "epoch": 3.2271498107084913,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043864162582032897,
      "loss": 5.0199,
      "mean_token_accuracy": 0.19566542357206346,
      "num_tokens": 6701163.0,
      "step": 2985
    },
    {
      "entropy": 5.3011407375335695,
      "epoch": 3.2325581395348837,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004383469514113784,
      "loss": 5.0126,
      "mean_token_accuracy": 0.20581124275922774,
      "num_tokens": 6712466.0,
      "step": 2990
    },
    {
      "entropy": 5.294154405593872,
      "epoch": 3.2379664683612766,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004380516833992289,
      "loss": 5.1264,
      "mean_token_accuracy": 0.19787818789482117,
      "num_tokens": 6724772.0,
      "step": 2995
    },
    {
      "entropy": 5.3787847518920895,
      "epoch": 3.243374797187669,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000437755822856885,
      "loss": 5.02,
      "mean_token_accuracy": 0.20379386097192764,
      "num_tokens": 6733953.0,
      "step": 3000
    },
    {
      "epoch": 3.243374797187669,
      "eval_entropy": 5.224143935333599,
      "eval_loss": 5.921028137207031,
      "eval_mean_token_accuracy": 0.17506341190839356,
      "eval_num_tokens": 6733953.0,
      "eval_runtime": 2.5049,
      "eval_samples_per_second": 1403.274,
      "eval_steps_per_second": 175.659,
      "step": 3000
    }
  ],
  "logging_steps": 5,
  "max_steps": 9240,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.0572520679424e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}