{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.1287166945552838,
  "eval_steps": 500,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.742555713653564,
      "epoch": 0.00032179173638820956,
      "grad_norm": 4.59375,
      "learning_rate": 2e-06,
      "loss": 10.7932,
      "mean_token_accuracy": 9.85221704468131e-05,
      "num_tokens": 11506.0,
      "step": 5
    },
    {
      "entropy": 10.742585468292237,
      "epoch": 0.0006435834727764191,
      "grad_norm": 5.0,
      "learning_rate": 4.5e-06,
      "loss": 10.7664,
      "mean_token_accuracy": 7.942811935208738e-05,
      "num_tokens": 21973.0,
      "step": 10
    },
    {
      "entropy": 10.742585182189941,
      "epoch": 0.0009653752091646286,
      "grad_norm": 4.6875,
      "learning_rate": 7e-06,
      "loss": 10.7473,
      "mean_token_accuracy": 0.0,
      "num_tokens": 32763.0,
      "step": 15
    },
    {
      "entropy": 10.742647171020508,
      "epoch": 0.0012871669455528382,
      "grad_norm": 4.5,
      "learning_rate": 9.5e-06,
      "loss": 10.716,
      "mean_token_accuracy": 0.00042943707667291163,
      "num_tokens": 44354.0,
      "step": 20
    },
    {
      "entropy": 10.742597675323486,
      "epoch": 0.0016089586819410479,
      "grad_norm": 4.375,
      "learning_rate": 1.2e-05,
      "loss": 10.6219,
      "mean_token_accuracy": 0.0008890067110769451,
      "num_tokens": 56534.0,
      "step": 25
    },
    {
      "entropy": 10.74241762161255,
      "epoch": 0.0019307504183292573,
      "grad_norm": 3.859375,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 10.4911,
      "mean_token_accuracy": 0.009375615860335529,
      "num_tokens": 67106.0,
      "step": 30
    },
    {
      "entropy": 10.741784954071045,
      "epoch": 0.002252542154717467,
      "grad_norm": 3.265625,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.3831,
      "mean_token_accuracy": 0.021833056677132846,
      "num_tokens": 78801.0,
      "step": 35
    },
    {
      "entropy": 10.74044713973999,
      "epoch": 0.0025743338911056765,
      "grad_norm": 2.53125,
      "learning_rate": 1.95e-05,
      "loss": 10.2405,
      "mean_token_accuracy": 0.028899907879531384,
      "num_tokens": 90390.0,
      "step": 40
    },
    {
      "entropy": 10.738579750061035,
      "epoch": 0.002896125627493886,
      "grad_norm": 2.265625,
      "learning_rate": 2.2e-05,
      "loss": 10.1128,
      "mean_token_accuracy": 0.036285107396543025,
      "num_tokens": 101652.0,
      "step": 45
    },
    {
      "entropy": 10.737388038635254,
      "epoch": 0.0032179173638820957,
      "grad_norm": 2.0625,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 10.0384,
      "mean_token_accuracy": 0.03730290364474058,
      "num_tokens": 112068.0,
      "step": 50
    },
    {
      "entropy": 10.737480354309081,
      "epoch": 0.003539709100270305,
      "grad_norm": 1.90625,
      "learning_rate": 2.7e-05,
      "loss": 10.0098,
      "mean_token_accuracy": 0.04103992246091366,
      "num_tokens": 123500.0,
      "step": 55
    },
    {
      "entropy": 10.736884212493896,
      "epoch": 0.0038615008366585145,
      "grad_norm": 1.9296875,
      "learning_rate": 2.95e-05,
      "loss": 9.9544,
      "mean_token_accuracy": 0.0359757786616683,
      "num_tokens": 134456.0,
      "step": 60
    },
    {
      "entropy": 10.7358078956604,
      "epoch": 0.004183292573046725,
      "grad_norm": 1.765625,
      "learning_rate": 3.2e-05,
      "loss": 9.9017,
      "mean_token_accuracy": 0.03450035620480776,
      "num_tokens": 146299.0,
      "step": 65
    },
    {
      "entropy": 10.734182739257813,
      "epoch": 0.004505084309434934,
      "grad_norm": 1.9375,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.8076,
      "mean_token_accuracy": 0.037839466519653794,
      "num_tokens": 157494.0,
      "step": 70
    },
    {
      "entropy": 10.73177366256714,
      "epoch": 0.004826876045823143,
      "grad_norm": 1.7734375,
      "learning_rate": 3.7e-05,
      "loss": 9.7775,
      "mean_token_accuracy": 0.03611580468714237,
      "num_tokens": 168226.0,
      "step": 75
    },
    {
      "entropy": 10.728845882415772,
      "epoch": 0.005148667782211353,
      "grad_norm": 1.9296875,
      "learning_rate": 3.95e-05,
      "loss": 9.6556,
      "mean_token_accuracy": 0.04190883524715901,
      "num_tokens": 179311.0,
      "step": 80
    },
    {
      "entropy": 10.725195121765136,
      "epoch": 0.005470459518599562,
      "grad_norm": 1.7421875,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.6267,
      "mean_token_accuracy": 0.043489571660757065,
      "num_tokens": 191513.0,
      "step": 85
    },
    {
      "entropy": 10.719332885742187,
      "epoch": 0.005792251254987772,
      "grad_norm": 1.8828125,
      "learning_rate": 4.45e-05,
      "loss": 9.5397,
      "mean_token_accuracy": 0.046276621893048285,
      "num_tokens": 204631.0,
      "step": 90
    },
    {
      "entropy": 10.710766220092774,
      "epoch": 0.006114042991375981,
      "grad_norm": 1.7578125,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 9.4486,
      "mean_token_accuracy": 0.0460438147187233,
      "num_tokens": 215901.0,
      "step": 95
    },
    {
      "entropy": 10.69977626800537,
      "epoch": 0.0064358347277641914,
      "grad_norm": 1.8203125,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 9.4013,
      "mean_token_accuracy": 0.044720832630991934,
      "num_tokens": 228928.0,
      "step": 100
    },
    {
      "entropy": 10.685317134857177,
      "epoch": 0.006757626464152401,
      "grad_norm": 1.765625,
      "learning_rate": 5.2e-05,
      "loss": 9.3138,
      "mean_token_accuracy": 0.044154060259461406,
      "num_tokens": 241543.0,
      "step": 105
    },
    {
      "entropy": 10.66606101989746,
      "epoch": 0.00707941820054061,
      "grad_norm": 1.8203125,
      "learning_rate": 5.45e-05,
      "loss": 9.1466,
      "mean_token_accuracy": 0.046302244812250135,
      "num_tokens": 253267.0,
      "step": 110
    },
    {
      "entropy": 10.636186695098877,
      "epoch": 0.00740120993692882,
      "grad_norm": 1.78125,
      "learning_rate": 5.7e-05,
      "loss": 9.0141,
      "mean_token_accuracy": 0.05802988223731518,
      "num_tokens": 264760.0,
      "step": 115
    },
    {
      "entropy": 10.5947696685791,
      "epoch": 0.007723001673317029,
      "grad_norm": 1.75,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.9075,
      "mean_token_accuracy": 0.051640312373638156,
      "num_tokens": 275080.0,
      "step": 120
    },
    {
      "entropy": 10.542396450042725,
      "epoch": 0.008044793409705238,
      "grad_norm": 1.7109375,
      "learning_rate": 6.2e-05,
      "loss": 8.7919,
      "mean_token_accuracy": 0.05795170105993748,
      "num_tokens": 287114.0,
      "step": 125
    },
    {
      "entropy": 10.490421962738036,
      "epoch": 0.00836658514609345,
      "grad_norm": 1.6796875,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.6367,
      "mean_token_accuracy": 0.06025359183549881,
      "num_tokens": 298732.0,
      "step": 130
    },
    {
      "entropy": 10.417353057861328,
      "epoch": 0.008688376882481658,
      "grad_norm": 1.65625,
      "learning_rate": 6.7e-05,
      "loss": 8.4641,
      "mean_token_accuracy": 0.06998355574905872,
      "num_tokens": 310227.0,
      "step": 135
    },
    {
      "entropy": 10.32230224609375,
      "epoch": 0.009010168618869868,
      "grad_norm": 1.4765625,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.3943,
      "mean_token_accuracy": 0.06697470992803574,
      "num_tokens": 322724.0,
      "step": 140
    },
    {
      "entropy": 10.228569889068604,
      "epoch": 0.009331960355258077,
      "grad_norm": 1.4375,
      "learning_rate": 7.2e-05,
      "loss": 8.3137,
      "mean_token_accuracy": 0.06024596691131592,
      "num_tokens": 335638.0,
      "step": 145
    },
    {
      "entropy": 10.116732311248779,
      "epoch": 0.009653752091646286,
      "grad_norm": 1.453125,
      "learning_rate": 7.45e-05,
      "loss": 8.1548,
      "mean_token_accuracy": 0.07251053638756275,
      "num_tokens": 346751.0,
      "step": 150
    },
    {
      "entropy": 9.953464126586914,
      "epoch": 0.009975543828034497,
      "grad_norm": 1.3359375,
      "learning_rate": 7.7e-05,
      "loss": 8.0894,
      "mean_token_accuracy": 0.07281172573566437,
      "num_tokens": 357809.0,
      "step": 155
    },
    {
      "entropy": 9.77461576461792,
      "epoch": 0.010297335564422706,
      "grad_norm": 1.21875,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.9142,
      "mean_token_accuracy": 0.06864561475813388,
      "num_tokens": 369631.0,
      "step": 160
    },
    {
      "entropy": 9.603110790252686,
      "epoch": 0.010619127300810915,
      "grad_norm": 1.0390625,
      "learning_rate": 8.2e-05,
      "loss": 7.9575,
      "mean_token_accuracy": 0.06908577904105187,
      "num_tokens": 380904.0,
      "step": 165
    },
    {
      "entropy": 9.377412414550781,
      "epoch": 0.010940919037199124,
      "grad_norm": 1.125,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.7359,
      "mean_token_accuracy": 0.07545260861515998,
      "num_tokens": 392898.0,
      "step": 170
    },
    {
      "entropy": 9.168265438079834,
      "epoch": 0.011262710773587334,
      "grad_norm": 1.125,
      "learning_rate": 8.7e-05,
      "loss": 7.6702,
      "mean_token_accuracy": 0.07459548935294151,
      "num_tokens": 404285.0,
      "step": 175
    },
    {
      "entropy": 8.926816463470459,
      "epoch": 0.011584502509975544,
      "grad_norm": 1.0,
      "learning_rate": 8.95e-05,
      "loss": 7.5717,
      "mean_token_accuracy": 0.07761912979185581,
      "num_tokens": 416443.0,
      "step": 180
    },
    {
      "entropy": 8.745314025878907,
      "epoch": 0.011906294246363754,
      "grad_norm": 0.96875,
      "learning_rate": 9.2e-05,
      "loss": 7.6426,
      "mean_token_accuracy": 0.0765433344990015,
      "num_tokens": 426558.0,
      "step": 185
    },
    {
      "entropy": 8.587053394317627,
      "epoch": 0.012228085982751963,
      "grad_norm": 1.1796875,
      "learning_rate": 9.45e-05,
      "loss": 7.6648,
      "mean_token_accuracy": 0.07643079459667206,
      "num_tokens": 437794.0,
      "step": 190
    },
    {
      "entropy": 8.517185878753661,
      "epoch": 0.012549877719140172,
      "grad_norm": 1.0546875,
      "learning_rate": 9.7e-05,
      "loss": 7.4958,
      "mean_token_accuracy": 0.07788795456290246,
      "num_tokens": 448682.0,
      "step": 195
    },
    {
      "entropy": 8.376048374176026,
      "epoch": 0.012871669455528383,
      "grad_norm": 1.1640625,
      "learning_rate": 9.95e-05,
      "loss": 7.417,
      "mean_token_accuracy": 0.08175441175699234,
      "num_tokens": 459699.0,
      "step": 200
    },
    {
      "entropy": 8.276328945159912,
      "epoch": 0.013193461191916592,
      "grad_norm": 1.125,
      "learning_rate": 0.000102,
      "loss": 7.3537,
      "mean_token_accuracy": 0.08016520962119103,
      "num_tokens": 472857.0,
      "step": 205
    },
    {
      "entropy": 8.274100303649902,
      "epoch": 0.013515252928304801,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.4194,
      "mean_token_accuracy": 0.08162855952978135,
      "num_tokens": 484983.0,
      "step": 210
    },
    {
      "entropy": 8.180224132537841,
      "epoch": 0.01383704466469301,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000107,
      "loss": 7.3626,
      "mean_token_accuracy": 0.08450362831354141,
      "num_tokens": 495973.0,
      "step": 215
    },
    {
      "entropy": 8.187409973144531,
      "epoch": 0.01415883640108122,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001095,
      "loss": 7.2816,
      "mean_token_accuracy": 0.090892593562603,
      "num_tokens": 507017.0,
      "step": 220
    },
    {
      "entropy": 8.120441341400147,
      "epoch": 0.01448062813746943,
      "grad_norm": 1.1875,
      "learning_rate": 0.000112,
      "loss": 7.36,
      "mean_token_accuracy": 0.08900276795029641,
      "num_tokens": 517522.0,
      "step": 225
    },
    {
      "entropy": 8.129059314727783,
      "epoch": 0.01480241987385764,
      "grad_norm": 1.015625,
      "learning_rate": 0.0001145,
      "loss": 7.3701,
      "mean_token_accuracy": 0.09131013453006745,
      "num_tokens": 528240.0,
      "step": 230
    },
    {
      "entropy": 7.994166135787964,
      "epoch": 0.015124211610245849,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.328,
      "mean_token_accuracy": 0.0881452701985836,
      "num_tokens": 538875.0,
      "step": 235
    },
    {
      "entropy": 8.10927906036377,
      "epoch": 0.015446003346634058,
      "grad_norm": 1.109375,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.3722,
      "mean_token_accuracy": 0.08998035714030266,
      "num_tokens": 549386.0,
      "step": 240
    },
    {
      "entropy": 8.068589448928833,
      "epoch": 0.015767795083022267,
      "grad_norm": 1.234375,
      "learning_rate": 0.000122,
      "loss": 7.3029,
      "mean_token_accuracy": 0.08936483487486839,
      "num_tokens": 559643.0,
      "step": 245
    },
    {
      "entropy": 8.049430751800537,
      "epoch": 0.016089586819410476,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0001245,
      "loss": 7.2267,
      "mean_token_accuracy": 0.09506414234638214,
      "num_tokens": 570932.0,
      "step": 250
    },
    {
      "entropy": 7.93134560585022,
      "epoch": 0.016411378555798686,
      "grad_norm": 1.296875,
      "learning_rate": 0.000127,
      "loss": 7.2019,
      "mean_token_accuracy": 0.09725837931036949,
      "num_tokens": 580258.0,
      "step": 255
    },
    {
      "entropy": 7.893705558776856,
      "epoch": 0.0167331702921869,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001295,
      "loss": 7.1337,
      "mean_token_accuracy": 0.09558385238051414,
      "num_tokens": 590622.0,
      "step": 260
    },
    {
      "entropy": 7.884620428085327,
      "epoch": 0.017054962028575107,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000132,
      "loss": 7.1827,
      "mean_token_accuracy": 0.09120650067925454,
      "num_tokens": 602202.0,
      "step": 265
    },
    {
      "entropy": 7.823669528961181,
      "epoch": 0.017376753764963317,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.1309,
      "mean_token_accuracy": 0.0979801706969738,
      "num_tokens": 613864.0,
      "step": 270
    },
    {
      "entropy": 7.87017560005188,
      "epoch": 0.017698545501351526,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.2524,
      "mean_token_accuracy": 0.09466890171170235,
      "num_tokens": 626884.0,
      "step": 275
    },
    {
      "entropy": 7.857308626174927,
      "epoch": 0.018020337237739735,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001395,
      "loss": 7.1612,
      "mean_token_accuracy": 0.09996400326490402,
      "num_tokens": 638418.0,
      "step": 280
    },
    {
      "entropy": 7.856085300445557,
      "epoch": 0.018342128974127944,
      "grad_norm": 1.125,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.1939,
      "mean_token_accuracy": 0.09103256016969681,
      "num_tokens": 648376.0,
      "step": 285
    },
    {
      "entropy": 7.9166460037231445,
      "epoch": 0.018663920710516153,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001445,
      "loss": 7.2464,
      "mean_token_accuracy": 0.09389530643820762,
      "num_tokens": 659268.0,
      "step": 290
    },
    {
      "entropy": 7.788627672195434,
      "epoch": 0.018985712446904363,
      "grad_norm": 1.03125,
      "learning_rate": 0.000147,
      "loss": 7.0223,
      "mean_token_accuracy": 0.10249627456068992,
      "num_tokens": 669736.0,
      "step": 295
    },
    {
      "entropy": 7.768741464614868,
      "epoch": 0.01930750418329257,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0001495,
      "loss": 7.1198,
      "mean_token_accuracy": 0.0995169997215271,
      "num_tokens": 681571.0,
      "step": 300
    },
    {
      "entropy": 7.723494529724121,
      "epoch": 0.01962929591968078,
      "grad_norm": 0.9375,
      "learning_rate": 0.000152,
      "loss": 7.0189,
      "mean_token_accuracy": 0.10438807159662247,
      "num_tokens": 692816.0,
      "step": 305
    },
    {
      "entropy": 7.758736324310303,
      "epoch": 0.019951087656068994,
      "grad_norm": 1.078125,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.1546,
      "mean_token_accuracy": 0.10238918289542198,
      "num_tokens": 705089.0,
      "step": 310
    },
    {
      "entropy": 7.666284465789795,
      "epoch": 0.020272879392457203,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000157,
      "loss": 7.0252,
      "mean_token_accuracy": 0.10788851976394653,
      "num_tokens": 716874.0,
      "step": 315
    },
    {
      "entropy": 7.531011438369751,
      "epoch": 0.020594671128845412,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001595,
      "loss": 6.9807,
      "mean_token_accuracy": 0.10204816684126854,
      "num_tokens": 728981.0,
      "step": 320
    },
    {
      "entropy": 7.782111930847168,
      "epoch": 0.02091646286523362,
      "grad_norm": 1.109375,
      "learning_rate": 0.000162,
      "loss": 7.0871,
      "mean_token_accuracy": 0.10574771091341972,
      "num_tokens": 739619.0,
      "step": 325
    },
    {
      "entropy": 7.599967575073242,
      "epoch": 0.02123825460162183,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.0537,
      "mean_token_accuracy": 0.10123220011591912,
      "num_tokens": 751363.0,
      "step": 330
    },
    {
      "entropy": 7.667736959457398,
      "epoch": 0.02156004633801004,
      "grad_norm": 1.15625,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.0722,
      "mean_token_accuracy": 0.10424386709928513,
      "num_tokens": 762953.0,
      "step": 335
    },
    {
      "entropy": 7.489533710479736,
      "epoch": 0.02188183807439825,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.9886,
      "mean_token_accuracy": 0.10709020495414734,
      "num_tokens": 775237.0,
      "step": 340
    },
    {
      "entropy": 7.635052871704102,
      "epoch": 0.022203629810786458,
      "grad_norm": 1.109375,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.9133,
      "mean_token_accuracy": 0.11272823959589004,
      "num_tokens": 786262.0,
      "step": 345
    },
    {
      "entropy": 7.458449077606201,
      "epoch": 0.022525421547174667,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.9036,
      "mean_token_accuracy": 0.11083822324872017,
      "num_tokens": 797201.0,
      "step": 350
    },
    {
      "entropy": 7.504961967468262,
      "epoch": 0.02284721328356288,
      "grad_norm": 1.09375,
      "learning_rate": 0.000177,
      "loss": 6.8593,
      "mean_token_accuracy": 0.11371754407882691,
      "num_tokens": 807671.0,
      "step": 355
    },
    {
      "entropy": 7.4689586639404295,
      "epoch": 0.02316900501995109,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001795,
      "loss": 6.8925,
      "mean_token_accuracy": 0.109658332914114,
      "num_tokens": 819142.0,
      "step": 360
    },
    {
      "entropy": 7.5210926055908205,
      "epoch": 0.023490796756339298,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000182,
      "loss": 6.8902,
      "mean_token_accuracy": 0.11664389371871949,
      "num_tokens": 829634.0,
      "step": 365
    },
    {
      "entropy": 7.562618398666382,
      "epoch": 0.023812588492727507,
      "grad_norm": 1.0625,
      "learning_rate": 0.0001845,
      "loss": 6.9457,
      "mean_token_accuracy": 0.11507059112191201,
      "num_tokens": 840723.0,
      "step": 370
    },
    {
      "entropy": 7.44709038734436,
      "epoch": 0.024134380229115716,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000187,
      "loss": 6.9014,
      "mean_token_accuracy": 0.11120686307549477,
      "num_tokens": 852299.0,
      "step": 375
    },
    {
      "entropy": 7.542909240722656,
      "epoch": 0.024456171965503926,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001895,
      "loss": 6.8845,
      "mean_token_accuracy": 0.11418962627649307,
      "num_tokens": 864427.0,
      "step": 380
    },
    {
      "entropy": 7.269602823257446,
      "epoch": 0.024777963701892135,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000192,
      "loss": 6.8457,
      "mean_token_accuracy": 0.11314126327633858,
      "num_tokens": 876188.0,
      "step": 385
    },
    {
      "entropy": 7.539252614974975,
      "epoch": 0.025099755438280344,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001945,
      "loss": 6.9751,
      "mean_token_accuracy": 0.10968251451849938,
      "num_tokens": 886835.0,
      "step": 390
    },
    {
      "entropy": 7.462894248962402,
      "epoch": 0.025421547174668553,
      "grad_norm": 1.09375,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.8758,
      "mean_token_accuracy": 0.1175595462322235,
      "num_tokens": 899609.0,
      "step": 395
    },
    {
      "entropy": 7.514445400238037,
      "epoch": 0.025743338911056766,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.9867,
      "mean_token_accuracy": 0.10821878015995026,
      "num_tokens": 912114.0,
      "step": 400
    },
    {
      "entropy": 7.434127473831177,
      "epoch": 0.026065130647444975,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000202,
      "loss": 6.9875,
      "mean_token_accuracy": 0.1139479123055935,
      "num_tokens": 923788.0,
      "step": 405
    },
    {
      "entropy": 7.338033294677734,
      "epoch": 0.026386922383833184,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.6791,
      "mean_token_accuracy": 0.1255500763654709,
      "num_tokens": 934952.0,
      "step": 410
    },
    {
      "entropy": 7.422152519226074,
      "epoch": 0.026708714120221393,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000207,
      "loss": 6.7758,
      "mean_token_accuracy": 0.11906903460621834,
      "num_tokens": 945780.0,
      "step": 415
    },
    {
      "entropy": 7.29374885559082,
      "epoch": 0.027030505856609603,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002095,
      "loss": 6.7765,
      "mean_token_accuracy": 0.12052410319447518,
      "num_tokens": 958502.0,
      "step": 420
    },
    {
      "entropy": 7.422839784622193,
      "epoch": 0.02735229759299781,
      "grad_norm": 1.46875,
      "learning_rate": 0.000212,
      "loss": 6.8635,
      "mean_token_accuracy": 0.11772049516439438,
      "num_tokens": 969339.0,
      "step": 425
    },
    {
      "entropy": 7.215829658508301,
      "epoch": 0.02767408932938602,
      "grad_norm": 1.125,
      "learning_rate": 0.0002145,
      "loss": 6.7171,
      "mean_token_accuracy": 0.12233989164233208,
      "num_tokens": 980260.0,
      "step": 430
    },
    {
      "entropy": 7.2986561298370365,
      "epoch": 0.02799588106577423,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.6963,
      "mean_token_accuracy": 0.11892440766096116,
      "num_tokens": 991960.0,
      "step": 435
    },
    {
      "entropy": 7.193790006637573,
      "epoch": 0.02831767280216244,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002195,
      "loss": 6.7273,
      "mean_token_accuracy": 0.11868971958756447,
      "num_tokens": 1002083.0,
      "step": 440
    },
    {
      "entropy": 7.308983230590821,
      "epoch": 0.02863946453855065,
      "grad_norm": 1.109375,
      "learning_rate": 0.000222,
      "loss": 6.7375,
      "mean_token_accuracy": 0.11723077073693275,
      "num_tokens": 1012896.0,
      "step": 445
    },
    {
      "entropy": 7.334416151046753,
      "epoch": 0.02896125627493886,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002245,
      "loss": 6.8189,
      "mean_token_accuracy": 0.12101824656128883,
      "num_tokens": 1024570.0,
      "step": 450
    },
    {
      "entropy": 7.2581829071044925,
      "epoch": 0.02928304801132707,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.6596,
      "mean_token_accuracy": 0.1274547331035137,
      "num_tokens": 1035934.0,
      "step": 455
    },
    {
      "entropy": 7.214243841171265,
      "epoch": 0.02960483974771528,
      "grad_norm": 1.109375,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.8555,
      "mean_token_accuracy": 0.11035279110074044,
      "num_tokens": 1047557.0,
      "step": 460
    },
    {
      "entropy": 7.284564542770386,
      "epoch": 0.02992663148410349,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.8011,
      "mean_token_accuracy": 0.11536358669400215,
      "num_tokens": 1059387.0,
      "step": 465
    },
    {
      "entropy": 7.255402135848999,
      "epoch": 0.030248423220491698,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.8003,
      "mean_token_accuracy": 0.11578696817159653,
      "num_tokens": 1070756.0,
      "step": 470
    },
    {
      "entropy": 7.274528503417969,
      "epoch": 0.030570214956879907,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000237,
      "loss": 6.7873,
      "mean_token_accuracy": 0.1295149154961109,
      "num_tokens": 1083147.0,
      "step": 475
    },
    {
      "entropy": 7.171627616882324,
      "epoch": 0.030892006693268116,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002395,
      "loss": 6.6541,
      "mean_token_accuracy": 0.12174521461129188,
      "num_tokens": 1094300.0,
      "step": 480
    },
    {
      "entropy": 7.192223072052002,
      "epoch": 0.031213798429656325,
      "grad_norm": 1.015625,
      "learning_rate": 0.000242,
      "loss": 6.7988,
      "mean_token_accuracy": 0.11465005055069924,
      "num_tokens": 1105991.0,
      "step": 485
    },
    {
      "entropy": 7.2635111808776855,
      "epoch": 0.031535590166044535,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002445,
      "loss": 6.8278,
      "mean_token_accuracy": 0.11537677273154259,
      "num_tokens": 1118065.0,
      "step": 490
    },
    {
      "entropy": 7.217528486251831,
      "epoch": 0.031857381902432744,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000247,
      "loss": 6.772,
      "mean_token_accuracy": 0.12680590748786927,
      "num_tokens": 1130521.0,
      "step": 495
    },
    {
      "entropy": 7.099181175231934,
      "epoch": 0.03217917363882095,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002495,
      "loss": 6.7436,
      "mean_token_accuracy": 0.11461131051182746,
      "num_tokens": 1143241.0,
      "step": 500
    },
    {
      "entropy": 7.2616432189941404,
      "epoch": 0.03250096537520916,
      "grad_norm": 1.078125,
      "learning_rate": 0.000252,
      "loss": 6.698,
      "mean_token_accuracy": 0.1210433527827263,
      "num_tokens": 1154240.0,
      "step": 505
    },
    {
      "entropy": 7.08572096824646,
      "epoch": 0.03282275711159737,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002545,
      "loss": 6.6583,
      "mean_token_accuracy": 0.12035112828016281,
      "num_tokens": 1166134.0,
      "step": 510
    },
    {
      "entropy": 7.13088059425354,
      "epoch": 0.03314454884798558,
      "grad_norm": 1.140625,
      "learning_rate": 0.000257,
      "loss": 6.6952,
      "mean_token_accuracy": 0.12703741267323493,
      "num_tokens": 1177183.0,
      "step": 515
    },
    {
      "entropy": 7.25070538520813,
      "epoch": 0.0334663405843738,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002595,
      "loss": 6.8061,
      "mean_token_accuracy": 0.11820513978600503,
      "num_tokens": 1189266.0,
      "step": 520
    },
    {
      "entropy": 7.051825141906738,
      "epoch": 0.033788132320762006,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000262,
      "loss": 6.6082,
      "mean_token_accuracy": 0.12360260263085365,
      "num_tokens": 1199718.0,
      "step": 525
    },
    {
      "entropy": 7.207192850112915,
      "epoch": 0.034109924057150215,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.7443,
      "mean_token_accuracy": 0.1214994914829731,
      "num_tokens": 1210278.0,
      "step": 530
    },
    {
      "entropy": 7.183014011383056,
      "epoch": 0.034431715793538424,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.6847,
      "mean_token_accuracy": 0.12363504841923714,
      "num_tokens": 1221012.0,
      "step": 535
    },
    {
      "entropy": 7.097731018066407,
      "epoch": 0.03475350752992663,
      "grad_norm": 1.078125,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.6674,
      "mean_token_accuracy": 0.1205046035349369,
      "num_tokens": 1232687.0,
      "step": 540
    },
    {
      "entropy": 7.153340673446655,
      "epoch": 0.03507529926631484,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.6428,
      "mean_token_accuracy": 0.127882944047451,
      "num_tokens": 1243994.0,
      "step": 545
    },
    {
      "entropy": 7.048250007629394,
      "epoch": 0.03539709100270305,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002745,
      "loss": 6.6121,
      "mean_token_accuracy": 0.1261216700077057,
      "num_tokens": 1254705.0,
      "step": 550
    },
    {
      "entropy": 7.134361505508423,
      "epoch": 0.03571888273909126,
      "grad_norm": 1.15625,
      "learning_rate": 0.000277,
      "loss": 6.7227,
      "mean_token_accuracy": 0.12077815011143685,
      "num_tokens": 1266367.0,
      "step": 555
    },
    {
      "entropy": 7.047533082962036,
      "epoch": 0.03604067447547947,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002795,
      "loss": 6.6958,
      "mean_token_accuracy": 0.12296305298805237,
      "num_tokens": 1279004.0,
      "step": 560
    },
    {
      "entropy": 7.034592580795288,
      "epoch": 0.03636246621186768,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.5162,
      "mean_token_accuracy": 0.12805850878357888,
      "num_tokens": 1290980.0,
      "step": 565
    },
    {
      "entropy": 7.040336656570434,
      "epoch": 0.03668425794825589,
      "grad_norm": 0.9375,
      "learning_rate": 0.0002845,
      "loss": 6.6749,
      "mean_token_accuracy": 0.12128347530961037,
      "num_tokens": 1302396.0,
      "step": 570
    },
    {
      "entropy": 7.105546283721924,
      "epoch": 0.0370060496846441,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000287,
      "loss": 6.6356,
      "mean_token_accuracy": 0.1311293475329876,
      "num_tokens": 1312304.0,
      "step": 575
    },
    {
      "entropy": 7.081285810470581,
      "epoch": 0.03732784142103231,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002895,
      "loss": 6.7769,
      "mean_token_accuracy": 0.12450022101402283,
      "num_tokens": 1325487.0,
      "step": 580
    },
    {
      "entropy": 7.005313682556152,
      "epoch": 0.037649633157420516,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000292,
      "loss": 6.6247,
      "mean_token_accuracy": 0.126084253937006,
      "num_tokens": 1336754.0,
      "step": 585
    },
    {
      "entropy": 7.028646039962768,
      "epoch": 0.037971424893808725,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002945,
      "loss": 6.6309,
      "mean_token_accuracy": 0.12374395728111268,
      "num_tokens": 1347918.0,
      "step": 590
    },
    {
      "entropy": 7.05291862487793,
      "epoch": 0.038293216630196934,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000297,
      "loss": 6.6336,
      "mean_token_accuracy": 0.13012145236134529,
      "num_tokens": 1359981.0,
      "step": 595
    },
    {
      "entropy": 7.028671169281006,
      "epoch": 0.03861500836658514,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002995,
      "loss": 6.6423,
      "mean_token_accuracy": 0.1260800801217556,
      "num_tokens": 1371359.0,
      "step": 600
    },
    {
      "entropy": 7.067501354217529,
      "epoch": 0.03893680010297335,
      "grad_norm": 1.125,
      "learning_rate": 0.000302,
      "loss": 6.6909,
      "mean_token_accuracy": 0.11782569289207459,
      "num_tokens": 1383829.0,
      "step": 605
    },
    {
      "entropy": 7.0067911624908445,
      "epoch": 0.03925859183936156,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003045,
      "loss": 6.5645,
      "mean_token_accuracy": 0.13620257899165153,
      "num_tokens": 1394266.0,
      "step": 610
    },
    {
      "entropy": 6.985071325302124,
      "epoch": 0.03958038357574978,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000307,
      "loss": 6.5691,
      "mean_token_accuracy": 0.1302391119301319,
      "num_tokens": 1405798.0,
      "step": 615
    },
    {
      "entropy": 6.973874282836914,
      "epoch": 0.03990217531213799,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003095,
      "loss": 6.5689,
      "mean_token_accuracy": 0.13187146857380866,
      "num_tokens": 1417411.0,
      "step": 620
    },
    {
      "entropy": 7.0129351139068605,
      "epoch": 0.040223967048526196,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000312,
      "loss": 6.6911,
      "mean_token_accuracy": 0.12921356335282325,
      "num_tokens": 1428987.0,
      "step": 625
    },
    {
      "entropy": 6.938956356048584,
      "epoch": 0.040545758784914405,
      "grad_norm": 1.25,
      "learning_rate": 0.0003145,
      "loss": 6.6049,
      "mean_token_accuracy": 0.12962101995944977,
      "num_tokens": 1440023.0,
      "step": 630
    },
    {
      "entropy": 7.100988531112671,
      "epoch": 0.040867550521302615,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000317,
      "loss": 6.6762,
      "mean_token_accuracy": 0.1187206856906414,
      "num_tokens": 1451078.0,
      "step": 635
    },
    {
      "entropy": 7.031370496749878,
      "epoch": 0.041189342257690824,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003195,
      "loss": 6.6551,
      "mean_token_accuracy": 0.12284370735287667,
      "num_tokens": 1462446.0,
      "step": 640
    },
    {
      "entropy": 6.888606119155884,
      "epoch": 0.04151113399407903,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000322,
      "loss": 6.5127,
      "mean_token_accuracy": 0.131097362190485,
      "num_tokens": 1472084.0,
      "step": 645
    },
    {
      "entropy": 6.948457860946656,
      "epoch": 0.04183292573046724,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.553,
      "mean_token_accuracy": 0.1207494542002678,
      "num_tokens": 1482464.0,
      "step": 650
    },
    {
      "entropy": 6.960102081298828,
      "epoch": 0.04215471746685545,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.592,
      "mean_token_accuracy": 0.12484122142195701,
      "num_tokens": 1492789.0,
      "step": 655
    },
    {
      "entropy": 6.917847633361816,
      "epoch": 0.04247650920324366,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.5871,
      "mean_token_accuracy": 0.1326150640845299,
      "num_tokens": 1504438.0,
      "step": 660
    },
    {
      "entropy": 7.005792331695557,
      "epoch": 0.04279830093963187,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.5294,
      "mean_token_accuracy": 0.1296870678663254,
      "num_tokens": 1515430.0,
      "step": 665
    },
    {
      "entropy": 6.8461448669433596,
      "epoch": 0.04312009267602008,
      "grad_norm": 1.125,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.4998,
      "mean_token_accuracy": 0.13402296751737594,
      "num_tokens": 1525804.0,
      "step": 670
    },
    {
      "entropy": 6.924587154388428,
      "epoch": 0.04344188441240829,
      "grad_norm": 1.0625,
      "learning_rate": 0.000337,
      "loss": 6.5119,
      "mean_token_accuracy": 0.12666192203760146,
      "num_tokens": 1537638.0,
      "step": 675
    },
    {
      "entropy": 6.82023057937622,
      "epoch": 0.0437636761487965,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003395,
      "loss": 6.4856,
      "mean_token_accuracy": 0.13445887267589568,
      "num_tokens": 1548232.0,
      "step": 680
    },
    {
      "entropy": 6.93139066696167,
      "epoch": 0.044085467885184706,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000342,
      "loss": 6.5018,
      "mean_token_accuracy": 0.13019829168915747,
      "num_tokens": 1558781.0,
      "step": 685
    },
    {
      "entropy": 6.913920307159424,
      "epoch": 0.044407259621572916,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.4656,
      "mean_token_accuracy": 0.1352573812007904,
      "num_tokens": 1571387.0,
      "step": 690
    },
    {
      "entropy": 6.8577742099761965,
      "epoch": 0.044729051357961125,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000347,
      "loss": 6.5251,
      "mean_token_accuracy": 0.13332423493266105,
      "num_tokens": 1582471.0,
      "step": 695
    },
    {
      "entropy": 6.825550699234009,
      "epoch": 0.045050843094349334,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003495,
      "loss": 6.484,
      "mean_token_accuracy": 0.13259591534733772,
      "num_tokens": 1592967.0,
      "step": 700
    },
    {
      "entropy": 6.84607195854187,
      "epoch": 0.04537263483073754,
      "grad_norm": 1.15625,
      "learning_rate": 0.000352,
      "loss": 6.4834,
      "mean_token_accuracy": 0.1336855672299862,
      "num_tokens": 1603772.0,
      "step": 705
    },
    {
      "entropy": 6.906696319580078,
      "epoch": 0.04569442656712576,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003545,
      "loss": 6.4789,
      "mean_token_accuracy": 0.12700057551264762,
      "num_tokens": 1614924.0,
      "step": 710
    },
    {
      "entropy": 6.7945763111114506,
      "epoch": 0.04601621830351397,
      "grad_norm": 0.953125,
      "learning_rate": 0.000357,
      "loss": 6.4715,
      "mean_token_accuracy": 0.13077478110790253,
      "num_tokens": 1626947.0,
      "step": 715
    },
    {
      "entropy": 6.796080684661865,
      "epoch": 0.04633801003990218,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003595,
      "loss": 6.3518,
      "mean_token_accuracy": 0.13691332638263704,
      "num_tokens": 1638651.0,
      "step": 720
    },
    {
      "entropy": 6.753641128540039,
      "epoch": 0.04665980177629039,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000362,
      "loss": 6.4204,
      "mean_token_accuracy": 0.1394519940018654,
      "num_tokens": 1650157.0,
      "step": 725
    },
    {
      "entropy": 6.7741881847381595,
      "epoch": 0.046981593512678596,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003645,
      "loss": 6.418,
      "mean_token_accuracy": 0.133515302836895,
      "num_tokens": 1661763.0,
      "step": 730
    },
    {
      "entropy": 6.971210908889771,
      "epoch": 0.047303385249066805,
      "grad_norm": 1.109375,
      "learning_rate": 0.000367,
      "loss": 6.6075,
      "mean_token_accuracy": 0.12115295752882957,
      "num_tokens": 1674051.0,
      "step": 735
    },
    {
      "entropy": 6.856794738769532,
      "epoch": 0.047625176985455014,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003695,
      "loss": 6.4626,
      "mean_token_accuracy": 0.1338093623518944,
      "num_tokens": 1685240.0,
      "step": 740
    },
    {
      "entropy": 6.777393484115601,
      "epoch": 0.047946968721843224,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000372,
      "loss": 6.4435,
      "mean_token_accuracy": 0.1303575173020363,
      "num_tokens": 1697472.0,
      "step": 745
    },
    {
      "entropy": 6.796992444992066,
      "epoch": 0.04826876045823143,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003745,
      "loss": 6.4543,
      "mean_token_accuracy": 0.1324995703995228,
      "num_tokens": 1708383.0,
      "step": 750
    },
    {
      "entropy": 6.788374280929565,
      "epoch": 0.04859055219461964,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000377,
      "loss": 6.4631,
      "mean_token_accuracy": 0.13005806729197503,
      "num_tokens": 1720367.0,
      "step": 755
    },
    {
      "entropy": 6.780432558059692,
      "epoch": 0.04891234393100785,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003795,
      "loss": 6.445,
      "mean_token_accuracy": 0.13468715846538543,
      "num_tokens": 1731252.0,
      "step": 760
    },
    {
      "entropy": 6.793448638916016,
      "epoch": 0.04923413566739606,
      "grad_norm": 1.140625,
      "learning_rate": 0.000382,
      "loss": 6.4776,
      "mean_token_accuracy": 0.1371725916862488,
      "num_tokens": 1741865.0,
      "step": 765
    },
    {
      "entropy": 6.7858991622924805,
      "epoch": 0.04955592740378427,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003845,
      "loss": 6.4394,
      "mean_token_accuracy": 0.13146187737584114,
      "num_tokens": 1754036.0,
      "step": 770
    },
    {
      "entropy": 6.814309310913086,
      "epoch": 0.04987771914017248,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.4836,
      "mean_token_accuracy": 0.12894209176301957,
      "num_tokens": 1764717.0,
      "step": 775
    },
    {
      "entropy": 6.754202890396118,
      "epoch": 0.05019951087656069,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.383,
      "mean_token_accuracy": 0.13913563415408134,
      "num_tokens": 1777078.0,
      "step": 780
    },
    {
      "entropy": 6.852421665191651,
      "epoch": 0.0505213026129489,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.453,
      "mean_token_accuracy": 0.12606742605566978,
      "num_tokens": 1788393.0,
      "step": 785
    },
    {
      "entropy": 6.727867794036865,
      "epoch": 0.050843094349337106,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.5638,
      "mean_token_accuracy": 0.12575344145298004,
      "num_tokens": 1799535.0,
      "step": 790
    },
    {
      "entropy": 6.849759912490844,
      "epoch": 0.051164886085725315,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.3297,
      "mean_token_accuracy": 0.13775620311498643,
      "num_tokens": 1809733.0,
      "step": 795
    },
    {
      "entropy": 6.681720924377442,
      "epoch": 0.05148667782211353,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003995,
      "loss": 6.3793,
      "mean_token_accuracy": 0.13796758204698562,
      "num_tokens": 1821535.0,
      "step": 800
    },
    {
      "entropy": 6.674613428115845,
      "epoch": 0.05180846955850174,
      "grad_norm": 1.140625,
      "learning_rate": 0.000402,
      "loss": 6.3629,
      "mean_token_accuracy": 0.13800609484314919,
      "num_tokens": 1832355.0,
      "step": 805
    },
    {
      "entropy": 6.773230028152466,
      "epoch": 0.05213026129488995,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004045,
      "loss": 6.3386,
      "mean_token_accuracy": 0.1425774149596691,
      "num_tokens": 1843340.0,
      "step": 810
    },
    {
      "entropy": 6.642430782318115,
      "epoch": 0.05245205303127816,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.3684,
      "mean_token_accuracy": 0.13916084170341492,
      "num_tokens": 1854233.0,
      "step": 815
    },
    {
      "entropy": 6.633238887786865,
      "epoch": 0.05277384476766637,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004095,
      "loss": 6.3114,
      "mean_token_accuracy": 0.13679933026432992,
      "num_tokens": 1864860.0,
      "step": 820
    },
    {
      "entropy": 6.780126476287842,
      "epoch": 0.05309563650405458,
      "grad_norm": 1.0625,
      "learning_rate": 0.000412,
      "loss": 6.2908,
      "mean_token_accuracy": 0.14247430860996246,
      "num_tokens": 1875981.0,
      "step": 825
    },
    {
      "entropy": 6.616991996765137,
      "epoch": 0.05341742824044279,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004145,
      "loss": 6.4106,
      "mean_token_accuracy": 0.13161869645118712,
      "num_tokens": 1889147.0,
      "step": 830
    },
    {
      "entropy": 6.857424640655518,
      "epoch": 0.053739219976830996,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000417,
      "loss": 6.4176,
      "mean_token_accuracy": 0.13287485763430595,
      "num_tokens": 1900823.0,
      "step": 835
    },
    {
      "entropy": 6.622591257095337,
      "epoch": 0.054061011713219205,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004195,
      "loss": 6.3323,
      "mean_token_accuracy": 0.13574840426445006,
      "num_tokens": 1910882.0,
      "step": 840
    },
    {
      "entropy": 6.52352614402771,
      "epoch": 0.054382803449607414,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000422,
      "loss": 6.2885,
      "mean_token_accuracy": 0.1370130032300949,
      "num_tokens": 1922228.0,
      "step": 845
    },
    {
      "entropy": 6.675861597061157,
      "epoch": 0.05470459518599562,
      "grad_norm": 1.125,
      "learning_rate": 0.0004245,
      "loss": 6.3254,
      "mean_token_accuracy": 0.13999047502875328,
      "num_tokens": 1933400.0,
      "step": 850
    },
    {
      "entropy": 6.6365453720092775,
      "epoch": 0.05502638692238383,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000427,
      "loss": 6.3747,
      "mean_token_accuracy": 0.13206338435411452,
      "num_tokens": 1944717.0,
      "step": 855
    },
    {
      "entropy": 6.806672430038452,
      "epoch": 0.05534817865877204,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004295,
      "loss": 6.4465,
      "mean_token_accuracy": 0.1298723392188549,
      "num_tokens": 1956116.0,
      "step": 860
    },
    {
      "entropy": 6.787279891967773,
      "epoch": 0.05566997039516025,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000432,
      "loss": 6.4613,
      "mean_token_accuracy": 0.13620107173919677,
      "num_tokens": 1967183.0,
      "step": 865
    },
    {
      "entropy": 6.522842454910278,
      "epoch": 0.05599176213154846,
      "grad_norm": 1.0,
      "learning_rate": 0.0004345,
      "loss": 6.2558,
      "mean_token_accuracy": 0.14262447282671928,
      "num_tokens": 1978943.0,
      "step": 870
    },
    {
      "entropy": 6.580580043792724,
      "epoch": 0.05631355386793667,
      "grad_norm": 1.109375,
      "learning_rate": 0.000437,
      "loss": 6.2737,
      "mean_token_accuracy": 0.14003785327076912,
      "num_tokens": 1990818.0,
      "step": 875
    },
    {
      "entropy": 6.574157285690307,
      "epoch": 0.05663534560432488,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004395,
      "loss": 6.2641,
      "mean_token_accuracy": 0.14258711114525796,
      "num_tokens": 2002309.0,
      "step": 880
    },
    {
      "entropy": 6.655279111862183,
      "epoch": 0.05695713734071309,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000442,
      "loss": 6.4362,
      "mean_token_accuracy": 0.1311865270137787,
      "num_tokens": 2013131.0,
      "step": 885
    },
    {
      "entropy": 6.861530065536499,
      "epoch": 0.0572789290771013,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004445,
      "loss": 6.5215,
      "mean_token_accuracy": 0.12429815754294396,
      "num_tokens": 2024881.0,
      "step": 890
    },
    {
      "entropy": 6.6120974063873295,
      "epoch": 0.05760072081348951,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000447,
      "loss": 6.3878,
      "mean_token_accuracy": 0.13595437854528428,
      "num_tokens": 2035633.0,
      "step": 895
    },
    {
      "entropy": 6.62742280960083,
      "epoch": 0.05792251254987772,
      "grad_norm": 1.125,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.2356,
      "mean_token_accuracy": 0.14306803271174431,
      "num_tokens": 2046237.0,
      "step": 900
    },
    {
      "entropy": 6.552933835983277,
      "epoch": 0.05824430428626593,
      "grad_norm": 1.25,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.2148,
      "mean_token_accuracy": 0.14152271002531053,
      "num_tokens": 2058158.0,
      "step": 905
    },
    {
      "entropy": 6.638343429565429,
      "epoch": 0.05856609602265414,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.4302,
      "mean_token_accuracy": 0.13081190586090088,
      "num_tokens": 2068364.0,
      "step": 910
    },
    {
      "entropy": 6.667097568511963,
      "epoch": 0.05888788775904235,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.3678,
      "mean_token_accuracy": 0.12357937470078469,
      "num_tokens": 2080168.0,
      "step": 915
    },
    {
      "entropy": 6.584589195251465,
      "epoch": 0.05920967949543056,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.2459,
      "mean_token_accuracy": 0.14396512135863304,
      "num_tokens": 2089908.0,
      "step": 920
    },
    {
      "entropy": 6.65417799949646,
      "epoch": 0.05953147123181877,
      "grad_norm": 1.0625,
      "learning_rate": 0.000462,
      "loss": 6.3424,
      "mean_token_accuracy": 0.13211728110909463,
      "num_tokens": 2101403.0,
      "step": 925
    },
    {
      "entropy": 6.49568543434143,
      "epoch": 0.05985326296820698,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004645,
      "loss": 6.2153,
      "mean_token_accuracy": 0.14081560969352722,
      "num_tokens": 2113940.0,
      "step": 930
    },
    {
      "entropy": 6.544306850433349,
      "epoch": 0.060175054704595186,
      "grad_norm": 1.046875,
      "learning_rate": 0.000467,
      "loss": 6.2557,
      "mean_token_accuracy": 0.1435394138097763,
      "num_tokens": 2124467.0,
      "step": 935
    },
    {
      "entropy": 6.588162183761597,
      "epoch": 0.060496846440983396,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004695,
      "loss": 6.2399,
      "mean_token_accuracy": 0.1443529948592186,
      "num_tokens": 2136147.0,
      "step": 940
    },
    {
      "entropy": 6.542601013183594,
      "epoch": 0.060818638177371605,
      "grad_norm": 1.078125,
      "learning_rate": 0.000472,
      "loss": 6.2855,
      "mean_token_accuracy": 0.14248571097850798,
      "num_tokens": 2147494.0,
      "step": 945
    },
    {
      "entropy": 6.513602781295776,
      "epoch": 0.061140429913759814,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004745,
      "loss": 6.1673,
      "mean_token_accuracy": 0.14396455883979797,
      "num_tokens": 2157660.0,
      "step": 950
    },
    {
      "entropy": 6.415520524978637,
      "epoch": 0.06146222165014802,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000477,
      "loss": 6.2775,
      "mean_token_accuracy": 0.1388951927423477,
      "num_tokens": 2170629.0,
      "step": 955
    },
    {
      "entropy": 6.536985874176025,
      "epoch": 0.06178401338653623,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004795,
      "loss": 6.1999,
      "mean_token_accuracy": 0.14611724615097046,
      "num_tokens": 2181945.0,
      "step": 960
    },
    {
      "entropy": 6.555672550201416,
      "epoch": 0.06210580512292444,
      "grad_norm": 1.03125,
      "learning_rate": 0.000482,
      "loss": 6.2884,
      "mean_token_accuracy": 0.13931862711906434,
      "num_tokens": 2192975.0,
      "step": 965
    },
    {
      "entropy": 6.523555850982666,
      "epoch": 0.06242759685931265,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004845,
      "loss": 6.2292,
      "mean_token_accuracy": 0.14297346845269204,
      "num_tokens": 2203929.0,
      "step": 970
    },
    {
      "entropy": 6.551759815216064,
      "epoch": 0.06274938859570087,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000487,
      "loss": 6.2444,
      "mean_token_accuracy": 0.14618593230843543,
      "num_tokens": 2214296.0,
      "step": 975
    },
    {
      "entropy": 6.542734003067016,
      "epoch": 0.06307118033208907,
      "grad_norm": 1.125,
      "learning_rate": 0.0004895,
      "loss": 6.2657,
      "mean_token_accuracy": 0.1420005366206169,
      "num_tokens": 2224834.0,
      "step": 980
    },
    {
      "entropy": 6.490674257278442,
      "epoch": 0.06339297206847729,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000492,
      "loss": 6.1984,
      "mean_token_accuracy": 0.14470900744199752,
      "num_tokens": 2236261.0,
      "step": 985
    },
    {
      "entropy": 6.574054622650147,
      "epoch": 0.06371476380486549,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004945,
      "loss": 6.445,
      "mean_token_accuracy": 0.1385732315480709,
      "num_tokens": 2248056.0,
      "step": 990
    },
    {
      "entropy": 6.560631370544433,
      "epoch": 0.0640365555412537,
      "grad_norm": 1.046875,
      "learning_rate": 0.000497,
      "loss": 6.2375,
      "mean_token_accuracy": 0.13548775166273117,
      "num_tokens": 2258304.0,
      "step": 995
    },
    {
      "entropy": 6.553973388671875,
      "epoch": 0.0643583472776419,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995,
      "loss": 6.3626,
      "mean_token_accuracy": 0.14208254665136338,
      "num_tokens": 2269541.0,
      "step": 1000
    },
    {
      "entropy": 6.420248031616211,
      "epoch": 0.06468013901403012,
      "grad_norm": 1.21875,
      "learning_rate": 0.000499998026082006,
      "loss": 6.1734,
      "mean_token_accuracy": 0.15447934120893478,
      "num_tokens": 2280812.0,
      "step": 1005
    },
    {
      "entropy": 6.664107942581177,
      "epoch": 0.06500193075041832,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999900070995136,
      "loss": 6.3506,
      "mean_token_accuracy": 0.14142308607697487,
      "num_tokens": 2291833.0,
      "step": 1010
    },
    {
      "entropy": 6.523155164718628,
      "epoch": 0.06532372248680654,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999758199023239,
      "loss": 6.2563,
      "mean_token_accuracy": 0.145108150690794,
      "num_tokens": 2302218.0,
      "step": 1015
    },
    {
      "entropy": 6.581012916564942,
      "epoch": 0.06564551422319474,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999554648793858,
      "loss": 6.3207,
      "mean_token_accuracy": 0.13909797295928,
      "num_tokens": 2313355.0,
      "step": 1020
    },
    {
      "entropy": 6.397714710235595,
      "epoch": 0.06596730595958296,
      "grad_norm": 1.25,
      "learning_rate": 0.0004999289425887425,
      "loss": 6.1485,
      "mean_token_accuracy": 0.145339135825634,
      "num_tokens": 2324652.0,
      "step": 1025
    },
    {
      "entropy": 6.521378660202027,
      "epoch": 0.06628909769597116,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998962537575161,
      "loss": 6.2546,
      "mean_token_accuracy": 0.1398783229291439,
      "num_tokens": 2335006.0,
      "step": 1030
    },
    {
      "entropy": 6.371570062637329,
      "epoch": 0.06661088943235938,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998573992818874,
      "loss": 6.1238,
      "mean_token_accuracy": 0.14918210208415986,
      "num_tokens": 2347261.0,
      "step": 1035
    },
    {
      "entropy": 6.545182466506958,
      "epoch": 0.0669326811687476,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998123802270715,
      "loss": 6.2314,
      "mean_token_accuracy": 0.14083297401666642,
      "num_tokens": 2358106.0,
      "step": 1040
    },
    {
      "entropy": 6.477479410171509,
      "epoch": 0.0672544729051358,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997611978272886,
      "loss": 6.2811,
      "mean_token_accuracy": 0.14108368679881095,
      "num_tokens": 2369182.0,
      "step": 1045
    },
    {
      "entropy": 6.415492010116577,
      "epoch": 0.06757626464152401,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997038534857298,
      "loss": 6.0323,
      "mean_token_accuracy": 0.150262039154768,
      "num_tokens": 2379298.0,
      "step": 1050
    },
    {
      "entropy": 6.391033077239991,
      "epoch": 0.06789805637791221,
      "grad_norm": 1.125,
      "learning_rate": 0.0004996403487745194,
      "loss": 6.1398,
      "mean_token_accuracy": 0.14792790189385413,
      "num_tokens": 2389920.0,
      "step": 1055
    },
    {
      "entropy": 6.480355501174927,
      "epoch": 0.06821984811430043,
      "grad_norm": 1.0,
      "learning_rate": 0.000499570685434671,
      "loss": 6.2449,
      "mean_token_accuracy": 0.13905822485685349,
      "num_tokens": 2401743.0,
      "step": 1060
    },
    {
      "entropy": 6.476265621185303,
      "epoch": 0.06854163985068863,
      "grad_norm": 1.125,
      "learning_rate": 0.0004994948653760405,
      "loss": 6.2046,
      "mean_token_accuracy": 0.14363889396190643,
      "num_tokens": 2411770.0,
      "step": 1065
    },
    {
      "entropy": 6.361932802200317,
      "epoch": 0.06886343158707685,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004994128906772729,
      "loss": 6.1164,
      "mean_token_accuracy": 0.14849536269903182,
      "num_tokens": 2422448.0,
      "step": 1070
    },
    {
      "entropy": 6.517858362197876,
      "epoch": 0.06918522332346505,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499324763585746,
      "loss": 6.1835,
      "mean_token_accuracy": 0.14678900837898254,
      "num_tokens": 2433428.0,
      "step": 1075
    },
    {
      "entropy": 6.331155014038086,
      "epoch": 0.06950701505985327,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004992304865175085,
      "loss": 6.1545,
      "mean_token_accuracy": 0.14461304917931556,
      "num_tokens": 2445538.0,
      "step": 1080
    },
    {
      "entropy": 6.442676591873169,
      "epoch": 0.06982880679624147,
      "grad_norm": 1.125,
      "learning_rate": 0.0004991300620572138,
      "loss": 6.1283,
      "mean_token_accuracy": 0.14782453551888466,
      "num_tokens": 2455656.0,
      "step": 1085
    },
    {
      "entropy": 6.333368253707886,
      "epoch": 0.07015059853262968,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004990234929580494,
      "loss": 6.1665,
      "mean_token_accuracy": 0.14236130341887474,
      "num_tokens": 2466933.0,
      "step": 1090
    },
    {
      "entropy": 6.4185097217559814,
      "epoch": 0.07047239026901789,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004989107821416609,
      "loss": 6.2137,
      "mean_token_accuracy": 0.14996169060468673,
      "num_tokens": 2479501.0,
      "step": 1095
    },
    {
      "entropy": 6.408093643188477,
      "epoch": 0.0707941820054061,
      "grad_norm": 1.0,
      "learning_rate": 0.0004987919326980723,
      "loss": 6.1955,
      "mean_token_accuracy": 0.13943222388625146,
      "num_tokens": 2491571.0,
      "step": 1100
    },
    {
      "entropy": 6.328987503051758,
      "epoch": 0.0711159737417943,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004986669478856011,
      "loss": 6.0414,
      "mean_token_accuracy": 0.15177379101514815,
      "num_tokens": 2502517.0,
      "step": 1105
    },
    {
      "entropy": 6.463210535049439,
      "epoch": 0.07143776547818252,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004985358311307688,
      "loss": 6.2415,
      "mean_token_accuracy": 0.14047156423330306,
      "num_tokens": 2514711.0,
      "step": 1110
    },
    {
      "entropy": 6.4414870262146,
      "epoch": 0.07175955721457072,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004983985860282081,
      "loss": 6.2103,
      "mean_token_accuracy": 0.14568767324090004,
      "num_tokens": 2526473.0,
      "step": 1115
    },
    {
      "entropy": 6.478610181808472,
      "epoch": 0.07208134895095894,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004982552163405623,
      "loss": 6.1723,
      "mean_token_accuracy": 0.14119128584861756,
      "num_tokens": 2537346.0,
      "step": 1120
    },
    {
      "entropy": 6.40332260131836,
      "epoch": 0.07240314068734714,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004981057259983839,
      "loss": 6.2217,
      "mean_token_accuracy": 0.13824794590473174,
      "num_tokens": 2549476.0,
      "step": 1125
    },
    {
      "entropy": 6.371269559860229,
      "epoch": 0.07272493242373536,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004979501191000262,
      "loss": 6.113,
      "mean_token_accuracy": 0.14911531060934066,
      "num_tokens": 2559766.0,
      "step": 1130
    },
    {
      "entropy": 6.425234127044678,
      "epoch": 0.07304672416012357,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004977883999115311,
      "loss": 6.2251,
      "mean_token_accuracy": 0.14199284091591835,
      "num_tokens": 2571559.0,
      "step": 1135
    },
    {
      "entropy": 6.431259536743164,
      "epoch": 0.07336851589651178,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004976205728665113,
      "loss": 6.1112,
      "mean_token_accuracy": 0.15070913657546042,
      "num_tokens": 2582530.0,
      "step": 1140
    },
    {
      "entropy": 6.24943962097168,
      "epoch": 0.07369030763289999,
      "grad_norm": 1.125,
      "learning_rate": 0.0004974466425660307,
      "loss": 6.0109,
      "mean_token_accuracy": 0.15389537215232849,
      "num_tokens": 2594100.0,
      "step": 1145
    },
    {
      "entropy": 6.366129446029663,
      "epoch": 0.0740120993692882,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004972666137784759,
      "loss": 6.2097,
      "mean_token_accuracy": 0.14721803665161132,
      "num_tokens": 2605853.0,
      "step": 1150
    },
    {
      "entropy": 6.466001224517822,
      "epoch": 0.07433389110567641,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004970804914394271,
      "loss": 6.1948,
      "mean_token_accuracy": 0.14637096226215363,
      "num_tokens": 2616298.0,
      "step": 1155
    },
    {
      "entropy": 6.396643543243409,
      "epoch": 0.07465568284206461,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004968882806515225,
      "loss": 6.185,
      "mean_token_accuracy": 0.14693560302257538,
      "num_tokens": 2627103.0,
      "step": 1160
    },
    {
      "entropy": 6.439416837692261,
      "epoch": 0.07497747457845283,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004966899866843177,
      "loss": 6.1133,
      "mean_token_accuracy": 0.14630427211523056,
      "num_tokens": 2638657.0,
      "step": 1165
    },
    {
      "entropy": 6.329187059402466,
      "epoch": 0.07529926631484103,
      "grad_norm": 1.015625,
      "learning_rate": 0.000496485614974142,
      "loss": 6.228,
      "mean_token_accuracy": 0.1431046985089779,
      "num_tokens": 2650170.0,
      "step": 1170
    },
    {
      "entropy": 6.464993619918824,
      "epoch": 0.07562105805122925,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004962751711239492,
      "loss": 6.1557,
      "mean_token_accuracy": 0.14459144622087478,
      "num_tokens": 2662060.0,
      "step": 1175
    },
    {
      "entropy": 6.411845684051514,
      "epoch": 0.07594284978761745,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004960586609031636,
      "loss": 6.2545,
      "mean_token_accuracy": 0.14064157232642174,
      "num_tokens": 2674031.0,
      "step": 1180
    },
    {
      "entropy": 6.344238042831421,
      "epoch": 0.07626464152400567,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004958360902475224,
      "loss": 6.0551,
      "mean_token_accuracy": 0.15323207825422286,
      "num_tokens": 2685133.0,
      "step": 1185
    },
    {
      "entropy": 6.328037738800049,
      "epoch": 0.07658643326039387,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004956074652589125,
      "loss": 6.19,
      "mean_token_accuracy": 0.14459021687507628,
      "num_tokens": 2695256.0,
      "step": 1190
    },
    {
      "entropy": 6.494858503341675,
      "epoch": 0.07690822499678208,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004953727922052035,
      "loss": 6.1907,
      "mean_token_accuracy": 0.1448133870959282,
      "num_tokens": 2705322.0,
      "step": 1195
    },
    {
      "entropy": 6.359596300125122,
      "epoch": 0.07723001673317029,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004951320775200756,
      "loss": 6.1101,
      "mean_token_accuracy": 0.14269352331757545,
      "num_tokens": 2717911.0,
      "step": 1200
    },
    {
      "entropy": 6.399364423751831,
      "epoch": 0.0775518084695585,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004948853278028436,
      "loss": 6.2189,
      "mean_token_accuracy": 0.13950742781162262,
      "num_tokens": 2729321.0,
      "step": 1205
    },
    {
      "entropy": 6.399625301361084,
      "epoch": 0.0778736002059467,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004946325498182755,
      "loss": 6.1091,
      "mean_token_accuracy": 0.14936656802892684,
      "num_tokens": 2740095.0,
      "step": 1210
    },
    {
      "entropy": 6.198732805252075,
      "epoch": 0.07819539194233492,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004943737504964076,
      "loss": 6.0177,
      "mean_token_accuracy": 0.1528451532125473,
      "num_tokens": 2750443.0,
      "step": 1215
    },
    {
      "entropy": 6.362048006057739,
      "epoch": 0.07851718367872312,
      "grad_norm": 1.0,
      "learning_rate": 0.000494108936932354,
      "loss": 6.0327,
      "mean_token_accuracy": 0.15278265327215196,
      "num_tokens": 2761578.0,
      "step": 1220
    },
    {
      "entropy": 6.290820360183716,
      "epoch": 0.07883897541511134,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004938381163861124,
      "loss": 6.1404,
      "mean_token_accuracy": 0.1517237439751625,
      "num_tokens": 2773410.0,
      "step": 1225
    },
    {
      "entropy": 6.38767557144165,
      "epoch": 0.07916076715149956,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004935612962823645,
      "loss": 6.1139,
      "mean_token_accuracy": 0.14652161300182343,
      "num_tokens": 2784507.0,
      "step": 1230
    },
    {
      "entropy": 6.147325468063355,
      "epoch": 0.07948255888788776,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004932784842102739,
      "loss": 6.0599,
      "mean_token_accuracy": 0.15777070224285125,
      "num_tokens": 2796075.0,
      "step": 1235
    },
    {
      "entropy": 6.353684997558593,
      "epoch": 0.07980435062427597,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004929896879232758,
      "loss": 6.0749,
      "mean_token_accuracy": 0.1501803681254387,
      "num_tokens": 2808124.0,
      "step": 1240
    },
    {
      "entropy": 6.35121603012085,
      "epoch": 0.08012614236066418,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004926949153388668,
      "loss": 6.127,
      "mean_token_accuracy": 0.1444758579134941,
      "num_tokens": 2819010.0,
      "step": 1245
    },
    {
      "entropy": 6.343505907058716,
      "epoch": 0.08044793409705239,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004923941745383859,
      "loss": 6.0505,
      "mean_token_accuracy": 0.15336280614137648,
      "num_tokens": 2829809.0,
      "step": 1250
    },
    {
      "entropy": 6.26406078338623,
      "epoch": 0.0807697258334406,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000492087473766794,
      "loss": 6.058,
      "mean_token_accuracy": 0.1509227141737938,
      "num_tokens": 2841237.0,
      "step": 1255
    },
    {
      "entropy": 6.227289295196533,
      "epoch": 0.08109151756982881,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000491774821432448,
      "loss": 6.0712,
      "mean_token_accuracy": 0.14665706008672713,
      "num_tokens": 2853664.0,
      "step": 1260
    },
    {
      "entropy": 6.25519208908081,
      "epoch": 0.08141330930621701,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004914562261068693,
      "loss": 6.0295,
      "mean_token_accuracy": 0.15138522982597352,
      "num_tokens": 2864821.0,
      "step": 1265
    },
    {
      "entropy": 6.467222690582275,
      "epoch": 0.08173510104260523,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004911316965245098,
      "loss": 6.2744,
      "mean_token_accuracy": 0.1357764020562172,
      "num_tokens": 2878847.0,
      "step": 1270
    },
    {
      "entropy": 6.337312507629394,
      "epoch": 0.08205689277899343,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000490801241582512,
      "loss": 6.1474,
      "mean_token_accuracy": 0.14485098645091057,
      "num_tokens": 2891172.0,
      "step": 1275
    },
    {
      "entropy": 6.462904167175293,
      "epoch": 0.08237868451538165,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000490464870340465,
      "loss": 6.2352,
      "mean_token_accuracy": 0.13823144510388374,
      "num_tokens": 2902218.0,
      "step": 1280
    },
    {
      "entropy": 6.270657682418824,
      "epoch": 0.08270047625176985,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004901225920201563,
      "loss": 6.1159,
      "mean_token_accuracy": 0.14343132004141806,
      "num_tokens": 2914062.0,
      "step": 1285
    },
    {
      "entropy": 6.360012531280518,
      "epoch": 0.08302226798815807,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000489774416005319,
      "loss": 6.091,
      "mean_token_accuracy": 0.1466546781361103,
      "num_tokens": 2924838.0,
      "step": 1290
    },
    {
      "entropy": 6.255312871932984,
      "epoch": 0.08334405972454627,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004894203518413742,
      "loss": 6.0781,
      "mean_token_accuracy": 0.14694482684135438,
      "num_tokens": 2936219.0,
      "step": 1295
    },
    {
      "entropy": 6.416478824615479,
      "epoch": 0.08366585146093448,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004890604092351701,
      "loss": 6.0683,
      "mean_token_accuracy": 0.14561905413866044,
      "num_tokens": 2947195.0,
      "step": 1300
    },
    {
      "entropy": 6.19344310760498,
      "epoch": 0.08398764319732269,
      "grad_norm": 1.09375,
      "learning_rate": 0.000488694598054715,
      "loss": 6.0338,
      "mean_token_accuracy": 0.14701765030622482,
      "num_tokens": 2959322.0,
      "step": 1305
    },
    {
      "entropy": 6.2918178081512455,
      "epoch": 0.0843094349337109,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004883229283289071,
      "loss": 6.0867,
      "mean_token_accuracy": 0.14712531566619874,
      "num_tokens": 2970237.0,
      "step": 1310
    },
    {
      "entropy": 6.260451936721802,
      "epoch": 0.0846312266700991,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048794541024725993,
      "loss": 6.0424,
      "mean_token_accuracy": 0.15119363814592363,
      "num_tokens": 2980701.0,
      "step": 1315
    },
    {
      "entropy": 6.25090856552124,
      "epoch": 0.08495301840648732,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004875620541596221,
      "loss": 5.9198,
      "mean_token_accuracy": 0.1514560803771019,
      "num_tokens": 2992712.0,
      "step": 1320
    },
    {
      "entropy": 6.234271287918091,
      "epoch": 0.08527481014287554,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048717287057589454,
      "loss": 6.0752,
      "mean_token_accuracy": 0.14440764486789703,
      "num_tokens": 3003754.0,
      "step": 1325
    },
    {
      "entropy": 6.255046749114991,
      "epoch": 0.08559660187926374,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004867778701657417,
      "loss": 6.0283,
      "mean_token_accuracy": 0.14669692739844323,
      "num_tokens": 3015715.0,
      "step": 1330
    },
    {
      "entropy": 6.198978328704834,
      "epoch": 0.08591839361565196,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048637706375829955,
      "loss": 6.0125,
      "mean_token_accuracy": 0.15693035572767258,
      "num_tokens": 3026781.0,
      "step": 1335
    },
    {
      "entropy": 6.3291237354278564,
      "epoch": 0.08624018535204016,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000485970462341878,
      "loss": 6.0391,
      "mean_token_accuracy": 0.14689479991793633,
      "num_tokens": 3037856.0,
      "step": 1340
    },
    {
      "entropy": 6.268665742874146,
      "epoch": 0.08656197708842837,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048555807706366044,
      "loss": 6.0847,
      "mean_token_accuracy": 0.1446255125105381,
      "num_tokens": 3050109.0,
      "step": 1345
    },
    {
      "entropy": 6.237447834014892,
      "epoch": 0.08688376882481658,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048513991922939756,
      "loss": 6.0616,
      "mean_token_accuracy": 0.14234503656625747,
      "num_tokens": 3060668.0,
      "step": 1350
    },
    {
      "entropy": 6.198616933822632,
      "epoch": 0.08720556056120479,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048471600030309744,
      "loss": 6.0421,
      "mean_token_accuracy": 0.14418703466653823,
      "num_tokens": 3071176.0,
      "step": 1355
    },
    {
      "entropy": 6.332860994338989,
      "epoch": 0.087527352297593,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048428633190671186,
      "loss": 6.0403,
      "mean_token_accuracy": 0.15211065858602524,
      "num_tokens": 3083180.0,
      "step": 1360
    },
    {
      "entropy": 6.218678760528564,
      "epoch": 0.08784914403398121,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004838509258198167,
      "loss": 5.9538,
      "mean_token_accuracy": 0.15536704435944557,
      "num_tokens": 3093049.0,
      "step": 1365
    },
    {
      "entropy": 6.163859224319458,
      "epoch": 0.08817093577036941,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048340979397929,
      "loss": 5.987,
      "mean_token_accuracy": 0.15278939306735992,
      "num_tokens": 3104979.0,
      "step": 1370
    },
    {
      "entropy": 6.196265602111817,
      "epoch": 0.08849272750675763,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048296294847898386,
      "loss": 5.9735,
      "mean_token_accuracy": 0.1589425176382065,
      "num_tokens": 3117098.0,
      "step": 1375
    },
    {
      "entropy": 6.2933696746826175,
      "epoch": 0.08881451924314583,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004825104015693934,
      "loss": 6.0967,
      "mean_token_accuracy": 0.14706607460975646,
      "num_tokens": 3127554.0,
      "step": 1380
    },
    {
      "entropy": 6.332402229309082,
      "epoch": 0.08913631097953405,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004820521656573208,
      "loss": 6.0697,
      "mean_token_accuracy": 0.15364808216691017,
      "num_tokens": 3139574.0,
      "step": 1385
    },
    {
      "entropy": 6.154071283340454,
      "epoch": 0.08945810271592225,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048158825330553505,
      "loss": 5.9452,
      "mean_token_accuracy": 0.16097584813833238,
      "num_tokens": 3151687.0,
      "step": 1390
    },
    {
      "entropy": 6.355695199966431,
      "epoch": 0.08977989445231047,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048111867723242763,
      "loss": 6.0481,
      "mean_token_accuracy": 0.15232885181903838,
      "num_tokens": 3161667.0,
      "step": 1395
    },
    {
      "entropy": 6.16862416267395,
      "epoch": 0.09010168618869867,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004806434503116637,
      "loss": 6.033,
      "mean_token_accuracy": 0.15161970853805543,
      "num_tokens": 3172797.0,
      "step": 1400
    },
    {
      "entropy": 6.239516496658325,
      "epoch": 0.09042347792508688,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004801625855718296,
      "loss": 5.9882,
      "mean_token_accuracy": 0.15210439413785934,
      "num_tokens": 3183970.0,
      "step": 1405
    },
    {
      "entropy": 6.291398382186889,
      "epoch": 0.09074526966147509,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047967609619607477,
      "loss": 6.0427,
      "mean_token_accuracy": 0.15301772952079773,
      "num_tokens": 3194461.0,
      "step": 1410
    },
    {
      "entropy": 6.2756983757019045,
      "epoch": 0.0910670613978633,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004791839955217513,
      "loss": 5.9568,
      "mean_token_accuracy": 0.15461016148328782,
      "num_tokens": 3204595.0,
      "step": 1415
    },
    {
      "entropy": 6.154314565658569,
      "epoch": 0.09138885313425152,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047868629704004786,
      "loss": 5.9462,
      "mean_token_accuracy": 0.1490132212638855,
      "num_tokens": 3215617.0,
      "step": 1420
    },
    {
      "entropy": 6.124943971633911,
      "epoch": 0.09171064487063972,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047818301439561965,
      "loss": 6.0813,
      "mean_token_accuracy": 0.1486785553395748,
      "num_tokens": 3227350.0,
      "step": 1425
    },
    {
      "entropy": 6.304641628265381,
      "epoch": 0.09203243660702794,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00047767416138621454,
      "loss": 6.2063,
      "mean_token_accuracy": 0.14140090346336365,
      "num_tokens": 3239473.0,
      "step": 1430
    },
    {
      "entropy": 6.324922275543213,
      "epoch": 0.09235422834341614,
      "grad_norm": 1.046875,
      "learning_rate": 0.000477159751962295,
      "loss": 5.9684,
      "mean_token_accuracy": 0.15401543006300927,
      "num_tokens": 3250360.0,
      "step": 1435
    },
    {
      "entropy": 6.12751088142395,
      "epoch": 0.09267602007980436,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047663980022665507,
      "loss": 5.9926,
      "mean_token_accuracy": 0.15175819844007493,
      "num_tokens": 3260227.0,
      "step": 1440
    },
    {
      "entropy": 6.237259864807129,
      "epoch": 0.09299781181619256,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047611432043403437,
      "loss": 5.9543,
      "mean_token_accuracy": 0.151527339220047,
      "num_tokens": 3271316.0,
      "step": 1445
    },
    {
      "entropy": 6.193226385116577,
      "epoch": 0.09331960355258077,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004755833269907267,
      "loss": 6.055,
      "mean_token_accuracy": 0.15415698066353797,
      "num_tokens": 3283194.0,
      "step": 1450
    },
    {
      "entropy": 6.226471662521362,
      "epoch": 0.09364139528896898,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004750468344541857,
      "loss": 5.9492,
      "mean_token_accuracy": 0.15627116858959197,
      "num_tokens": 3294337.0,
      "step": 1455
    },
    {
      "entropy": 6.095775890350342,
      "epoch": 0.09396318702535719,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047450485753262525,
      "loss": 6.0268,
      "mean_token_accuracy": 0.1549006626009941,
      "num_tokens": 3305661.0,
      "step": 1460
    },
    {
      "entropy": 6.316003370285034,
      "epoch": 0.0942849787617454,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047395741108461633,
      "loss": 5.9668,
      "mean_token_accuracy": 0.16676316410303116,
      "num_tokens": 3318457.0,
      "step": 1465
    },
    {
      "entropy": 6.10923490524292,
      "epoch": 0.09460677049813361,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047340451011867985,
      "loss": 5.9276,
      "mean_token_accuracy": 0.1575511649250984,
      "num_tokens": 3329208.0,
      "step": 1470
    },
    {
      "entropy": 6.169835376739502,
      "epoch": 0.09492856223452181,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047284616979287515,
      "loss": 5.9957,
      "mean_token_accuracy": 0.15678192675113678,
      "num_tokens": 3340131.0,
      "step": 1475
    },
    {
      "entropy": 6.147319173812866,
      "epoch": 0.09525035397091003,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047228240541438433,
      "loss": 5.927,
      "mean_token_accuracy": 0.15638188570737838,
      "num_tokens": 3351416.0,
      "step": 1480
    },
    {
      "entropy": 6.111018514633178,
      "epoch": 0.09557214570729823,
      "grad_norm": 0.9375,
      "learning_rate": 0.00047171323243909257,
      "loss": 5.9375,
      "mean_token_accuracy": 0.1552967607975006,
      "num_tokens": 3363734.0,
      "step": 1485
    },
    {
      "entropy": 6.22268123626709,
      "epoch": 0.09589393744368645,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047113866647116457,
      "loss": 6.0156,
      "mean_token_accuracy": 0.15189583972096443,
      "num_tokens": 3375194.0,
      "step": 1490
    },
    {
      "entropy": 6.290840578079224,
      "epoch": 0.09621572918007465,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004705587232626164,
      "loss": 6.1143,
      "mean_token_accuracy": 0.1482478827238083,
      "num_tokens": 3387005.0,
      "step": 1495
    },
    {
      "entropy": 6.177110862731934,
      "epoch": 0.09653752091646287,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046997341871288424,
      "loss": 5.9724,
      "mean_token_accuracy": 0.14636649787425995,
      "num_tokens": 3398902.0,
      "step": 1500
    },
    {
      "entropy": 6.236763906478882,
      "epoch": 0.09685931265285108,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004693827688683879,
      "loss": 5.9133,
      "mean_token_accuracy": 0.15838426500558853,
      "num_tokens": 3410716.0,
      "step": 1505
    },
    {
      "entropy": 6.154586696624756,
      "epoch": 0.09718110438923928,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004687867899220914,
      "loss": 5.961,
      "mean_token_accuracy": 0.15415489226579665,
      "num_tokens": 3422252.0,
      "step": 1510
    },
    {
      "entropy": 6.140632581710816,
      "epoch": 0.0975028961256275,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046818549821305846,
      "loss": 5.8691,
      "mean_token_accuracy": 0.1620650425553322,
      "num_tokens": 3433799.0,
      "step": 1515
    },
    {
      "entropy": 6.097218656539917,
      "epoch": 0.0978246878620157,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046757891022600494,
      "loss": 5.9079,
      "mean_token_accuracy": 0.15895169079303742,
      "num_tokens": 3444754.0,
      "step": 1520
    },
    {
      "entropy": 6.131570863723755,
      "epoch": 0.09814647959840392,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004669670425908471,
      "loss": 5.8836,
      "mean_token_accuracy": 0.15784632563591003,
      "num_tokens": 3456897.0,
      "step": 1525
    },
    {
      "entropy": 6.0702661037445065,
      "epoch": 0.09846827133479212,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004663499120822451,
      "loss": 5.835,
      "mean_token_accuracy": 0.16073963791131973,
      "num_tokens": 3468415.0,
      "step": 1530
    },
    {
      "entropy": 6.077900362014771,
      "epoch": 0.09879006307118034,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004657275356191437,
      "loss": 5.8369,
      "mean_token_accuracy": 0.15783393234014512,
      "num_tokens": 3478914.0,
      "step": 1535
    },
    {
      "entropy": 6.247751140594483,
      "epoch": 0.09911185480756854,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046509993026430804,
      "loss": 6.0328,
      "mean_token_accuracy": 0.15122698694467546,
      "num_tokens": 3490617.0,
      "step": 1540
    },
    {
      "entropy": 6.107704210281372,
      "epoch": 0.09943364654395676,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004644671132238558,
      "loss": 5.9323,
      "mean_token_accuracy": 0.15626108199357985,
      "num_tokens": 3503145.0,
      "step": 1545
    },
    {
      "entropy": 6.2650384426116945,
      "epoch": 0.09975543828034496,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046382910184678585,
      "loss": 5.9544,
      "mean_token_accuracy": 0.1508778765797615,
      "num_tokens": 3514378.0,
      "step": 1550
    },
    {
      "entropy": 6.066193294525147,
      "epoch": 0.10007723001673317,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004631859136245025,
      "loss": 5.9536,
      "mean_token_accuracy": 0.15337267220020295,
      "num_tokens": 3524820.0,
      "step": 1555
    },
    {
      "entropy": 6.2549515724182125,
      "epoch": 0.10039902175312138,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004625375661903357,
      "loss": 6.0194,
      "mean_token_accuracy": 0.15266060531139375,
      "num_tokens": 3537117.0,
      "step": 1560
    },
    {
      "entropy": 6.1690354347229,
      "epoch": 0.10072081348950959,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046188407731905787,
      "loss": 5.9786,
      "mean_token_accuracy": 0.15118198990821838,
      "num_tokens": 3549741.0,
      "step": 1565
    },
    {
      "entropy": 5.9546397686004635,
      "epoch": 0.1010426052258978,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046122546492639643,
      "loss": 5.7664,
      "mean_token_accuracy": 0.16974862962961196,
      "num_tokens": 3560009.0,
      "step": 1570
    },
    {
      "entropy": 6.1606494903564455,
      "epoch": 0.10136439696228601,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000460561747068543,
      "loss": 5.8662,
      "mean_token_accuracy": 0.15560854971408844,
      "num_tokens": 3570005.0,
      "step": 1575
    },
    {
      "entropy": 6.085360765457153,
      "epoch": 0.10168618869867421,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004598929419416578,
      "loss": 5.8876,
      "mean_token_accuracy": 0.1590146005153656,
      "num_tokens": 3581017.0,
      "step": 1580
    },
    {
      "entropy": 6.100130701065064,
      "epoch": 0.10200798043506243,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045921906788137123,
      "loss": 5.8961,
      "mean_token_accuracy": 0.16178194135427476,
      "num_tokens": 3592102.0,
      "step": 1585
    },
    {
      "entropy": 6.079522705078125,
      "epoch": 0.10232977217145063,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045854014336228115,
      "loss": 5.8172,
      "mean_token_accuracy": 0.1636298730969429,
      "num_tokens": 3602860.0,
      "step": 1590
    },
    {
      "entropy": 6.100549745559692,
      "epoch": 0.10265156390783885,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00045785618699744615,
      "loss": 5.8923,
      "mean_token_accuracy": 0.15646297633647918,
      "num_tokens": 3614265.0,
      "step": 1595
    },
    {
      "entropy": 6.137320423126221,
      "epoch": 0.10297335564422706,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045716721753787543,
      "loss": 5.9092,
      "mean_token_accuracy": 0.1642071709036827,
      "num_tokens": 3624783.0,
      "step": 1600
    },
    {
      "entropy": 6.103617238998413,
      "epoch": 0.10329514738061527,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004564732538720148,
      "loss": 5.9007,
      "mean_token_accuracy": 0.15452659055590628,
      "num_tokens": 3636671.0,
      "step": 1605
    },
    {
      "entropy": 6.162163400650025,
      "epoch": 0.10361693911700348,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045577431502522877,
      "loss": 5.9502,
      "mean_token_accuracy": 0.16308289170265197,
      "num_tokens": 3648249.0,
      "step": 1610
    },
    {
      "entropy": 6.2008871078491214,
      "epoch": 0.10393873085339168,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004550704201592787,
      "loss": 5.9927,
      "mean_token_accuracy": 0.1535536527633667,
      "num_tokens": 3659625.0,
      "step": 1615
    },
    {
      "entropy": 6.1465507507324215,
      "epoch": 0.1042605225897799,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004543615885717981,
      "loss": 5.9086,
      "mean_token_accuracy": 0.160139761865139,
      "num_tokens": 3670233.0,
      "step": 1620
    },
    {
      "entropy": 6.0907965183258055,
      "epoch": 0.1045823143261681,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045364783969576296,
      "loss": 5.8771,
      "mean_token_accuracy": 0.1571106120944023,
      "num_tokens": 3680793.0,
      "step": 1625
    },
    {
      "entropy": 6.108975076675415,
      "epoch": 0.10490410606255632,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004529291930989592,
      "loss": 5.883,
      "mean_token_accuracy": 0.17130757123231888,
      "num_tokens": 3692561.0,
      "step": 1630
    },
    {
      "entropy": 6.049561834335327,
      "epoch": 0.10522589779894452,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004522056684834464,
      "loss": 5.887,
      "mean_token_accuracy": 0.1572432816028595,
      "num_tokens": 3704306.0,
      "step": 1635
    },
    {
      "entropy": 6.081239366531372,
      "epoch": 0.10554768953533274,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004514772856850173,
      "loss": 5.8848,
      "mean_token_accuracy": 0.16130839735269548,
      "num_tokens": 3715276.0,
      "step": 1640
    },
    {
      "entropy": 6.184744596481323,
      "epoch": 0.10586948127172094,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004507440646726542,
      "loss": 6.025,
      "mean_token_accuracy": 0.1553824484348297,
      "num_tokens": 3726799.0,
      "step": 1645
    },
    {
      "entropy": 6.106659936904907,
      "epoch": 0.10619127300810915,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004500060255479818,
      "loss": 5.9149,
      "mean_token_accuracy": 0.15892969965934753,
      "num_tokens": 3736648.0,
      "step": 1650
    },
    {
      "entropy": 6.1206361770629885,
      "epoch": 0.10651306474449736,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004492631885447151,
      "loss": 5.9212,
      "mean_token_accuracy": 0.1520114503800869,
      "num_tokens": 3746940.0,
      "step": 1655
    },
    {
      "entropy": 6.142186498641967,
      "epoch": 0.10683485648088557,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044851557402810616,
      "loss": 5.86,
      "mean_token_accuracy": 0.15836223363876342,
      "num_tokens": 3758099.0,
      "step": 1660
    },
    {
      "entropy": 5.9857617855072025,
      "epoch": 0.10715664821727378,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044776320249438444,
      "loss": 5.8378,
      "mean_token_accuracy": 0.16005610525608063,
      "num_tokens": 3769881.0,
      "step": 1665
    },
    {
      "entropy": 6.105796766281128,
      "epoch": 0.10747843995366199,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044700609457019565,
      "loss": 5.8723,
      "mean_token_accuracy": 0.1569559782743454,
      "num_tokens": 3780925.0,
      "step": 1670
    },
    {
      "entropy": 6.123667669296265,
      "epoch": 0.1078002316900502,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004462442710120359,
      "loss": 5.978,
      "mean_token_accuracy": 0.15348947197198867,
      "num_tokens": 3791941.0,
      "step": 1675
    },
    {
      "entropy": 6.140316152572632,
      "epoch": 0.10812202342643841,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000445477752705683,
      "loss": 5.9025,
      "mean_token_accuracy": 0.15857508778572083,
      "num_tokens": 3803698.0,
      "step": 1680
    },
    {
      "entropy": 6.062806749343872,
      "epoch": 0.10844381516282661,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00044470656066562336,
      "loss": 5.8514,
      "mean_token_accuracy": 0.1542954534292221,
      "num_tokens": 3815681.0,
      "step": 1685
    },
    {
      "entropy": 6.0847492694854735,
      "epoch": 0.10876560689921483,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004439307160344765,
      "loss": 5.8906,
      "mean_token_accuracy": 0.15737924128770828,
      "num_tokens": 3828074.0,
      "step": 1690
    },
    {
      "entropy": 6.040540266036987,
      "epoch": 0.10908739863560304,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044315024008241473,
      "loss": 5.8193,
      "mean_token_accuracy": 0.16169758141040802,
      "num_tokens": 3838928.0,
      "step": 1695
    },
    {
      "entropy": 6.152355909347534,
      "epoch": 0.10940919037199125,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004423651542065806,
      "loss": 5.972,
      "mean_token_accuracy": 0.14924763143062592,
      "num_tokens": 3850148.0,
      "step": 1700
    },
    {
      "entropy": 6.210541772842407,
      "epoch": 0.10973098210837946,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044157547993050006,
      "loss": 5.9284,
      "mean_token_accuracy": 0.15215876325964928,
      "num_tokens": 3861099.0,
      "step": 1705
    },
    {
      "entropy": 6.108062696456909,
      "epoch": 0.11005277384476767,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044078123890349227,
      "loss": 5.8861,
      "mean_token_accuracy": 0.16254641860723495,
      "num_tokens": 3872742.0,
      "step": 1710
    },
    {
      "entropy": 6.034341526031494,
      "epoch": 0.11037456558115588,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043998245290007606,
      "loss": 5.8189,
      "mean_token_accuracy": 0.16008612066507338,
      "num_tokens": 3883573.0,
      "step": 1715
    },
    {
      "entropy": 6.092084360122681,
      "epoch": 0.11069635731754408,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043917914381937323,
      "loss": 5.8707,
      "mean_token_accuracy": 0.15305377990007402,
      "num_tokens": 3894614.0,
      "step": 1720
    },
    {
      "entropy": 5.947681093215943,
      "epoch": 0.1110181490539323,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00043837133368450815,
      "loss": 5.6755,
      "mean_token_accuracy": 0.17631727159023286,
      "num_tokens": 3904217.0,
      "step": 1725
    },
    {
      "entropy": 5.988527441024781,
      "epoch": 0.1113399407903205,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004375590446420037,
      "loss": 5.8423,
      "mean_token_accuracy": 0.15377960056066514,
      "num_tokens": 3916066.0,
      "step": 1730
    },
    {
      "entropy": 6.192314147949219,
      "epoch": 0.11166173252670872,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004367422989611743,
      "loss": 5.8629,
      "mean_token_accuracy": 0.16659451127052308,
      "num_tokens": 3928106.0,
      "step": 1735
    },
    {
      "entropy": 5.978959846496582,
      "epoch": 0.11198352426309692,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004359211190335153,
      "loss": 5.7561,
      "mean_token_accuracy": 0.16876690536737443,
      "num_tokens": 3940176.0,
      "step": 1740
    },
    {
      "entropy": 5.962417316436768,
      "epoch": 0.11230531599948514,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043509552737208923,
      "loss": 5.8668,
      "mean_token_accuracy": 0.1604593113064766,
      "num_tokens": 3951938.0,
      "step": 1745
    },
    {
      "entropy": 6.1542033672332765,
      "epoch": 0.11262710773587334,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00043426554661090853,
      "loss": 5.8896,
      "mean_token_accuracy": 0.15797159373760222,
      "num_tokens": 3962731.0,
      "step": 1750
    },
    {
      "entropy": 6.078965711593628,
      "epoch": 0.11294889947226155,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00043343119950431516,
      "loss": 5.8989,
      "mean_token_accuracy": 0.1685984030365944,
      "num_tokens": 3973841.0,
      "step": 1755
    },
    {
      "entropy": 6.0909887790679935,
      "epoch": 0.11327069120864976,
      "grad_norm": 0.90625,
      "learning_rate": 0.00043259250892635644,
      "loss": 5.849,
      "mean_token_accuracy": 0.15423968136310579,
      "num_tokens": 3985723.0,
      "step": 1760
    },
    {
      "entropy": 5.981694746017456,
      "epoch": 0.11359248294503797,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004317494978701582,
      "loss": 5.7654,
      "mean_token_accuracy": 0.16921489387750627,
      "num_tokens": 3996719.0,
      "step": 1765
    },
    {
      "entropy": 6.0790181159973145,
      "epoch": 0.11391427468142618,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004309021894472943,
      "loss": 5.8408,
      "mean_token_accuracy": 0.1637660652399063,
      "num_tokens": 4007178.0,
      "step": 1770
    },
    {
      "entropy": 6.193857336044312,
      "epoch": 0.11423606641781439,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004300506068871534,
      "loss": 5.9561,
      "mean_token_accuracy": 0.15705759078264236,
      "num_tokens": 4017441.0,
      "step": 1775
    },
    {
      "entropy": 6.0972432613372805,
      "epoch": 0.1145578581542026,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00042919477353630135,
      "loss": 5.9267,
      "mean_token_accuracy": 0.15945902168750764,
      "num_tokens": 4028902.0,
      "step": 1780
    },
    {
      "entropy": 6.088161182403565,
      "epoch": 0.11487964989059081,
      "grad_norm": 1.078125,
      "learning_rate": 0.000428334712857842,
      "loss": 5.8481,
      "mean_token_accuracy": 0.15834707170724868,
      "num_tokens": 4039686.0,
      "step": 1785
    },
    {
      "entropy": 6.092359447479248,
      "epoch": 0.11520144162697903,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042747044843077304,
      "loss": 5.8735,
      "mean_token_accuracy": 0.15396366342902185,
      "num_tokens": 4052367.0,
      "step": 1790
    },
    {
      "entropy": 6.024484586715698,
      "epoch": 0.11552323336336723,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042660200394934047,
      "loss": 5.7914,
      "mean_token_accuracy": 0.15892019718885422,
      "num_tokens": 4063200.0,
      "step": 1795
    },
    {
      "entropy": 6.0296402931213375,
      "epoch": 0.11584502509975544,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042572940322238844,
      "loss": 5.7621,
      "mean_token_accuracy": 0.16620553135871888,
      "num_tokens": 4074896.0,
      "step": 1800
    },
    {
      "entropy": 6.014833593368531,
      "epoch": 0.11616681683614365,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00042485267017270664,
      "loss": 5.717,
      "mean_token_accuracy": 0.16428265124559402,
      "num_tokens": 4086622.0,
      "step": 1805
    },
    {
      "entropy": 5.952508020401001,
      "epoch": 0.11648860857253186,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004239718288363745,
      "loss": 5.8441,
      "mean_token_accuracy": 0.16071850210428237,
      "num_tokens": 4097172.0,
      "step": 1810
    },
    {
      "entropy": 6.073939561843872,
      "epoch": 0.11681040030892006,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004230869033621023,
      "loss": 5.8946,
      "mean_token_accuracy": 0.1668115332722664,
      "num_tokens": 4108735.0,
      "step": 1815
    },
    {
      "entropy": 6.041821050643921,
      "epoch": 0.11713219204530828,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004221979180105688,
      "loss": 5.837,
      "mean_token_accuracy": 0.1618662878870964,
      "num_tokens": 4119840.0,
      "step": 1820
    },
    {
      "entropy": 6.002658462524414,
      "epoch": 0.11745398378169648,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00042130489715375645,
      "loss": 5.8832,
      "mean_token_accuracy": 0.15968891829252244,
      "num_tokens": 4131999.0,
      "step": 1825
    },
    {
      "entropy": 6.1330491065979,
      "epoch": 0.1177757755180847,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00042040786527428335,
      "loss": 5.7651,
      "mean_token_accuracy": 0.16318420171737671,
      "num_tokens": 4143516.0,
      "step": 1830
    },
    {
      "entropy": 5.996764755249023,
      "epoch": 0.1180975672544729,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004195068469647315,
      "loss": 5.8386,
      "mean_token_accuracy": 0.15657123625278474,
      "num_tokens": 4153963.0,
      "step": 1835
    },
    {
      "entropy": 5.976077747344971,
      "epoch": 0.11841935899086112,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041860186692697297,
      "loss": 5.7268,
      "mean_token_accuracy": 0.1676751896739006,
      "num_tokens": 4165361.0,
      "step": 1840
    },
    {
      "entropy": 5.975963687896728,
      "epoch": 0.11874115072724932,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041769294997149264,
      "loss": 5.75,
      "mean_token_accuracy": 0.16740206331014634,
      "num_tokens": 4176608.0,
      "step": 1845
    },
    {
      "entropy": 5.930391788482666,
      "epoch": 0.11906294246363754,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004167801210167081,
      "loss": 5.7926,
      "mean_token_accuracy": 0.16247987747192383,
      "num_tokens": 4188709.0,
      "step": 1850
    },
    {
      "entropy": 6.0283245086669925,
      "epoch": 0.11938473420002574,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004158634050882861,
      "loss": 5.8113,
      "mean_token_accuracy": 0.1592404916882515,
      "num_tokens": 4199079.0,
      "step": 1855
    },
    {
      "entropy": 6.08579740524292,
      "epoch": 0.11970652593641395,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004149428273184569,
      "loss": 5.8994,
      "mean_token_accuracy": 0.1571430966258049,
      "num_tokens": 4209204.0,
      "step": 1860
    },
    {
      "entropy": 6.115895318984985,
      "epoch": 0.12002831767280216,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004140184129453253,
      "loss": 5.896,
      "mean_token_accuracy": 0.15317915976047516,
      "num_tokens": 4220270.0,
      "step": 1865
    },
    {
      "entropy": 5.977196741104126,
      "epoch": 0.12035010940919037,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000413090187312178,
      "loss": 5.7498,
      "mean_token_accuracy": 0.16569025069475174,
      "num_tokens": 4232996.0,
      "step": 1870
    },
    {
      "entropy": 5.974374628067016,
      "epoch": 0.12067190114557858,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004121581758667898,
      "loss": 5.7578,
      "mean_token_accuracy": 0.1671750284731388,
      "num_tokens": 4244089.0,
      "step": 1875
    },
    {
      "entropy": 6.00874342918396,
      "epoch": 0.12099369288196679,
      "grad_norm": 0.953125,
      "learning_rate": 0.00041122240416072533,
      "loss": 5.7989,
      "mean_token_accuracy": 0.16003866046667098,
      "num_tokens": 4254939.0,
      "step": 1880
    },
    {
      "entropy": 6.073114728927612,
      "epoch": 0.12131548461835501,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004102828978486385,
      "loss": 5.8726,
      "mean_token_accuracy": 0.15603156834840776,
      "num_tokens": 4266429.0,
      "step": 1885
    },
    {
      "entropy": 5.972569847106934,
      "epoch": 0.12163727635474321,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004093396826875695,
      "loss": 5.7964,
      "mean_token_accuracy": 0.15974399149417878,
      "num_tokens": 4278063.0,
      "step": 1890
    },
    {
      "entropy": 6.0056493282318115,
      "epoch": 0.12195906809113143,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040839278453623837,
      "loss": 5.8105,
      "mean_token_accuracy": 0.16451311111450195,
      "num_tokens": 4290281.0,
      "step": 1895
    },
    {
      "entropy": 6.038688039779663,
      "epoch": 0.12228085982751963,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004074422293543363,
      "loss": 5.804,
      "mean_token_accuracy": 0.15884712487459182,
      "num_tokens": 4301281.0,
      "step": 1900
    },
    {
      "entropy": 6.016970539093018,
      "epoch": 0.12260265156390784,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004064880432018137,
      "loss": 5.7658,
      "mean_token_accuracy": 0.15432919561862946,
      "num_tokens": 4312406.0,
      "step": 1905
    },
    {
      "entropy": 6.103054714202881,
      "epoch": 0.12292444330029605,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00040553025223816615,
      "loss": 5.8855,
      "mean_token_accuracy": 0.15506619587540627,
      "num_tokens": 4322831.0,
      "step": 1910
    },
    {
      "entropy": 6.005894613265991,
      "epoch": 0.12324623503668426,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00040456888272171653,
      "loss": 5.8042,
      "mean_token_accuracy": 0.1634931042790413,
      "num_tokens": 4334850.0,
      "step": 1915
    },
    {
      "entropy": 6.033585166931152,
      "epoch": 0.12356802677307246,
      "grad_norm": 0.859375,
      "learning_rate": 0.00040360396100889577,
      "loss": 5.8797,
      "mean_token_accuracy": 0.1673731878399849,
      "num_tokens": 4346913.0,
      "step": 1920
    },
    {
      "entropy": 6.02968282699585,
      "epoch": 0.12388981850946068,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004026355135535202,
      "loss": 5.7898,
      "mean_token_accuracy": 0.16513559669256211,
      "num_tokens": 4357030.0,
      "step": 1925
    },
    {
      "entropy": 5.9573445320129395,
      "epoch": 0.12421161024584888,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000401663566906066,
      "loss": 5.7356,
      "mean_token_accuracy": 0.16273822635412216,
      "num_tokens": 4368562.0,
      "step": 1930
    },
    {
      "entropy": 6.075269746780395,
      "epoch": 0.1245334019822371,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00040068814771294134,
      "loss": 5.8348,
      "mean_token_accuracy": 0.1631076753139496,
      "num_tokens": 4380499.0,
      "step": 1935
    },
    {
      "entropy": 5.973206567764282,
      "epoch": 0.1248551937186253,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003997092827157562,
      "loss": 5.6817,
      "mean_token_accuracy": 0.17137495726346968,
      "num_tokens": 4391288.0,
      "step": 1940
    },
    {
      "entropy": 6.008097457885742,
      "epoch": 0.12517698545501352,
      "grad_norm": 1.015625,
      "learning_rate": 0.000398726998750589,
      "loss": 5.8549,
      "mean_token_accuracy": 0.1622577652335167,
      "num_tokens": 4402371.0,
      "step": 1945
    },
    {
      "entropy": 6.058371210098267,
      "epoch": 0.12549877719140173,
      "grad_norm": 0.953125,
      "learning_rate": 0.00039774132274725076,
      "loss": 5.8899,
      "mean_token_accuracy": 0.16455032378435136,
      "num_tokens": 4414095.0,
      "step": 1950
    },
    {
      "entropy": 5.990140342712403,
      "epoch": 0.12582056892778992,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039675228172854707,
      "loss": 5.8451,
      "mean_token_accuracy": 0.1557239845395088,
      "num_tokens": 4425878.0,
      "step": 1955
    },
    {
      "entropy": 6.079447650909424,
      "epoch": 0.12614236066417814,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003957599028095371,
      "loss": 5.7468,
      "mean_token_accuracy": 0.16118832379579545,
      "num_tokens": 4437387.0,
      "step": 1960
    },
    {
      "entropy": 6.004099607467651,
      "epoch": 0.12646415240056635,
      "grad_norm": 0.953125,
      "learning_rate": 0.00039476421319679017,
      "loss": 5.7359,
      "mean_token_accuracy": 0.1670518174767494,
      "num_tokens": 4448457.0,
      "step": 1965
    },
    {
      "entropy": 6.00559983253479,
      "epoch": 0.12678594413695457,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00039376524018764,
      "loss": 5.7616,
      "mean_token_accuracy": 0.17010354697704316,
      "num_tokens": 4460368.0,
      "step": 1970
    },
    {
      "entropy": 5.861134004592896,
      "epoch": 0.12710773587334276,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039276301116943616,
      "loss": 5.7011,
      "mean_token_accuracy": 0.16875038146972657,
      "num_tokens": 4472436.0,
      "step": 1975
    },
    {
      "entropy": 5.954335880279541,
      "epoch": 0.12742952760973097,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003917575536187936,
      "loss": 5.745,
      "mean_token_accuracy": 0.16061961352825166,
      "num_tokens": 4483945.0,
      "step": 1980
    },
    {
      "entropy": 5.945883464813233,
      "epoch": 0.1277513193461192,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039074889510083894,
      "loss": 5.7885,
      "mean_token_accuracy": 0.1619923420250416,
      "num_tokens": 4494694.0,
      "step": 1985
    },
    {
      "entropy": 6.025341606140136,
      "epoch": 0.1280731110825074,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038973706326845495,
      "loss": 5.8042,
      "mean_token_accuracy": 0.16122044026851653,
      "num_tokens": 4506454.0,
      "step": 1990
    },
    {
      "entropy": 6.062008476257324,
      "epoch": 0.12839490281889562,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003887220858615225,
      "loss": 5.7555,
      "mean_token_accuracy": 0.168570613861084,
      "num_tokens": 4516921.0,
      "step": 1995
    },
    {
      "entropy": 5.976399898529053,
      "epoch": 0.1287166945552838,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003877039907061597,
      "loss": 5.7396,
      "mean_token_accuracy": 0.16865382939577103,
      "num_tokens": 4527936.0,
      "step": 2000
    }
  ],
  "logging_steps": 5,
  "max_steps": 4000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1065543491911680.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}