rus-cyrl-10mb-after-ppt-Dp-100mb-ckpt500_seed3407/checkpoint-12000/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0486760464913047,
  "eval_steps": 3000,
  "global_step": 12000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 7.897289562225342,
      "epoch": 0.0004369483527047103,
      "grad_norm": 0.890625,
      "learning_rate": 2e-06,
      "loss": 7.6042,
      "mean_token_accuracy": 0.08350983113050461,
      "num_tokens": 10478.0,
      "step": 5
    },
    {
      "entropy": 7.906166076660156,
      "epoch": 0.0008738967054094206,
      "grad_norm": 0.76953125,
      "learning_rate": 4.5e-06,
      "loss": 7.6188,
      "mean_token_accuracy": 0.08587298914790154,
      "num_tokens": 21152.0,
      "step": 10
    },
    {
      "entropy": 7.886682033538818,
      "epoch": 0.0013108450581141308,
      "grad_norm": 0.984375,
      "learning_rate": 7e-06,
      "loss": 7.6304,
      "mean_token_accuracy": 0.08478806018829346,
      "num_tokens": 31246.0,
      "step": 15
    },
    {
      "entropy": 7.918443632125855,
      "epoch": 0.0017477934108188413,
      "grad_norm": 0.82421875,
      "learning_rate": 9.5e-06,
      "loss": 7.6787,
      "mean_token_accuracy": 0.07867336831986904,
      "num_tokens": 41934.0,
      "step": 20
    },
    {
      "entropy": 7.970422983169556,
      "epoch": 0.0021847417635235517,
      "grad_norm": 0.79296875,
      "learning_rate": 1.2e-05,
      "loss": 7.6469,
      "mean_token_accuracy": 0.08144963085651398,
      "num_tokens": 52830.0,
      "step": 25
    },
    {
      "entropy": 7.949476909637451,
      "epoch": 0.0026216901162282617,
      "grad_norm": 0.828125,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 7.5996,
      "mean_token_accuracy": 0.08676403239369393,
      "num_tokens": 62706.0,
      "step": 30
    },
    {
      "entropy": 7.977545309066772,
      "epoch": 0.003058638468932972,
      "grad_norm": 0.86328125,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 7.6506,
      "mean_token_accuracy": 0.086336649954319,
      "num_tokens": 72619.0,
      "step": 35
    },
    {
      "entropy": 8.052342128753661,
      "epoch": 0.0034955868216376825,
      "grad_norm": 1.1484375,
      "learning_rate": 1.95e-05,
      "loss": 7.6618,
      "mean_token_accuracy": 0.08166376948356628,
      "num_tokens": 82903.0,
      "step": 40
    },
    {
      "entropy": 8.004082489013673,
      "epoch": 0.003932535174342393,
      "grad_norm": 0.796875,
      "learning_rate": 2.2e-05,
      "loss": 7.6376,
      "mean_token_accuracy": 0.09148860350251198,
      "num_tokens": 94243.0,
      "step": 45
    },
    {
      "entropy": 7.982316017150879,
      "epoch": 0.004369483527047103,
      "grad_norm": 0.8671875,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 7.5827,
      "mean_token_accuracy": 0.08447664603590965,
      "num_tokens": 103618.0,
      "step": 50
    },
    {
      "entropy": 7.994159317016601,
      "epoch": 0.004806431879751813,
      "grad_norm": 0.8359375,
      "learning_rate": 2.7e-05,
      "loss": 7.6458,
      "mean_token_accuracy": 0.0843130737543106,
      "num_tokens": 112180.0,
      "step": 55
    },
    {
      "entropy": 7.965240716934204,
      "epoch": 0.005243380232456523,
      "grad_norm": 0.87890625,
      "learning_rate": 2.95e-05,
      "loss": 7.6378,
      "mean_token_accuracy": 0.0797391913831234,
      "num_tokens": 122462.0,
      "step": 60
    },
    {
      "entropy": 7.984555625915528,
      "epoch": 0.005680328585161234,
      "grad_norm": 0.82421875,
      "learning_rate": 3.2e-05,
      "loss": 7.675,
      "mean_token_accuracy": 0.07956810221076012,
      "num_tokens": 132741.0,
      "step": 65
    },
    {
      "entropy": 8.007879114151,
      "epoch": 0.006117276937865944,
      "grad_norm": 0.828125,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 7.4766,
      "mean_token_accuracy": 0.09351350888609886,
      "num_tokens": 142973.0,
      "step": 70
    },
    {
      "entropy": 8.00863881111145,
      "epoch": 0.006554225290570655,
      "grad_norm": 0.89453125,
      "learning_rate": 3.7e-05,
      "loss": 7.7046,
      "mean_token_accuracy": 0.08399961739778519,
      "num_tokens": 153199.0,
      "step": 75
    },
    {
      "entropy": 7.956274366378784,
      "epoch": 0.006991173643275365,
      "grad_norm": 0.890625,
      "learning_rate": 3.95e-05,
      "loss": 7.5526,
      "mean_token_accuracy": 0.08937041833996773,
      "num_tokens": 163079.0,
      "step": 80
    },
    {
      "entropy": 7.965800809860229,
      "epoch": 0.0074281219959800755,
      "grad_norm": 0.82421875,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 7.5914,
      "mean_token_accuracy": 0.08658739626407623,
      "num_tokens": 173308.0,
      "step": 85
    },
    {
      "entropy": 7.9384160995483395,
      "epoch": 0.007865070348684786,
      "grad_norm": 0.9375,
      "learning_rate": 4.45e-05,
      "loss": 7.6535,
      "mean_token_accuracy": 0.09069974273443222,
      "num_tokens": 183856.0,
      "step": 90
    },
    {
      "entropy": 7.94856538772583,
      "epoch": 0.008302018701389496,
      "grad_norm": 0.87109375,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 7.5749,
      "mean_token_accuracy": 0.08630532398819923,
      "num_tokens": 193368.0,
      "step": 95
    },
    {
      "entropy": 7.945826864242553,
      "epoch": 0.008738967054094207,
      "grad_norm": 0.8046875,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 7.5981,
      "mean_token_accuracy": 0.08861325606703759,
      "num_tokens": 202686.0,
      "step": 100
    },
    {
      "entropy": 7.925296783447266,
      "epoch": 0.009175915406798917,
      "grad_norm": 0.89453125,
      "learning_rate": 5.2e-05,
      "loss": 7.5568,
      "mean_token_accuracy": 0.09273463487625122,
      "num_tokens": 213237.0,
      "step": 105
    },
    {
      "entropy": 7.953626537322998,
      "epoch": 0.009612863759503626,
      "grad_norm": 0.8515625,
      "learning_rate": 5.45e-05,
      "loss": 7.5719,
      "mean_token_accuracy": 0.08897444158792496,
      "num_tokens": 222406.0,
      "step": 110
    },
    {
      "entropy": 7.976461410522461,
      "epoch": 0.010049812112208336,
      "grad_norm": 0.96875,
      "learning_rate": 5.7e-05,
      "loss": 7.6179,
      "mean_token_accuracy": 0.08532986417412758,
      "num_tokens": 232139.0,
      "step": 115
    },
    {
      "entropy": 7.922226190567017,
      "epoch": 0.010486760464913047,
      "grad_norm": 0.91796875,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 7.5876,
      "mean_token_accuracy": 0.08453975692391395,
      "num_tokens": 242438.0,
      "step": 120
    },
    {
      "entropy": 7.840370607376099,
      "epoch": 0.010923708817617757,
      "grad_norm": 0.8828125,
      "learning_rate": 6.2e-05,
      "loss": 7.589,
      "mean_token_accuracy": 0.08821919411420823,
      "num_tokens": 253082.0,
      "step": 125
    },
    {
      "entropy": 7.940424108505249,
      "epoch": 0.011360657170322468,
      "grad_norm": 1.0,
      "learning_rate": 6.450000000000001e-05,
      "loss": 7.5948,
      "mean_token_accuracy": 0.0919061541557312,
      "num_tokens": 264067.0,
      "step": 130
    },
    {
      "entropy": 8.062803173065186,
      "epoch": 0.011797605523027178,
      "grad_norm": 0.82421875,
      "learning_rate": 6.7e-05,
      "loss": 7.5772,
      "mean_token_accuracy": 0.08630142137408256,
      "num_tokens": 274571.0,
      "step": 135
    },
    {
      "entropy": 7.8331855773925785,
      "epoch": 0.012234553875731888,
      "grad_norm": 0.79296875,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.5644,
      "mean_token_accuracy": 0.09022602662444115,
      "num_tokens": 284561.0,
      "step": 140
    },
    {
      "entropy": 7.9677777767181395,
      "epoch": 0.012671502228436599,
      "grad_norm": 0.81640625,
      "learning_rate": 7.2e-05,
      "loss": 7.4588,
      "mean_token_accuracy": 0.08896260559558869,
      "num_tokens": 295366.0,
      "step": 145
    },
    {
      "entropy": 7.901757526397705,
      "epoch": 0.01310845058114131,
      "grad_norm": 1.4453125,
      "learning_rate": 7.45e-05,
      "loss": 7.6533,
      "mean_token_accuracy": 0.08677535206079483,
      "num_tokens": 305423.0,
      "step": 150
    },
    {
      "entropy": 7.8698560237884525,
      "epoch": 0.01354539893384602,
      "grad_norm": 0.8515625,
      "learning_rate": 7.7e-05,
      "loss": 7.5374,
      "mean_token_accuracy": 0.09062277376651764,
      "num_tokens": 315182.0,
      "step": 155
    },
    {
      "entropy": 7.939631605148316,
      "epoch": 0.01398234728655073,
      "grad_norm": 0.8984375,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.5551,
      "mean_token_accuracy": 0.08309731036424636,
      "num_tokens": 324780.0,
      "step": 160
    },
    {
      "entropy": 7.941270208358764,
      "epoch": 0.01441929563925544,
      "grad_norm": 0.828125,
      "learning_rate": 8.2e-05,
      "loss": 7.617,
      "mean_token_accuracy": 0.08777398616075516,
      "num_tokens": 335104.0,
      "step": 165
    },
    {
      "entropy": 7.986670398712159,
      "epoch": 0.014856243991960151,
      "grad_norm": 1.0078125,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.6081,
      "mean_token_accuracy": 0.0891649942845106,
      "num_tokens": 345073.0,
      "step": 170
    },
    {
      "entropy": 8.023071241378783,
      "epoch": 0.015293192344664861,
      "grad_norm": 0.94140625,
      "learning_rate": 8.7e-05,
      "loss": 7.6144,
      "mean_token_accuracy": 0.08408513441681861,
      "num_tokens": 354342.0,
      "step": 175
    },
    {
      "entropy": 7.945475244522095,
      "epoch": 0.015730140697369572,
      "grad_norm": 0.8828125,
      "learning_rate": 8.95e-05,
      "loss": 7.5353,
      "mean_token_accuracy": 0.09062551632523537,
      "num_tokens": 364441.0,
      "step": 180
    },
    {
      "entropy": 7.928923225402832,
      "epoch": 0.01616708905007428,
      "grad_norm": 0.90625,
      "learning_rate": 9.2e-05,
      "loss": 7.5542,
      "mean_token_accuracy": 0.08356907293200493,
      "num_tokens": 373918.0,
      "step": 185
    },
    {
      "entropy": 7.914482545852661,
      "epoch": 0.016604037402778993,
      "grad_norm": 0.921875,
      "learning_rate": 9.45e-05,
      "loss": 7.6305,
      "mean_token_accuracy": 0.08665620759129525,
      "num_tokens": 383273.0,
      "step": 190
    },
    {
      "entropy": 7.877908754348755,
      "epoch": 0.0170409857554837,
      "grad_norm": 0.98046875,
      "learning_rate": 9.7e-05,
      "loss": 7.4795,
      "mean_token_accuracy": 0.09727039858698845,
      "num_tokens": 393262.0,
      "step": 195
    },
    {
      "entropy": 7.961901712417602,
      "epoch": 0.017477934108188414,
      "grad_norm": 1.0,
      "learning_rate": 9.95e-05,
      "loss": 7.6166,
      "mean_token_accuracy": 0.08374508917331695,
      "num_tokens": 402389.0,
      "step": 200
    },
    {
      "entropy": 7.935757064819336,
      "epoch": 0.017914882460893122,
      "grad_norm": 0.953125,
      "learning_rate": 0.000102,
      "loss": 7.5619,
      "mean_token_accuracy": 0.0888036273419857,
      "num_tokens": 412997.0,
      "step": 205
    },
    {
      "entropy": 7.9856915950775145,
      "epoch": 0.018351830813597834,
      "grad_norm": 1.03125,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.5661,
      "mean_token_accuracy": 0.08725668415427208,
      "num_tokens": 422222.0,
      "step": 210
    },
    {
      "entropy": 7.811623525619507,
      "epoch": 0.018788779166302543,
      "grad_norm": 0.8203125,
      "learning_rate": 0.000107,
      "loss": 7.5068,
      "mean_token_accuracy": 0.09225667044520378,
      "num_tokens": 432539.0,
      "step": 215
    },
    {
      "entropy": 7.9459527969360355,
      "epoch": 0.019225727519007252,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0001095,
      "loss": 7.5453,
      "mean_token_accuracy": 0.08999547138810157,
      "num_tokens": 442311.0,
      "step": 220
    },
    {
      "entropy": 7.893007898330689,
      "epoch": 0.019662675871711964,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000112,
      "loss": 7.5167,
      "mean_token_accuracy": 0.08521599024534225,
      "num_tokens": 452727.0,
      "step": 225
    },
    {
      "entropy": 7.965888023376465,
      "epoch": 0.020099624224416673,
      "grad_norm": 1.03125,
      "learning_rate": 0.0001145,
      "loss": 7.5489,
      "mean_token_accuracy": 0.0897920735180378,
      "num_tokens": 462461.0,
      "step": 230
    },
    {
      "entropy": 7.924414014816284,
      "epoch": 0.020536572577121385,
      "grad_norm": 1.03125,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.5427,
      "mean_token_accuracy": 0.0894696906208992,
      "num_tokens": 472362.0,
      "step": 235
    },
    {
      "entropy": 7.880691957473755,
      "epoch": 0.020973520929826094,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.5639,
      "mean_token_accuracy": 0.08265788778662682,
      "num_tokens": 482718.0,
      "step": 240
    },
    {
      "entropy": 7.892111539840698,
      "epoch": 0.021410469282530806,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000122,
      "loss": 7.5248,
      "mean_token_accuracy": 0.08400834277272225,
      "num_tokens": 492947.0,
      "step": 245
    },
    {
      "entropy": 7.998752784729004,
      "epoch": 0.021847417635235514,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0001245,
      "loss": 7.5869,
      "mean_token_accuracy": 0.08245678767561912,
      "num_tokens": 502957.0,
      "step": 250
    },
    {
      "entropy": 7.905310440063476,
      "epoch": 0.022284365987940227,
      "grad_norm": 1.0625,
      "learning_rate": 0.000127,
      "loss": 7.5414,
      "mean_token_accuracy": 0.08397668749094009,
      "num_tokens": 513220.0,
      "step": 255
    },
    {
      "entropy": 7.939359521865844,
      "epoch": 0.022721314340644935,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0001295,
      "loss": 7.5046,
      "mean_token_accuracy": 0.0867614358663559,
      "num_tokens": 523217.0,
      "step": 260
    },
    {
      "entropy": 7.874382209777832,
      "epoch": 0.023158262693349647,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000132,
      "loss": 7.5321,
      "mean_token_accuracy": 0.08652917370200157,
      "num_tokens": 533789.0,
      "step": 265
    },
    {
      "entropy": 7.985016393661499,
      "epoch": 0.023595211046054356,
      "grad_norm": 0.9375,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.4971,
      "mean_token_accuracy": 0.09434628114104271,
      "num_tokens": 544146.0,
      "step": 270
    },
    {
      "entropy": 7.840573167800903,
      "epoch": 0.02403215939875907,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.5389,
      "mean_token_accuracy": 0.0874963328242302,
      "num_tokens": 554215.0,
      "step": 275
    },
    {
      "entropy": 7.910647535324097,
      "epoch": 0.024469107751463777,
      "grad_norm": 1.078125,
      "learning_rate": 0.0001395,
      "loss": 7.524,
      "mean_token_accuracy": 0.08785401061177253,
      "num_tokens": 564587.0,
      "step": 280
    },
    {
      "entropy": 8.014890098571778,
      "epoch": 0.024906056104168486,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.5682,
      "mean_token_accuracy": 0.08573365658521652,
      "num_tokens": 574819.0,
      "step": 285
    },
    {
      "entropy": 7.8606682300567625,
      "epoch": 0.025343004456873198,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0001445,
      "loss": 7.5076,
      "mean_token_accuracy": 0.08299032188951969,
      "num_tokens": 585178.0,
      "step": 290
    },
    {
      "entropy": 7.878801870346069,
      "epoch": 0.025779952809577907,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000147,
      "loss": 7.4975,
      "mean_token_accuracy": 0.08870302848517894,
      "num_tokens": 595086.0,
      "step": 295
    },
    {
      "entropy": 7.8693853378295895,
      "epoch": 0.02621690116228262,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0001495,
      "loss": 7.5201,
      "mean_token_accuracy": 0.08878342881798744,
      "num_tokens": 604942.0,
      "step": 300
    },
    {
      "entropy": 7.927869176864624,
      "epoch": 0.026653849514987327,
      "grad_norm": 1.015625,
      "learning_rate": 0.000152,
      "loss": 7.4803,
      "mean_token_accuracy": 0.0899667464196682,
      "num_tokens": 616082.0,
      "step": 305
    },
    {
      "entropy": 7.873331737518311,
      "epoch": 0.02709079786769204,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.5467,
      "mean_token_accuracy": 0.08932502046227456,
      "num_tokens": 626375.0,
      "step": 310
    },
    {
      "entropy": 7.859624433517456,
      "epoch": 0.027527746220396748,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000157,
      "loss": 7.4942,
      "mean_token_accuracy": 0.08501280173659324,
      "num_tokens": 637563.0,
      "step": 315
    },
    {
      "entropy": 7.986823797225952,
      "epoch": 0.02796469457310146,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0001595,
      "loss": 7.5055,
      "mean_token_accuracy": 0.09258699566125869,
      "num_tokens": 647256.0,
      "step": 320
    },
    {
      "entropy": 7.865791273117066,
      "epoch": 0.02840164292580617,
      "grad_norm": 0.984375,
      "learning_rate": 0.000162,
      "loss": 7.5538,
      "mean_token_accuracy": 0.08355396911501885,
      "num_tokens": 657247.0,
      "step": 325
    },
    {
      "entropy": 7.960019207000732,
      "epoch": 0.02883859127851088,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.4561,
      "mean_token_accuracy": 0.08626900538802147,
      "num_tokens": 667613.0,
      "step": 330
    },
    {
      "entropy": 7.887899160385132,
      "epoch": 0.02927553963121559,
      "grad_norm": 1.09375,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.5092,
      "mean_token_accuracy": 0.08869680538773536,
      "num_tokens": 678048.0,
      "step": 335
    },
    {
      "entropy": 7.926455736160278,
      "epoch": 0.029712487983920302,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.4952,
      "mean_token_accuracy": 0.09030729234218597,
      "num_tokens": 688468.0,
      "step": 340
    },
    {
      "entropy": 7.849487924575806,
      "epoch": 0.03014943633662501,
      "grad_norm": 0.9375,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.4801,
      "mean_token_accuracy": 0.08931280449032783,
      "num_tokens": 699809.0,
      "step": 345
    },
    {
      "entropy": 7.818565845489502,
      "epoch": 0.030586384689329723,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.4455,
      "mean_token_accuracy": 0.09213222786784173,
      "num_tokens": 709227.0,
      "step": 350
    },
    {
      "entropy": 7.93096079826355,
      "epoch": 0.03102333304203443,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000177,
      "loss": 7.5295,
      "mean_token_accuracy": 0.08403203263878822,
      "num_tokens": 720178.0,
      "step": 355
    },
    {
      "entropy": 7.879321956634522,
      "epoch": 0.031460281394739144,
      "grad_norm": 0.9375,
      "learning_rate": 0.0001795,
      "loss": 7.4737,
      "mean_token_accuracy": 0.09429880529642105,
      "num_tokens": 730514.0,
      "step": 360
    },
    {
      "entropy": 7.884739542007447,
      "epoch": 0.03189722974744385,
      "grad_norm": 1.03125,
      "learning_rate": 0.000182,
      "loss": 7.5358,
      "mean_token_accuracy": 0.08618768155574799,
      "num_tokens": 741345.0,
      "step": 365
    },
    {
      "entropy": 7.859089326858521,
      "epoch": 0.03233417810014856,
      "grad_norm": 1.046875,
      "learning_rate": 0.0001845,
      "loss": 7.4394,
      "mean_token_accuracy": 0.0918248526751995,
      "num_tokens": 751163.0,
      "step": 370
    },
    {
      "entropy": 7.865943956375122,
      "epoch": 0.03277112645285327,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000187,
      "loss": 7.4387,
      "mean_token_accuracy": 0.08565517961978912,
      "num_tokens": 762090.0,
      "step": 375
    },
    {
      "entropy": 7.882615852355957,
      "epoch": 0.033208074805557986,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001895,
      "loss": 7.4199,
      "mean_token_accuracy": 0.09661138281226159,
      "num_tokens": 772049.0,
      "step": 380
    },
    {
      "entropy": 7.8357421875,
      "epoch": 0.03364502315826269,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000192,
      "loss": 7.3948,
      "mean_token_accuracy": 0.09582874178886414,
      "num_tokens": 781280.0,
      "step": 385
    },
    {
      "entropy": 7.847851085662842,
      "epoch": 0.0340819715109674,
      "grad_norm": 1.109375,
      "learning_rate": 0.0001945,
      "loss": 7.4399,
      "mean_token_accuracy": 0.10038882195949554,
      "num_tokens": 791642.0,
      "step": 390
    },
    {
      "entropy": 7.808693170547485,
      "epoch": 0.034518919863672115,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.461,
      "mean_token_accuracy": 0.08796226382255554,
      "num_tokens": 801224.0,
      "step": 395
    },
    {
      "entropy": 7.909064292907715,
      "epoch": 0.03495586821637683,
      "grad_norm": 1.109375,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.4035,
      "mean_token_accuracy": 0.09300346970558167,
      "num_tokens": 810727.0,
      "step": 400
    },
    {
      "entropy": 7.90141282081604,
      "epoch": 0.03539281656908153,
      "grad_norm": 1.125,
      "learning_rate": 0.000202,
      "loss": 7.4981,
      "mean_token_accuracy": 0.08850462436676025,
      "num_tokens": 819943.0,
      "step": 405
    },
    {
      "entropy": 7.751878356933593,
      "epoch": 0.035829764921786245,
      "grad_norm": 1.078125,
      "learning_rate": 0.00020449999999999998,
      "loss": 7.4123,
      "mean_token_accuracy": 0.08893894106149673,
      "num_tokens": 829474.0,
      "step": 410
    },
    {
      "entropy": 7.881792497634888,
      "epoch": 0.03626671327449096,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000207,
      "loss": 7.3957,
      "mean_token_accuracy": 0.09066153168678284,
      "num_tokens": 839288.0,
      "step": 415
    },
    {
      "entropy": 7.823987722396851,
      "epoch": 0.03670366162719567,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002095,
      "loss": 7.3985,
      "mean_token_accuracy": 0.09359202459454537,
      "num_tokens": 850064.0,
      "step": 420
    },
    {
      "entropy": 7.700559186935425,
      "epoch": 0.037140609979900374,
      "grad_norm": 1.125,
      "learning_rate": 0.000212,
      "loss": 7.4049,
      "mean_token_accuracy": 0.08930439800024033,
      "num_tokens": 860732.0,
      "step": 425
    },
    {
      "entropy": 7.845630502700805,
      "epoch": 0.037577558332605086,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002145,
      "loss": 7.3843,
      "mean_token_accuracy": 0.09245424419641494,
      "num_tokens": 870076.0,
      "step": 430
    },
    {
      "entropy": 7.678752708435058,
      "epoch": 0.0380145066853098,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00021700000000000002,
      "loss": 7.2926,
      "mean_token_accuracy": 0.09747275933623314,
      "num_tokens": 880201.0,
      "step": 435
    },
    {
      "entropy": 7.746017980575561,
      "epoch": 0.038451455038014504,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002195,
      "loss": 7.365,
      "mean_token_accuracy": 0.09477003365755081,
      "num_tokens": 890344.0,
      "step": 440
    },
    {
      "entropy": 7.766406536102295,
      "epoch": 0.038888403390719216,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000222,
      "loss": 7.3579,
      "mean_token_accuracy": 0.09475301504135132,
      "num_tokens": 900443.0,
      "step": 445
    },
    {
      "entropy": 7.832936811447143,
      "epoch": 0.03932535174342393,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002245,
      "loss": 7.3912,
      "mean_token_accuracy": 0.093039420992136,
      "num_tokens": 910284.0,
      "step": 450
    },
    {
      "entropy": 7.858904647827148,
      "epoch": 0.03976230009612864,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022700000000000002,
      "loss": 7.4642,
      "mean_token_accuracy": 0.09033563137054443,
      "num_tokens": 920986.0,
      "step": 455
    },
    {
      "entropy": 7.757792663574219,
      "epoch": 0.040199248448833345,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00022950000000000002,
      "loss": 7.3627,
      "mean_token_accuracy": 0.09344023391604424,
      "num_tokens": 930481.0,
      "step": 460
    },
    {
      "entropy": 7.816108894348145,
      "epoch": 0.04063619680153806,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00023200000000000003,
      "loss": 7.3938,
      "mean_token_accuracy": 0.09392273649573327,
      "num_tokens": 940554.0,
      "step": 465
    },
    {
      "entropy": 7.7882520198822025,
      "epoch": 0.04107314515424277,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023449999999999998,
      "loss": 7.3473,
      "mean_token_accuracy": 0.09586822837591172,
      "num_tokens": 950005.0,
      "step": 470
    },
    {
      "entropy": 7.681087589263916,
      "epoch": 0.04151009350694748,
      "grad_norm": 1.046875,
      "learning_rate": 0.000237,
      "loss": 7.3382,
      "mean_token_accuracy": 0.09327205643057823,
      "num_tokens": 959414.0,
      "step": 475
    },
    {
      "entropy": 7.797640085220337,
      "epoch": 0.04194704185965219,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002395,
      "loss": 7.3684,
      "mean_token_accuracy": 0.08881603106856346,
      "num_tokens": 969352.0,
      "step": 480
    },
    {
      "entropy": 7.834505796432495,
      "epoch": 0.0423839902123569,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000242,
      "loss": 7.4022,
      "mean_token_accuracy": 0.09166060984134675,
      "num_tokens": 980294.0,
      "step": 485
    },
    {
      "entropy": 7.640561914443969,
      "epoch": 0.04282093856506161,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002445,
      "loss": 7.2872,
      "mean_token_accuracy": 0.09981269985437394,
      "num_tokens": 990453.0,
      "step": 490
    },
    {
      "entropy": 7.763474369049073,
      "epoch": 0.04325788691776632,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000247,
      "loss": 7.2957,
      "mean_token_accuracy": 0.09954148381948472,
      "num_tokens": 1000509.0,
      "step": 495
    },
    {
      "entropy": 7.786115121841431,
      "epoch": 0.04369483527047103,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002495,
      "loss": 7.4391,
      "mean_token_accuracy": 0.08776586875319481,
      "num_tokens": 1011747.0,
      "step": 500
    },
    {
      "entropy": 7.71697096824646,
      "epoch": 0.04413178362317574,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000252,
      "loss": 7.3987,
      "mean_token_accuracy": 0.08907804265618324,
      "num_tokens": 1021511.0,
      "step": 505
    },
    {
      "entropy": 7.781257009506225,
      "epoch": 0.04456873197588045,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002545,
      "loss": 7.3892,
      "mean_token_accuracy": 0.08882547542452812,
      "num_tokens": 1032116.0,
      "step": 510
    },
    {
      "entropy": 7.699050712585449,
      "epoch": 0.04500568032858516,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000257,
      "loss": 7.2936,
      "mean_token_accuracy": 0.0988588310778141,
      "num_tokens": 1041356.0,
      "step": 515
    },
    {
      "entropy": 7.724139022827148,
      "epoch": 0.04544262868128987,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002595,
      "loss": 7.3613,
      "mean_token_accuracy": 0.09164860397577286,
      "num_tokens": 1051314.0,
      "step": 520
    },
    {
      "entropy": 7.725086450576782,
      "epoch": 0.04587957703399458,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000262,
      "loss": 7.3286,
      "mean_token_accuracy": 0.09037298560142518,
      "num_tokens": 1060857.0,
      "step": 525
    },
    {
      "entropy": 7.730894708633423,
      "epoch": 0.046316525386699295,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00026450000000000003,
      "loss": 7.3401,
      "mean_token_accuracy": 0.09307361096143722,
      "num_tokens": 1071744.0,
      "step": 530
    },
    {
      "entropy": 7.750498676300049,
      "epoch": 0.046753473739404,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026700000000000004,
      "loss": 7.3204,
      "mean_token_accuracy": 0.0946407824754715,
      "num_tokens": 1080777.0,
      "step": 535
    },
    {
      "entropy": 7.702250289916992,
      "epoch": 0.04719042209210871,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026950000000000005,
      "loss": 7.3129,
      "mean_token_accuracy": 0.09728860557079315,
      "num_tokens": 1090951.0,
      "step": 540
    },
    {
      "entropy": 7.662070989608765,
      "epoch": 0.047627370444813424,
      "grad_norm": 1.140625,
      "learning_rate": 0.00027200000000000005,
      "loss": 7.2726,
      "mean_token_accuracy": 0.09608129486441612,
      "num_tokens": 1101667.0,
      "step": 545
    },
    {
      "entropy": 7.711210823059082,
      "epoch": 0.04806431879751814,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002745,
      "loss": 7.3992,
      "mean_token_accuracy": 0.08903621137142181,
      "num_tokens": 1112652.0,
      "step": 550
    },
    {
      "entropy": 7.703273153305053,
      "epoch": 0.04850126715022284,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000277,
      "loss": 7.337,
      "mean_token_accuracy": 0.09509981796145439,
      "num_tokens": 1122518.0,
      "step": 555
    },
    {
      "entropy": 7.7245645999908445,
      "epoch": 0.048938215502927554,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002795,
      "loss": 7.345,
      "mean_token_accuracy": 0.09325287640094757,
      "num_tokens": 1132867.0,
      "step": 560
    },
    {
      "entropy": 7.777069234848023,
      "epoch": 0.049375163855632266,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028199999999999997,
      "loss": 7.4122,
      "mean_token_accuracy": 0.08977549225091934,
      "num_tokens": 1142524.0,
      "step": 565
    },
    {
      "entropy": 7.790037679672241,
      "epoch": 0.04981211220833697,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002845,
      "loss": 7.2861,
      "mean_token_accuracy": 0.10011612400412559,
      "num_tokens": 1152983.0,
      "step": 570
    },
    {
      "entropy": 7.716982364654541,
      "epoch": 0.050249060561041684,
      "grad_norm": 1.078125,
      "learning_rate": 0.000287,
      "loss": 7.3616,
      "mean_token_accuracy": 0.09314811900258065,
      "num_tokens": 1164271.0,
      "step": 575
    },
    {
      "entropy": 7.647295999526977,
      "epoch": 0.050686008913746396,
      "grad_norm": 1.125,
      "learning_rate": 0.0002895,
      "loss": 7.2216,
      "mean_token_accuracy": 0.09704305902123452,
      "num_tokens": 1173567.0,
      "step": 580
    },
    {
      "entropy": 7.612508869171142,
      "epoch": 0.05112295726645111,
      "grad_norm": 1.140625,
      "learning_rate": 0.000292,
      "loss": 7.2779,
      "mean_token_accuracy": 0.10037864819169044,
      "num_tokens": 1183619.0,
      "step": 585
    },
    {
      "entropy": 7.655305528640747,
      "epoch": 0.05155990561915581,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002945,
      "loss": 7.233,
      "mean_token_accuracy": 0.09470883682370186,
      "num_tokens": 1193922.0,
      "step": 590
    },
    {
      "entropy": 7.737557649612427,
      "epoch": 0.051996853971860525,
      "grad_norm": 1.15625,
      "learning_rate": 0.000297,
      "loss": 7.3201,
      "mean_token_accuracy": 0.09405622780323028,
      "num_tokens": 1203665.0,
      "step": 595
    },
    {
      "entropy": 7.673120069503784,
      "epoch": 0.05243380232456524,
      "grad_norm": 1.0,
      "learning_rate": 0.0002995,
      "loss": 7.317,
      "mean_token_accuracy": 0.09930261299014091,
      "num_tokens": 1214119.0,
      "step": 600
    },
    {
      "entropy": 7.711197757720948,
      "epoch": 0.05287075067726995,
      "grad_norm": 1.125,
      "learning_rate": 0.000302,
      "loss": 7.3359,
      "mean_token_accuracy": 0.08920879289507866,
      "num_tokens": 1225275.0,
      "step": 605
    },
    {
      "entropy": 7.587856864929199,
      "epoch": 0.053307699029974655,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003045,
      "loss": 7.2069,
      "mean_token_accuracy": 0.09227376878261566,
      "num_tokens": 1235199.0,
      "step": 610
    },
    {
      "entropy": 7.658899211883545,
      "epoch": 0.05374464738267937,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000307,
      "loss": 7.2368,
      "mean_token_accuracy": 0.09821678847074508,
      "num_tokens": 1245648.0,
      "step": 615
    },
    {
      "entropy": 7.546068382263184,
      "epoch": 0.05418159573538408,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003095,
      "loss": 7.2031,
      "mean_token_accuracy": 0.10184830203652381,
      "num_tokens": 1254793.0,
      "step": 620
    },
    {
      "entropy": 7.690952110290527,
      "epoch": 0.05461854408808879,
      "grad_norm": 1.078125,
      "learning_rate": 0.000312,
      "loss": 7.2577,
      "mean_token_accuracy": 0.09762314632534981,
      "num_tokens": 1264429.0,
      "step": 625
    },
    {
      "entropy": 7.53085880279541,
      "epoch": 0.055055492440793496,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003145,
      "loss": 7.2252,
      "mean_token_accuracy": 0.09289011359214783,
      "num_tokens": 1274449.0,
      "step": 630
    },
    {
      "entropy": 7.681129884719849,
      "epoch": 0.05549244079349821,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000317,
      "loss": 7.3249,
      "mean_token_accuracy": 0.09160538911819457,
      "num_tokens": 1284987.0,
      "step": 635
    },
    {
      "entropy": 7.6850756168365475,
      "epoch": 0.05592938914620292,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003195,
      "loss": 7.321,
      "mean_token_accuracy": 0.08939207196235657,
      "num_tokens": 1295984.0,
      "step": 640
    },
    {
      "entropy": 7.543700742721557,
      "epoch": 0.056366337498907626,
      "grad_norm": 1.09375,
      "learning_rate": 0.000322,
      "loss": 7.2222,
      "mean_token_accuracy": 0.09924623519182205,
      "num_tokens": 1305738.0,
      "step": 645
    },
    {
      "entropy": 7.613607883453369,
      "epoch": 0.05680328585161234,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032450000000000003,
      "loss": 7.1722,
      "mean_token_accuracy": 0.09349776208400726,
      "num_tokens": 1315691.0,
      "step": 650
    },
    {
      "entropy": 7.554134130477905,
      "epoch": 0.05724023420431705,
      "grad_norm": 1.0,
      "learning_rate": 0.00032700000000000003,
      "loss": 7.1896,
      "mean_token_accuracy": 0.10020266026258469,
      "num_tokens": 1326195.0,
      "step": 655
    },
    {
      "entropy": 7.56112003326416,
      "epoch": 0.05767718255702176,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032950000000000004,
      "loss": 7.2129,
      "mean_token_accuracy": 0.09725727662444114,
      "num_tokens": 1336916.0,
      "step": 660
    },
    {
      "entropy": 7.603675413131714,
      "epoch": 0.05811413090972647,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033200000000000005,
      "loss": 7.247,
      "mean_token_accuracy": 0.09576455503702164,
      "num_tokens": 1347386.0,
      "step": 665
    },
    {
      "entropy": 7.609761524200439,
      "epoch": 0.05855107926243118,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033450000000000005,
      "loss": 7.196,
      "mean_token_accuracy": 0.09891333803534508,
      "num_tokens": 1357959.0,
      "step": 670
    },
    {
      "entropy": 7.543549108505249,
      "epoch": 0.05898802761513589,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000337,
      "loss": 7.2173,
      "mean_token_accuracy": 0.09141774773597718,
      "num_tokens": 1368135.0,
      "step": 675
    },
    {
      "entropy": 7.643839502334595,
      "epoch": 0.059424975967840604,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003395,
      "loss": 7.2466,
      "mean_token_accuracy": 0.09271192476153374,
      "num_tokens": 1378702.0,
      "step": 680
    },
    {
      "entropy": 7.408550214767456,
      "epoch": 0.05986192432054531,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000342,
      "loss": 7.1268,
      "mean_token_accuracy": 0.1002315916121006,
      "num_tokens": 1387902.0,
      "step": 685
    },
    {
      "entropy": 7.598399209976196,
      "epoch": 0.06029887267325002,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034449999999999997,
      "loss": 7.1424,
      "mean_token_accuracy": 0.10513755530118943,
      "num_tokens": 1397288.0,
      "step": 690
    },
    {
      "entropy": 7.5331981658935545,
      "epoch": 0.060735821025954734,
      "grad_norm": 1.0625,
      "learning_rate": 0.000347,
      "loss": 7.1805,
      "mean_token_accuracy": 0.1005538746714592,
      "num_tokens": 1407416.0,
      "step": 695
    },
    {
      "entropy": 7.626793670654297,
      "epoch": 0.061172769378659446,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003495,
      "loss": 7.2406,
      "mean_token_accuracy": 0.09300423189997672,
      "num_tokens": 1417733.0,
      "step": 700
    },
    {
      "entropy": 7.463755989074707,
      "epoch": 0.06160971773136415,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000352,
      "loss": 7.1325,
      "mean_token_accuracy": 0.09881449341773987,
      "num_tokens": 1426723.0,
      "step": 705
    },
    {
      "entropy": 7.4767341136932375,
      "epoch": 0.06204666608406886,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003545,
      "loss": 7.1811,
      "mean_token_accuracy": 0.09799282476305962,
      "num_tokens": 1437290.0,
      "step": 710
    },
    {
      "entropy": 7.5843759059906,
      "epoch": 0.062483614436773575,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000357,
      "loss": 7.1317,
      "mean_token_accuracy": 0.10022315308451653,
      "num_tokens": 1446756.0,
      "step": 715
    },
    {
      "entropy": 7.491819620132446,
      "epoch": 0.06292056278947829,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003595,
      "loss": 7.1987,
      "mean_token_accuracy": 0.09563518688082695,
      "num_tokens": 1457074.0,
      "step": 720
    },
    {
      "entropy": 7.548601818084717,
      "epoch": 0.06335751114218299,
      "grad_norm": 1.171875,
      "learning_rate": 0.000362,
      "loss": 7.2557,
      "mean_token_accuracy": 0.09054888933897018,
      "num_tokens": 1467785.0,
      "step": 725
    },
    {
      "entropy": 7.486869239807129,
      "epoch": 0.0637944594948877,
      "grad_norm": 1.0,
      "learning_rate": 0.0003645,
      "loss": 7.1532,
      "mean_token_accuracy": 0.09929404258728028,
      "num_tokens": 1477532.0,
      "step": 730
    },
    {
      "entropy": 7.470096015930176,
      "epoch": 0.06423140784759242,
      "grad_norm": 1.03125,
      "learning_rate": 0.000367,
      "loss": 7.0877,
      "mean_token_accuracy": 0.10287476405501365,
      "num_tokens": 1487102.0,
      "step": 735
    },
    {
      "entropy": 7.446895885467529,
      "epoch": 0.06466835620029712,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003695,
      "loss": 7.1666,
      "mean_token_accuracy": 0.09422165304422378,
      "num_tokens": 1497459.0,
      "step": 740
    },
    {
      "entropy": 7.516766309738159,
      "epoch": 0.06510530455300184,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000372,
      "loss": 7.168,
      "mean_token_accuracy": 0.09273407310247421,
      "num_tokens": 1508756.0,
      "step": 745
    },
    {
      "entropy": 7.497219371795654,
      "epoch": 0.06554225290570655,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003745,
      "loss": 7.0858,
      "mean_token_accuracy": 0.10392189174890518,
      "num_tokens": 1518735.0,
      "step": 750
    },
    {
      "entropy": 7.503822278976441,
      "epoch": 0.06597920125841125,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000377,
      "loss": 7.1207,
      "mean_token_accuracy": 0.09711763262748718,
      "num_tokens": 1528632.0,
      "step": 755
    },
    {
      "entropy": 7.355689430236817,
      "epoch": 0.06641614961111597,
      "grad_norm": 1.0,
      "learning_rate": 0.0003795,
      "loss": 7.1008,
      "mean_token_accuracy": 0.09904444217681885,
      "num_tokens": 1538180.0,
      "step": 760
    },
    {
      "entropy": 7.512228107452392,
      "epoch": 0.06685309796382068,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000382,
      "loss": 7.1416,
      "mean_token_accuracy": 0.09793454185128211,
      "num_tokens": 1547788.0,
      "step": 765
    },
    {
      "entropy": 7.426608943939209,
      "epoch": 0.06729004631652538,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003845,
      "loss": 7.0757,
      "mean_token_accuracy": 0.10199111700057983,
      "num_tokens": 1558008.0,
      "step": 770
    },
    {
      "entropy": 7.404742813110351,
      "epoch": 0.0677269946692301,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038700000000000003,
      "loss": 7.1458,
      "mean_token_accuracy": 0.09649223536252975,
      "num_tokens": 1567728.0,
      "step": 775
    },
    {
      "entropy": 7.468697881698608,
      "epoch": 0.0681639430219348,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00038950000000000003,
      "loss": 7.104,
      "mean_token_accuracy": 0.09830691069364547,
      "num_tokens": 1578488.0,
      "step": 780
    },
    {
      "entropy": 7.433673858642578,
      "epoch": 0.06860089137463951,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00039200000000000004,
      "loss": 7.071,
      "mean_token_accuracy": 0.1033280424773693,
      "num_tokens": 1588070.0,
      "step": 785
    },
    {
      "entropy": 7.444263696670532,
      "epoch": 0.06903783972734423,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00039450000000000005,
      "loss": 7.1528,
      "mean_token_accuracy": 0.09422910511493683,
      "num_tokens": 1598572.0,
      "step": 790
    },
    {
      "entropy": 7.519211387634277,
      "epoch": 0.06947478808004894,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039700000000000005,
      "loss": 7.1589,
      "mean_token_accuracy": 0.09713958278298378,
      "num_tokens": 1607982.0,
      "step": 795
    },
    {
      "entropy": 7.391828441619873,
      "epoch": 0.06991173643275365,
      "grad_norm": 1.125,
      "learning_rate": 0.0003995,
      "loss": 7.1658,
      "mean_token_accuracy": 0.0995835930109024,
      "num_tokens": 1618785.0,
      "step": 800
    },
    {
      "entropy": 7.510007715225219,
      "epoch": 0.07034868478545836,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000402,
      "loss": 7.0834,
      "mean_token_accuracy": 0.09890709817409515,
      "num_tokens": 1627988.0,
      "step": 805
    },
    {
      "entropy": 7.354162311553955,
      "epoch": 0.07078563313816306,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004045,
      "loss": 7.1707,
      "mean_token_accuracy": 0.09317449629306793,
      "num_tokens": 1638024.0,
      "step": 810
    },
    {
      "entropy": 7.494234275817871,
      "epoch": 0.07122258149086778,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040699999999999997,
      "loss": 7.1515,
      "mean_token_accuracy": 0.09472875967621804,
      "num_tokens": 1649215.0,
      "step": 815
    },
    {
      "entropy": 7.336769390106201,
      "epoch": 0.07165952984357249,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004095,
      "loss": 7.0398,
      "mean_token_accuracy": 0.10013310313224792,
      "num_tokens": 1659198.0,
      "step": 820
    },
    {
      "entropy": 7.461428451538086,
      "epoch": 0.0720964781962772,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000412,
      "loss": 7.0921,
      "mean_token_accuracy": 0.09403431341052056,
      "num_tokens": 1668972.0,
      "step": 825
    },
    {
      "entropy": 7.293336820602417,
      "epoch": 0.07253342654898191,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004145,
      "loss": 7.0585,
      "mean_token_accuracy": 0.10309704616665841,
      "num_tokens": 1679807.0,
      "step": 830
    },
    {
      "entropy": 7.416385793685913,
      "epoch": 0.07297037490168662,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000417,
      "loss": 7.125,
      "mean_token_accuracy": 0.09676878824830056,
      "num_tokens": 1690567.0,
      "step": 835
    },
    {
      "entropy": 7.312589406967163,
      "epoch": 0.07340732325439134,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004195,
      "loss": 7.0125,
      "mean_token_accuracy": 0.10345098450779915,
      "num_tokens": 1700939.0,
      "step": 840
    },
    {
      "entropy": 7.2877274513244625,
      "epoch": 0.07384427160709604,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000422,
      "loss": 7.0795,
      "mean_token_accuracy": 0.1037543036043644,
      "num_tokens": 1711181.0,
      "step": 845
    },
    {
      "entropy": 7.450462198257446,
      "epoch": 0.07428121995980075,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004245,
      "loss": 7.094,
      "mean_token_accuracy": 0.09376272410154343,
      "num_tokens": 1721392.0,
      "step": 850
    },
    {
      "entropy": 7.326985836029053,
      "epoch": 0.07471816831250547,
      "grad_norm": 1.078125,
      "learning_rate": 0.000427,
      "loss": 7.0237,
      "mean_token_accuracy": 0.10064763948321342,
      "num_tokens": 1731942.0,
      "step": 855
    },
    {
      "entropy": 7.383999919891357,
      "epoch": 0.07515511666521017,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004295,
      "loss": 7.1221,
      "mean_token_accuracy": 0.09718978181481361,
      "num_tokens": 1743409.0,
      "step": 860
    },
    {
      "entropy": 7.321899366378784,
      "epoch": 0.07559206501791488,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000432,
      "loss": 7.0248,
      "mean_token_accuracy": 0.10485066547989845,
      "num_tokens": 1753425.0,
      "step": 865
    },
    {
      "entropy": 7.311501312255859,
      "epoch": 0.0760290133706196,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004345,
      "loss": 7.0481,
      "mean_token_accuracy": 0.09691131561994552,
      "num_tokens": 1762722.0,
      "step": 870
    },
    {
      "entropy": 7.587531995773316,
      "epoch": 0.0764659617233243,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000437,
      "loss": 7.1176,
      "mean_token_accuracy": 0.09000161141157151,
      "num_tokens": 1771895.0,
      "step": 875
    },
    {
      "entropy": 7.24350414276123,
      "epoch": 0.07690291007602901,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004395,
      "loss": 7.0263,
      "mean_token_accuracy": 0.10476705506443977,
      "num_tokens": 1782175.0,
      "step": 880
    },
    {
      "entropy": 7.367937660217285,
      "epoch": 0.07733985842873373,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000442,
      "loss": 7.0853,
      "mean_token_accuracy": 0.09626216888427734,
      "num_tokens": 1792409.0,
      "step": 885
    },
    {
      "entropy": 7.413240146636963,
      "epoch": 0.07777680678143843,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004445,
      "loss": 7.1215,
      "mean_token_accuracy": 0.093594279140234,
      "num_tokens": 1802721.0,
      "step": 890
    },
    {
      "entropy": 7.249949169158936,
      "epoch": 0.07821375513414315,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000447,
      "loss": 6.9593,
      "mean_token_accuracy": 0.0994382567703724,
      "num_tokens": 1812064.0,
      "step": 895
    },
    {
      "entropy": 7.423052549362183,
      "epoch": 0.07865070348684786,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044950000000000003,
      "loss": 7.0682,
      "mean_token_accuracy": 0.09718895554542542,
      "num_tokens": 1822284.0,
      "step": 900
    },
    {
      "entropy": 7.293412828445435,
      "epoch": 0.07908765183955256,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.9757,
      "mean_token_accuracy": 0.10433312505483627,
      "num_tokens": 1832150.0,
      "step": 905
    },
    {
      "entropy": 7.383054161071778,
      "epoch": 0.07952460019225728,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045450000000000004,
      "loss": 7.0895,
      "mean_token_accuracy": 0.09755966365337372,
      "num_tokens": 1842870.0,
      "step": 910
    },
    {
      "entropy": 7.263940238952637,
      "epoch": 0.07996154854496199,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.9138,
      "mean_token_accuracy": 0.10788114890456199,
      "num_tokens": 1852771.0,
      "step": 915
    },
    {
      "entropy": 7.258758020401001,
      "epoch": 0.08039849689766669,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045950000000000006,
      "loss": 7.0202,
      "mean_token_accuracy": 0.10123156458139419,
      "num_tokens": 1863095.0,
      "step": 920
    },
    {
      "entropy": 7.30932240486145,
      "epoch": 0.08083544525037141,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000462,
      "loss": 7.0345,
      "mean_token_accuracy": 0.10289572551846504,
      "num_tokens": 1873060.0,
      "step": 925
    },
    {
      "entropy": 7.286193132400513,
      "epoch": 0.08127239360307612,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004645,
      "loss": 6.9689,
      "mean_token_accuracy": 0.10647596344351769,
      "num_tokens": 1882574.0,
      "step": 930
    },
    {
      "entropy": 7.255521631240844,
      "epoch": 0.08170934195578082,
      "grad_norm": 1.0,
      "learning_rate": 0.000467,
      "loss": 7.0261,
      "mean_token_accuracy": 0.10283842384815216,
      "num_tokens": 1892227.0,
      "step": 935
    },
    {
      "entropy": 7.333770799636841,
      "epoch": 0.08214629030848554,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004695,
      "loss": 7.0544,
      "mean_token_accuracy": 0.09969500303268433,
      "num_tokens": 1902868.0,
      "step": 940
    },
    {
      "entropy": 7.258380460739136,
      "epoch": 0.08258323866119024,
      "grad_norm": 1.046875,
      "learning_rate": 0.000472,
      "loss": 7.0208,
      "mean_token_accuracy": 0.09542301744222641,
      "num_tokens": 1912781.0,
      "step": 945
    },
    {
      "entropy": 7.288229084014892,
      "epoch": 0.08302018701389496,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004745,
      "loss": 7.0122,
      "mean_token_accuracy": 0.10442236512899399,
      "num_tokens": 1923638.0,
      "step": 950
    },
    {
      "entropy": 7.2728832244873045,
      "epoch": 0.08345713536659967,
      "grad_norm": 1.0625,
      "learning_rate": 0.000477,
      "loss": 6.9727,
      "mean_token_accuracy": 0.1044746495783329,
      "num_tokens": 1933498.0,
      "step": 955
    },
    {
      "entropy": 7.165169334411621,
      "epoch": 0.08389408371930437,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004795,
      "loss": 6.9808,
      "mean_token_accuracy": 0.10142417028546333,
      "num_tokens": 1943798.0,
      "step": 960
    },
    {
      "entropy": 7.275189161300659,
      "epoch": 0.0843310320720091,
      "grad_norm": 1.109375,
      "learning_rate": 0.000482,
      "loss": 7.0178,
      "mean_token_accuracy": 0.1035747729241848,
      "num_tokens": 1954235.0,
      "step": 965
    },
    {
      "entropy": 7.269236516952515,
      "epoch": 0.0847679804247138,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004845,
      "loss": 7.0173,
      "mean_token_accuracy": 0.09617507830262184,
      "num_tokens": 1964421.0,
      "step": 970
    },
    {
      "entropy": 7.1920867443084715,
      "epoch": 0.0852049287774185,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000487,
      "loss": 6.9847,
      "mean_token_accuracy": 0.105922382324934,
      "num_tokens": 1975862.0,
      "step": 975
    },
    {
      "entropy": 7.282237482070923,
      "epoch": 0.08564187713012322,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004895,
      "loss": 6.9777,
      "mean_token_accuracy": 0.10412709340453148,
      "num_tokens": 1985758.0,
      "step": 980
    },
    {
      "entropy": 7.226387071609497,
      "epoch": 0.08607882548282793,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000492,
      "loss": 6.9644,
      "mean_token_accuracy": 0.1005631759762764,
      "num_tokens": 1996904.0,
      "step": 985
    },
    {
      "entropy": 7.325346755981445,
      "epoch": 0.08651577383553263,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004945,
      "loss": 7.0478,
      "mean_token_accuracy": 0.1003109060227871,
      "num_tokens": 2006114.0,
      "step": 990
    },
    {
      "entropy": 7.19905686378479,
      "epoch": 0.08695272218823735,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000497,
      "loss": 6.9881,
      "mean_token_accuracy": 0.10641124993562698,
      "num_tokens": 2016349.0,
      "step": 995
    },
    {
      "entropy": 7.211451530456543,
      "epoch": 0.08738967054094206,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004995,
      "loss": 6.9353,
      "mean_token_accuracy": 0.10676476284861565,
      "num_tokens": 2026701.0,
      "step": 1000
    },
    {
      "entropy": 7.1555695056915285,
      "epoch": 0.08782661889364678,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999999986190023,
      "loss": 6.9266,
      "mean_token_accuracy": 0.10275837779045105,
      "num_tokens": 2036950.0,
      "step": 1005
    },
    {
      "entropy": 7.262877178192139,
      "epoch": 0.08826356724635148,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999999930086991,
      "loss": 6.9397,
      "mean_token_accuracy": 0.10614437088370324,
      "num_tokens": 2047655.0,
      "step": 1010
    },
    {
      "entropy": 7.181167459487915,
      "epoch": 0.08870051559905619,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499999983082778,
      "loss": 6.9041,
      "mean_token_accuracy": 0.10694134309887886,
      "num_tokens": 2056800.0,
      "step": 1015
    },
    {
      "entropy": 7.162737894058227,
      "epoch": 0.0891374639517609,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999999688412394,
      "loss": 6.9414,
      "mean_token_accuracy": 0.10635368078947068,
      "num_tokens": 2066862.0,
      "step": 1020
    },
    {
      "entropy": 7.205184698104858,
      "epoch": 0.08957441230446561,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999999502840836,
      "loss": 6.9621,
      "mean_token_accuracy": 0.10661176219582558,
      "num_tokens": 2076585.0,
      "step": 1025
    },
    {
      "entropy": 7.230246829986572,
      "epoch": 0.09001136065717032,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999999274113107,
      "loss": 7.0242,
      "mean_token_accuracy": 0.0967386618256569,
      "num_tokens": 2087768.0,
      "step": 1030
    },
    {
      "entropy": 7.244874048233032,
      "epoch": 0.09044830900987504,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999999002229214,
      "loss": 6.9319,
      "mean_token_accuracy": 0.10978882536292076,
      "num_tokens": 2097027.0,
      "step": 1035
    },
    {
      "entropy": 7.1607935428619385,
      "epoch": 0.09088525736257974,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999998687189161,
      "loss": 6.8632,
      "mean_token_accuracy": 0.1147397093474865,
      "num_tokens": 2107055.0,
      "step": 1040
    },
    {
      "entropy": 7.114990091323852,
      "epoch": 0.09132220571528446,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999998328992954,
      "loss": 6.8043,
      "mean_token_accuracy": 0.10847252830863,
      "num_tokens": 2116771.0,
      "step": 1045
    },
    {
      "entropy": 7.116586971282959,
      "epoch": 0.09175915406798917,
      "grad_norm": 1.109375,
      "learning_rate": 0.00049999979276406,
      "loss": 6.9405,
      "mean_token_accuracy": 0.10683408454060554,
      "num_tokens": 2126881.0,
      "step": 1050
    },
    {
      "entropy": 7.139071369171143,
      "epoch": 0.09219610242069387,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999997483132108,
      "loss": 6.9056,
      "mean_token_accuracy": 0.10297400355339051,
      "num_tokens": 2136964.0,
      "step": 1055
    },
    {
      "entropy": 7.256573295593261,
      "epoch": 0.09263305077339859,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999996995467483,
      "loss": 6.9573,
      "mean_token_accuracy": 0.10562608242034913,
      "num_tokens": 2147263.0,
      "step": 1060
    },
    {
      "entropy": 7.122767400741577,
      "epoch": 0.0930699991261033,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999996464646738,
      "loss": 6.9787,
      "mean_token_accuracy": 0.10480033457279206,
      "num_tokens": 2157474.0,
      "step": 1065
    },
    {
      "entropy": 7.128157997131348,
      "epoch": 0.093506947478808,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999995890669882,
      "loss": 6.9069,
      "mean_token_accuracy": 0.10734480768442153,
      "num_tokens": 2167701.0,
      "step": 1070
    },
    {
      "entropy": 7.108348178863525,
      "epoch": 0.09394389583151272,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999995273536925,
      "loss": 6.8695,
      "mean_token_accuracy": 0.10862331837415695,
      "num_tokens": 2178581.0,
      "step": 1075
    },
    {
      "entropy": 7.111027956008911,
      "epoch": 0.09438084418421742,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499999461324788,
      "loss": 6.8725,
      "mean_token_accuracy": 0.10809326171875,
      "num_tokens": 2189529.0,
      "step": 1080
    },
    {
      "entropy": 7.156719923019409,
      "epoch": 0.09481779253692213,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999993909802759,
      "loss": 6.9364,
      "mean_token_accuracy": 0.10107551366090775,
      "num_tokens": 2199805.0,
      "step": 1085
    },
    {
      "entropy": 7.0858289241790775,
      "epoch": 0.09525474088962685,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999993163201577,
      "loss": 6.8985,
      "mean_token_accuracy": 0.10579909533262252,
      "num_tokens": 2210108.0,
      "step": 1090
    },
    {
      "entropy": 7.26385760307312,
      "epoch": 0.09569168924233155,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999992373444347,
      "loss": 6.8966,
      "mean_token_accuracy": 0.10724986046552658,
      "num_tokens": 2220325.0,
      "step": 1095
    },
    {
      "entropy": 7.044133472442627,
      "epoch": 0.09612863759503627,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999991540531084,
      "loss": 6.8976,
      "mean_token_accuracy": 0.1008307620882988,
      "num_tokens": 2230521.0,
      "step": 1100
    },
    {
      "entropy": 7.1802715301513675,
      "epoch": 0.09656558594774098,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004999990664461804,
      "loss": 6.933,
      "mean_token_accuracy": 0.10014933347702026,
      "num_tokens": 2240372.0,
      "step": 1105
    },
    {
      "entropy": 7.06400728225708,
      "epoch": 0.09700253430044568,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999989745236524,
      "loss": 6.8761,
      "mean_token_accuracy": 0.11035806387662887,
      "num_tokens": 2250522.0,
      "step": 1110
    },
    {
      "entropy": 7.090740442276001,
      "epoch": 0.0974394826531504,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999988782855261,
      "loss": 6.8382,
      "mean_token_accuracy": 0.10595676749944687,
      "num_tokens": 2260142.0,
      "step": 1115
    },
    {
      "entropy": 7.09924898147583,
      "epoch": 0.09787643100585511,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999987777318035,
      "loss": 6.852,
      "mean_token_accuracy": 0.10673597902059555,
      "num_tokens": 2270372.0,
      "step": 1120
    },
    {
      "entropy": 6.989193630218506,
      "epoch": 0.09831337935855981,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999986728624865,
      "loss": 6.7996,
      "mean_token_accuracy": 0.11145754531025887,
      "num_tokens": 2280517.0,
      "step": 1125
    },
    {
      "entropy": 7.1385228633880615,
      "epoch": 0.09875032771126453,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499998563677577,
      "loss": 6.8951,
      "mean_token_accuracy": 0.10448078438639641,
      "num_tokens": 2290778.0,
      "step": 1130
    },
    {
      "entropy": 7.123292922973633,
      "epoch": 0.09918727606396924,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999984501770771,
      "loss": 6.9369,
      "mean_token_accuracy": 0.10169804915785789,
      "num_tokens": 2301692.0,
      "step": 1135
    },
    {
      "entropy": 7.095804166793823,
      "epoch": 0.09962422441667394,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004999983323609891,
      "loss": 6.8209,
      "mean_token_accuracy": 0.10772261098027229,
      "num_tokens": 2312759.0,
      "step": 1140
    },
    {
      "entropy": 7.069378757476807,
      "epoch": 0.10006117276937866,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999982102293153,
      "loss": 6.8021,
      "mean_token_accuracy": 0.10774984061717988,
      "num_tokens": 2322016.0,
      "step": 1145
    },
    {
      "entropy": 7.055843925476074,
      "epoch": 0.10049812112208337,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999980837820578,
      "loss": 6.8062,
      "mean_token_accuracy": 0.11227297484874725,
      "num_tokens": 2332476.0,
      "step": 1150
    },
    {
      "entropy": 7.105614280700683,
      "epoch": 0.10093506947478809,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999979530192192,
      "loss": 6.8314,
      "mean_token_accuracy": 0.10820274353027344,
      "num_tokens": 2342914.0,
      "step": 1155
    },
    {
      "entropy": 7.032117033004761,
      "epoch": 0.10137201782749279,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999978179408019,
      "loss": 6.8902,
      "mean_token_accuracy": 0.10032764598727226,
      "num_tokens": 2353381.0,
      "step": 1160
    },
    {
      "entropy": 7.120272445678711,
      "epoch": 0.1018089661801975,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999976785468085,
      "loss": 6.8674,
      "mean_token_accuracy": 0.10994301959872246,
      "num_tokens": 2364207.0,
      "step": 1165
    },
    {
      "entropy": 6.9931823253631595,
      "epoch": 0.10224591453290222,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999975348372418,
      "loss": 6.7968,
      "mean_token_accuracy": 0.11197755932807922,
      "num_tokens": 2374985.0,
      "step": 1170
    },
    {
      "entropy": 7.06548662185669,
      "epoch": 0.10268286288560692,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999973868121044,
      "loss": 6.8752,
      "mean_token_accuracy": 0.10842390954494477,
      "num_tokens": 2385684.0,
      "step": 1175
    },
    {
      "entropy": 7.193864440917968,
      "epoch": 0.10311981123831163,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999972344713994,
      "loss": 6.7953,
      "mean_token_accuracy": 0.10968969836831093,
      "num_tokens": 2395177.0,
      "step": 1180
    },
    {
      "entropy": 6.892039108276367,
      "epoch": 0.10355675959101635,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999970778151295,
      "loss": 6.8292,
      "mean_token_accuracy": 0.1131418652832508,
      "num_tokens": 2405154.0,
      "step": 1185
    },
    {
      "entropy": 7.109063673019409,
      "epoch": 0.10399370794372105,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999969168432976,
      "loss": 6.8128,
      "mean_token_accuracy": 0.10942742675542831,
      "num_tokens": 2415818.0,
      "step": 1190
    },
    {
      "entropy": 6.981149959564209,
      "epoch": 0.10443065629642576,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999967515559071,
      "loss": 6.7937,
      "mean_token_accuracy": 0.10865649655461311,
      "num_tokens": 2425072.0,
      "step": 1195
    },
    {
      "entropy": 7.053072881698609,
      "epoch": 0.10486760464913047,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499996581952961,
      "loss": 6.8427,
      "mean_token_accuracy": 0.11198535412549973,
      "num_tokens": 2435055.0,
      "step": 1200
    },
    {
      "entropy": 7.097857189178467,
      "epoch": 0.10530455300183518,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999964080344625,
      "loss": 6.7474,
      "mean_token_accuracy": 0.11387341618537902,
      "num_tokens": 2444434.0,
      "step": 1205
    },
    {
      "entropy": 6.945683431625366,
      "epoch": 0.1057415013545399,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999962298004151,
      "loss": 6.7673,
      "mean_token_accuracy": 0.10980340838432312,
      "num_tokens": 2453860.0,
      "step": 1210
    },
    {
      "entropy": 7.091061687469482,
      "epoch": 0.1061784497072446,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999960472508219,
      "loss": 6.8713,
      "mean_token_accuracy": 0.10719245076179504,
      "num_tokens": 2464235.0,
      "step": 1215
    },
    {
      "entropy": 7.024345874786377,
      "epoch": 0.10661539805994931,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999958603856869,
      "loss": 6.8067,
      "mean_token_accuracy": 0.1112932026386261,
      "num_tokens": 2474561.0,
      "step": 1220
    },
    {
      "entropy": 6.958724117279052,
      "epoch": 0.10705234641265403,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999956692050132,
      "loss": 6.7833,
      "mean_token_accuracy": 0.11405704021453858,
      "num_tokens": 2484574.0,
      "step": 1225
    },
    {
      "entropy": 7.0886390209198,
      "epoch": 0.10748929476535873,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999954737088048,
      "loss": 6.7501,
      "mean_token_accuracy": 0.11529114991426467,
      "num_tokens": 2494146.0,
      "step": 1230
    },
    {
      "entropy": 6.948289823532105,
      "epoch": 0.10792624311806344,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999952738970652,
      "loss": 6.8264,
      "mean_token_accuracy": 0.11386599391698837,
      "num_tokens": 2504157.0,
      "step": 1235
    },
    {
      "entropy": 7.057093334197998,
      "epoch": 0.10836319147076816,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999950697697983,
      "loss": 6.7766,
      "mean_token_accuracy": 0.1121026560664177,
      "num_tokens": 2512979.0,
      "step": 1240
    },
    {
      "entropy": 6.975002336502075,
      "epoch": 0.10880013982347286,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999948613270081,
      "loss": 6.8346,
      "mean_token_accuracy": 0.10475755855441093,
      "num_tokens": 2523202.0,
      "step": 1245
    },
    {
      "entropy": 7.0157585620880125,
      "epoch": 0.10923708817617758,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999946485686987,
      "loss": 6.7808,
      "mean_token_accuracy": 0.11152637600898743,
      "num_tokens": 2532790.0,
      "step": 1250
    },
    {
      "entropy": 6.999054861068726,
      "epoch": 0.10967403652888229,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499994431494874,
      "loss": 6.7737,
      "mean_token_accuracy": 0.1105453684926033,
      "num_tokens": 2542337.0,
      "step": 1255
    },
    {
      "entropy": 6.91253628730774,
      "epoch": 0.11011098488158699,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999942101055381,
      "loss": 6.7273,
      "mean_token_accuracy": 0.11472218930721283,
      "num_tokens": 2552762.0,
      "step": 1260
    },
    {
      "entropy": 6.998731660842895,
      "epoch": 0.11054793323429171,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999939844006955,
      "loss": 6.7637,
      "mean_token_accuracy": 0.11208643317222595,
      "num_tokens": 2563261.0,
      "step": 1265
    },
    {
      "entropy": 7.025073385238647,
      "epoch": 0.11098488158699642,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999937543803503,
      "loss": 6.7246,
      "mean_token_accuracy": 0.11664522662758828,
      "num_tokens": 2573089.0,
      "step": 1270
    },
    {
      "entropy": 6.890957021713257,
      "epoch": 0.11142182993970112,
      "grad_norm": 1.0,
      "learning_rate": 0.000499993520044507,
      "loss": 6.8066,
      "mean_token_accuracy": 0.10803473964333535,
      "num_tokens": 2584358.0,
      "step": 1275
    },
    {
      "entropy": 7.083640241622925,
      "epoch": 0.11185877829240584,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999932813931702,
      "loss": 6.7817,
      "mean_token_accuracy": 0.11540437862277031,
      "num_tokens": 2594981.0,
      "step": 1280
    },
    {
      "entropy": 6.970252609252929,
      "epoch": 0.11229572664511055,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999930384263441,
      "loss": 6.7769,
      "mean_token_accuracy": 0.1091542400419712,
      "num_tokens": 2603508.0,
      "step": 1285
    },
    {
      "entropy": 7.003110408782959,
      "epoch": 0.11273267499781525,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499992791144034,
      "loss": 6.747,
      "mean_token_accuracy": 0.11488912031054496,
      "num_tokens": 2613666.0,
      "step": 1290
    },
    {
      "entropy": 7.111478185653686,
      "epoch": 0.11316962335051997,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999925395462439,
      "loss": 6.8063,
      "mean_token_accuracy": 0.10856303349137306,
      "num_tokens": 2624441.0,
      "step": 1295
    },
    {
      "entropy": 6.943297529220581,
      "epoch": 0.11360657170322468,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999922836329793,
      "loss": 6.7665,
      "mean_token_accuracy": 0.10978462025523186,
      "num_tokens": 2635261.0,
      "step": 1300
    },
    {
      "entropy": 6.897740364074707,
      "epoch": 0.1140435200559294,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999920234042446,
      "loss": 6.7374,
      "mean_token_accuracy": 0.11765253841876984,
      "num_tokens": 2645493.0,
      "step": 1305
    },
    {
      "entropy": 6.958786058425903,
      "epoch": 0.1144804684086341,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999917588600451,
      "loss": 6.7047,
      "mean_token_accuracy": 0.11853755936026573,
      "num_tokens": 2655430.0,
      "step": 1310
    },
    {
      "entropy": 6.939574718475342,
      "epoch": 0.1149174167613388,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999914900003857,
      "loss": 6.7597,
      "mean_token_accuracy": 0.11579839065670967,
      "num_tokens": 2665675.0,
      "step": 1315
    },
    {
      "entropy": 6.983940458297729,
      "epoch": 0.11535436511404352,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999912168252717,
      "loss": 6.7285,
      "mean_token_accuracy": 0.11347154676914215,
      "num_tokens": 2675962.0,
      "step": 1320
    },
    {
      "entropy": 6.963703870773315,
      "epoch": 0.11579131346674823,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999909393347083,
      "loss": 6.766,
      "mean_token_accuracy": 0.11305825337767601,
      "num_tokens": 2687015.0,
      "step": 1325
    },
    {
      "entropy": 6.978190517425537,
      "epoch": 0.11622826181945294,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999906575287006,
      "loss": 6.7953,
      "mean_token_accuracy": 0.10686544924974442,
      "num_tokens": 2698065.0,
      "step": 1330
    },
    {
      "entropy": 6.9220874309539795,
      "epoch": 0.11666521017215765,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999903714072544,
      "loss": 6.6976,
      "mean_token_accuracy": 0.1245916247367859,
      "num_tokens": 2709046.0,
      "step": 1335
    },
    {
      "entropy": 7.0129210472106935,
      "epoch": 0.11710215852486236,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999900809703749,
      "loss": 6.7733,
      "mean_token_accuracy": 0.11275611147284507,
      "num_tokens": 2719705.0,
      "step": 1340
    },
    {
      "entropy": 6.906162214279175,
      "epoch": 0.11753910687756706,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004999897862180678,
      "loss": 6.636,
      "mean_token_accuracy": 0.1237392172217369,
      "num_tokens": 2729556.0,
      "step": 1345
    },
    {
      "entropy": 6.873116540908813,
      "epoch": 0.11797605523027178,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999894871503386,
      "loss": 6.7608,
      "mean_token_accuracy": 0.1130751185119152,
      "num_tokens": 2741144.0,
      "step": 1350
    },
    {
      "entropy": 7.004008674621582,
      "epoch": 0.11841300358297649,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999891837671932,
      "loss": 6.6835,
      "mean_token_accuracy": 0.11774843856692314,
      "num_tokens": 2751384.0,
      "step": 1355
    },
    {
      "entropy": 6.905232238769531,
      "epoch": 0.11884995193568121,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999888760686375,
      "loss": 6.6982,
      "mean_token_accuracy": 0.10932691991329194,
      "num_tokens": 2761044.0,
      "step": 1360
    },
    {
      "entropy": 6.811592769622803,
      "epoch": 0.11928690028838591,
      "grad_norm": 0.953125,
      "learning_rate": 0.000499988564054677,
      "loss": 6.665,
      "mean_token_accuracy": 0.1154528170824051,
      "num_tokens": 2770946.0,
      "step": 1365
    },
    {
      "entropy": 7.016284084320068,
      "epoch": 0.11972384864109062,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999882477253182,
      "loss": 6.7196,
      "mean_token_accuracy": 0.109063258767128,
      "num_tokens": 2781247.0,
      "step": 1370
    },
    {
      "entropy": 6.83001856803894,
      "epoch": 0.12016079699379534,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999879270805667,
      "loss": 6.5671,
      "mean_token_accuracy": 0.1252991333603859,
      "num_tokens": 2792078.0,
      "step": 1375
    },
    {
      "entropy": 6.864751482009888,
      "epoch": 0.12059774534650004,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999876021204291,
      "loss": 6.6585,
      "mean_token_accuracy": 0.11526466608047485,
      "num_tokens": 2802458.0,
      "step": 1380
    },
    {
      "entropy": 6.941031503677368,
      "epoch": 0.12103469369920475,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999872728449112,
      "loss": 6.8076,
      "mean_token_accuracy": 0.11191245838999749,
      "num_tokens": 2813352.0,
      "step": 1385
    },
    {
      "entropy": 6.937870597839355,
      "epoch": 0.12147164205190947,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999869392540195,
      "loss": 6.6947,
      "mean_token_accuracy": 0.11170903369784355,
      "num_tokens": 2822954.0,
      "step": 1390
    },
    {
      "entropy": 6.951033449172973,
      "epoch": 0.12190859040461417,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999866013477605,
      "loss": 6.727,
      "mean_token_accuracy": 0.11035996600985527,
      "num_tokens": 2833375.0,
      "step": 1395
    },
    {
      "entropy": 6.83228759765625,
      "epoch": 0.12234553875731889,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999862591261407,
      "loss": 6.6502,
      "mean_token_accuracy": 0.11303703859448433,
      "num_tokens": 2843103.0,
      "step": 1400
    },
    {
      "entropy": 6.9420592308044435,
      "epoch": 0.1227824871100236,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999859125891664,
      "loss": 6.732,
      "mean_token_accuracy": 0.11628735661506653,
      "num_tokens": 2853801.0,
      "step": 1405
    },
    {
      "entropy": 6.89582667350769,
      "epoch": 0.1232194354627283,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999855617368445,
      "loss": 6.7202,
      "mean_token_accuracy": 0.10793796628713608,
      "num_tokens": 2863851.0,
      "step": 1410
    },
    {
      "entropy": 6.8485987186431885,
      "epoch": 0.12365638381543302,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999852065691816,
      "loss": 6.656,
      "mean_token_accuracy": 0.11803558021783829,
      "num_tokens": 2875073.0,
      "step": 1415
    },
    {
      "entropy": 6.966313171386719,
      "epoch": 0.12409333216813773,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999848470861846,
      "loss": 6.6892,
      "mean_token_accuracy": 0.11483604982495307,
      "num_tokens": 2885676.0,
      "step": 1420
    },
    {
      "entropy": 6.886367988586426,
      "epoch": 0.12453028052084243,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999844832878602,
      "loss": 6.6692,
      "mean_token_accuracy": 0.12404601722955703,
      "num_tokens": 2896663.0,
      "step": 1425
    },
    {
      "entropy": 6.865903234481811,
      "epoch": 0.12496722887354715,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999841151742156,
      "loss": 6.6836,
      "mean_token_accuracy": 0.11751426309347153,
      "num_tokens": 2907555.0,
      "step": 1430
    },
    {
      "entropy": 6.834966945648193,
      "epoch": 0.12540417722625186,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999837427452578,
      "loss": 6.6711,
      "mean_token_accuracy": 0.1152021735906601,
      "num_tokens": 2917173.0,
      "step": 1435
    },
    {
      "entropy": 6.887197113037109,
      "epoch": 0.12584112557895658,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499983366000994,
      "loss": 6.6828,
      "mean_token_accuracy": 0.11685618609189988,
      "num_tokens": 2928268.0,
      "step": 1440
    },
    {
      "entropy": 6.919593954086304,
      "epoch": 0.12627807393166127,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999829849414312,
      "loss": 6.6339,
      "mean_token_accuracy": 0.12221134454011917,
      "num_tokens": 2938203.0,
      "step": 1445
    },
    {
      "entropy": 6.8582720279693605,
      "epoch": 0.12671502228436599,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499982599566577,
      "loss": 6.6622,
      "mean_token_accuracy": 0.12187671884894372,
      "num_tokens": 2948094.0,
      "step": 1450
    },
    {
      "entropy": 6.868987035751343,
      "epoch": 0.1271519706370707,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999822098764385,
      "loss": 6.6657,
      "mean_token_accuracy": 0.11611272245645524,
      "num_tokens": 2957865.0,
      "step": 1455
    },
    {
      "entropy": 6.846604824066162,
      "epoch": 0.1275889189897754,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999818158710234,
      "loss": 6.5737,
      "mean_token_accuracy": 0.12412638366222381,
      "num_tokens": 2967880.0,
      "step": 1460
    },
    {
      "entropy": 6.874258327484131,
      "epoch": 0.12802586734248012,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999814175503392,
      "loss": 6.6387,
      "mean_token_accuracy": 0.11633199378848076,
      "num_tokens": 2979537.0,
      "step": 1465
    },
    {
      "entropy": 6.812403917312622,
      "epoch": 0.12846281569518483,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999810149143935,
      "loss": 6.6275,
      "mean_token_accuracy": 0.11614747941493989,
      "num_tokens": 2989700.0,
      "step": 1470
    },
    {
      "entropy": 6.771956157684326,
      "epoch": 0.12889976404788953,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499980607963194,
      "loss": 6.639,
      "mean_token_accuracy": 0.11743173152208328,
      "num_tokens": 2999141.0,
      "step": 1475
    },
    {
      "entropy": 6.867529582977295,
      "epoch": 0.12933671240059424,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999801966967486,
      "loss": 6.593,
      "mean_token_accuracy": 0.12583683878183366,
      "num_tokens": 3008259.0,
      "step": 1480
    },
    {
      "entropy": 6.905298471450806,
      "epoch": 0.12977366075329896,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999797811150652,
      "loss": 6.634,
      "mean_token_accuracy": 0.11574389562010765,
      "num_tokens": 3018281.0,
      "step": 1485
    },
    {
      "entropy": 6.850580358505249,
      "epoch": 0.13021060910600368,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999793612181516,
      "loss": 6.6416,
      "mean_token_accuracy": 0.12190348654985428,
      "num_tokens": 3029062.0,
      "step": 1490
    },
    {
      "entropy": 6.823162078857422,
      "epoch": 0.13064755745870837,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999789370060161,
      "loss": 6.6053,
      "mean_token_accuracy": 0.12013282999396324,
      "num_tokens": 3038142.0,
      "step": 1495
    },
    {
      "entropy": 6.878135681152344,
      "epoch": 0.1310845058114131,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999785084786667,
      "loss": 6.7104,
      "mean_token_accuracy": 0.10974604934453965,
      "num_tokens": 3049123.0,
      "step": 1500
    },
    {
      "entropy": 6.827748155593872,
      "epoch": 0.1315214541641178,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999780756361114,
      "loss": 6.6109,
      "mean_token_accuracy": 0.11498570144176483,
      "num_tokens": 3059031.0,
      "step": 1505
    },
    {
      "entropy": 6.89024248123169,
      "epoch": 0.1319584025168225,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999776384783589,
      "loss": 6.6644,
      "mean_token_accuracy": 0.11741800382733344,
      "num_tokens": 3068830.0,
      "step": 1510
    },
    {
      "entropy": 6.721803903579712,
      "epoch": 0.13239535086952722,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004999771970054174,
      "loss": 6.6362,
      "mean_token_accuracy": 0.11696416139602661,
      "num_tokens": 3080124.0,
      "step": 1515
    },
    {
      "entropy": 6.899826955795288,
      "epoch": 0.13283229922223194,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999767512172953,
      "loss": 6.638,
      "mean_token_accuracy": 0.11275093853473664,
      "num_tokens": 3089607.0,
      "step": 1520
    },
    {
      "entropy": 6.817293930053711,
      "epoch": 0.13326924757493663,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999763011140013,
      "loss": 6.7252,
      "mean_token_accuracy": 0.10601238310337066,
      "num_tokens": 3099363.0,
      "step": 1525
    },
    {
      "entropy": 6.935261344909668,
      "epoch": 0.13370619592764135,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999758466955439,
      "loss": 6.6219,
      "mean_token_accuracy": 0.1220114767551422,
      "num_tokens": 3109552.0,
      "step": 1530
    },
    {
      "entropy": 6.773904275894165,
      "epoch": 0.13414314428034607,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999753879619319,
      "loss": 6.6,
      "mean_token_accuracy": 0.12055370435118676,
      "num_tokens": 3119536.0,
      "step": 1535
    },
    {
      "entropy": 6.843937015533447,
      "epoch": 0.13458009263305076,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999749249131741,
      "loss": 6.6517,
      "mean_token_accuracy": 0.1186549685895443,
      "num_tokens": 3130552.0,
      "step": 1540
    },
    {
      "entropy": 6.893322515487671,
      "epoch": 0.13501704098575548,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999744575492793,
      "loss": 6.5489,
      "mean_token_accuracy": 0.11952240914106368,
      "num_tokens": 3140723.0,
      "step": 1545
    },
    {
      "entropy": 6.793356323242188,
      "epoch": 0.1354539893384602,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999739858702567,
      "loss": 6.5877,
      "mean_token_accuracy": 0.12382742390036583,
      "num_tokens": 3151031.0,
      "step": 1550
    },
    {
      "entropy": 6.760280799865723,
      "epoch": 0.1358909376911649,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499973509876115,
      "loss": 6.5922,
      "mean_token_accuracy": 0.12205319106578827,
      "num_tokens": 3161373.0,
      "step": 1555
    },
    {
      "entropy": 6.842703056335449,
      "epoch": 0.1363278860438696,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999730295668634,
      "loss": 6.6059,
      "mean_token_accuracy": 0.11810858622193336,
      "num_tokens": 3171094.0,
      "step": 1560
    },
    {
      "entropy": 6.792533588409424,
      "epoch": 0.13676483439657433,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999725449425114,
      "loss": 6.5583,
      "mean_token_accuracy": 0.1213751420378685,
      "num_tokens": 3181646.0,
      "step": 1565
    },
    {
      "entropy": 6.758139514923096,
      "epoch": 0.13720178274927902,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499972056003068,
      "loss": 6.5366,
      "mean_token_accuracy": 0.12806674242019653,
      "num_tokens": 3190727.0,
      "step": 1570
    },
    {
      "entropy": 6.871710777282715,
      "epoch": 0.13763873110198374,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999715627485426,
      "loss": 6.6295,
      "mean_token_accuracy": 0.11181031242012977,
      "num_tokens": 3200316.0,
      "step": 1575
    },
    {
      "entropy": 6.744039487838745,
      "epoch": 0.13807567945468846,
      "grad_norm": 0.9375,
      "learning_rate": 0.000499971065178945,
      "loss": 6.5833,
      "mean_token_accuracy": 0.11728300899267197,
      "num_tokens": 3211830.0,
      "step": 1580
    },
    {
      "entropy": 6.80518102645874,
      "epoch": 0.13851262780739318,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999705632942843,
      "loss": 6.6093,
      "mean_token_accuracy": 0.11741095557808875,
      "num_tokens": 3222711.0,
      "step": 1585
    },
    {
      "entropy": 6.751020383834839,
      "epoch": 0.13894957616009787,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999700570945704,
      "loss": 6.5137,
      "mean_token_accuracy": 0.13087992370128632,
      "num_tokens": 3232810.0,
      "step": 1590
    },
    {
      "entropy": 6.756484413146973,
      "epoch": 0.1393865245128026,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999695465798128,
      "loss": 6.593,
      "mean_token_accuracy": 0.12206040173768998,
      "num_tokens": 3243366.0,
      "step": 1595
    },
    {
      "entropy": 6.843332004547119,
      "epoch": 0.1398234728655073,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999690317500215,
      "loss": 6.5984,
      "mean_token_accuracy": 0.12029016464948654,
      "num_tokens": 3254000.0,
      "step": 1600
    },
    {
      "entropy": 6.8639579772949215,
      "epoch": 0.140260421218212,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999685126052063,
      "loss": 6.6187,
      "mean_token_accuracy": 0.11791635006666183,
      "num_tokens": 3263601.0,
      "step": 1605
    },
    {
      "entropy": 6.737782335281372,
      "epoch": 0.14069736957091672,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999679891453772,
      "loss": 6.6507,
      "mean_token_accuracy": 0.11550768241286277,
      "num_tokens": 3273243.0,
      "step": 1610
    },
    {
      "entropy": 6.825797033309937,
      "epoch": 0.14113431792362144,
      "grad_norm": 0.9375,
      "learning_rate": 0.000499967461370544,
      "loss": 6.5905,
      "mean_token_accuracy": 0.11994630172848701,
      "num_tokens": 3284208.0,
      "step": 1615
    },
    {
      "entropy": 6.765414667129517,
      "epoch": 0.14157126627632613,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999669292807172,
      "loss": 6.5572,
      "mean_token_accuracy": 0.12147776633501053,
      "num_tokens": 3294216.0,
      "step": 1620
    },
    {
      "entropy": 6.838953018188477,
      "epoch": 0.14200821462903085,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999663928759066,
      "loss": 6.6214,
      "mean_token_accuracy": 0.11713840439915657,
      "num_tokens": 3304576.0,
      "step": 1625
    },
    {
      "entropy": 6.773513078689575,
      "epoch": 0.14244516298173557,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999658521561229,
      "loss": 6.6058,
      "mean_token_accuracy": 0.11238482668995857,
      "num_tokens": 3314214.0,
      "step": 1630
    },
    {
      "entropy": 6.8341858863830565,
      "epoch": 0.14288211133444026,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999653071213763,
      "loss": 6.6128,
      "mean_token_accuracy": 0.11472874283790588,
      "num_tokens": 3325527.0,
      "step": 1635
    },
    {
      "entropy": 6.689016628265381,
      "epoch": 0.14331905968714498,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999647577716771,
      "loss": 6.5222,
      "mean_token_accuracy": 0.12063723728060723,
      "num_tokens": 3334511.0,
      "step": 1640
    },
    {
      "entropy": 6.802826976776123,
      "epoch": 0.1437560080398497,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999642041070361,
      "loss": 6.5073,
      "mean_token_accuracy": 0.12565877586603164,
      "num_tokens": 3343922.0,
      "step": 1645
    },
    {
      "entropy": 6.721828937530518,
      "epoch": 0.1441929563925544,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999636461274637,
      "loss": 6.5278,
      "mean_token_accuracy": 0.11740849018096924,
      "num_tokens": 3353895.0,
      "step": 1650
    },
    {
      "entropy": 6.7812549591064455,
      "epoch": 0.1446299047452591,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999630838329706,
      "loss": 6.6036,
      "mean_token_accuracy": 0.11610466390848159,
      "num_tokens": 3364637.0,
      "step": 1655
    },
    {
      "entropy": 6.796862030029297,
      "epoch": 0.14506685309796383,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999625172235679,
      "loss": 6.5517,
      "mean_token_accuracy": 0.11584209352731704,
      "num_tokens": 3374308.0,
      "step": 1660
    },
    {
      "entropy": 6.753171968460083,
      "epoch": 0.14550380145066852,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499961946299266,
      "loss": 6.5295,
      "mean_token_accuracy": 0.12038384228944779,
      "num_tokens": 3383637.0,
      "step": 1665
    },
    {
      "entropy": 6.769080781936646,
      "epoch": 0.14594074980337324,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999613710600763,
      "loss": 6.6257,
      "mean_token_accuracy": 0.11186694428324699,
      "num_tokens": 3394150.0,
      "step": 1670
    },
    {
      "entropy": 6.839853715896607,
      "epoch": 0.14637769815607796,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999607915060094,
      "loss": 6.5858,
      "mean_token_accuracy": 0.11677002832293511,
      "num_tokens": 3404445.0,
      "step": 1675
    },
    {
      "entropy": 6.678188371658325,
      "epoch": 0.14681464650878268,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999602076370768,
      "loss": 6.4738,
      "mean_token_accuracy": 0.1292397640645504,
      "num_tokens": 3415183.0,
      "step": 1680
    },
    {
      "entropy": 6.721586894989014,
      "epoch": 0.14725159486148737,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999596194532895,
      "loss": 6.4685,
      "mean_token_accuracy": 0.12550642862915992,
      "num_tokens": 3424824.0,
      "step": 1685
    },
    {
      "entropy": 6.744260597229004,
      "epoch": 0.1476885432141921,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999590269546588,
      "loss": 6.5775,
      "mean_token_accuracy": 0.11547422781586647,
      "num_tokens": 3434789.0,
      "step": 1690
    },
    {
      "entropy": 6.6185546875,
      "epoch": 0.1481254915668968,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999584301411961,
      "loss": 6.4484,
      "mean_token_accuracy": 0.14269705563783647,
      "num_tokens": 3445810.0,
      "step": 1695
    },
    {
      "entropy": 6.805194997787476,
      "epoch": 0.1485624399196015,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999578290129129,
      "loss": 6.5926,
      "mean_token_accuracy": 0.11843555644154549,
      "num_tokens": 3455510.0,
      "step": 1700
    },
    {
      "entropy": 6.758487224578857,
      "epoch": 0.14899938827230622,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999572235698206,
      "loss": 6.5666,
      "mean_token_accuracy": 0.11734471693634987,
      "num_tokens": 3465785.0,
      "step": 1705
    },
    {
      "entropy": 6.741208553314209,
      "epoch": 0.14943633662501093,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999566138119308,
      "loss": 6.6292,
      "mean_token_accuracy": 0.11335316821932792,
      "num_tokens": 3477568.0,
      "step": 1710
    },
    {
      "entropy": 6.705219745635986,
      "epoch": 0.14987328497771563,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999559997392554,
      "loss": 6.4506,
      "mean_token_accuracy": 0.14270611405372619,
      "num_tokens": 3487703.0,
      "step": 1715
    },
    {
      "entropy": 6.711113119125367,
      "epoch": 0.15031023333042035,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999553813518061,
      "loss": 6.5284,
      "mean_token_accuracy": 0.11559826582670212,
      "num_tokens": 3497757.0,
      "step": 1720
    },
    {
      "entropy": 6.668199300765991,
      "epoch": 0.15074718168312506,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999547586495946,
      "loss": 6.4834,
      "mean_token_accuracy": 0.12288279235363006,
      "num_tokens": 3507594.0,
      "step": 1725
    },
    {
      "entropy": 6.773939371109009,
      "epoch": 0.15118413003582976,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999541316326329,
      "loss": 6.5773,
      "mean_token_accuracy": 0.1167720839381218,
      "num_tokens": 3517889.0,
      "step": 1730
    },
    {
      "entropy": 6.674737501144409,
      "epoch": 0.15162107838853447,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999535003009332,
      "loss": 6.5629,
      "mean_token_accuracy": 0.12246896028518676,
      "num_tokens": 3528158.0,
      "step": 1735
    },
    {
      "entropy": 6.775814867019653,
      "epoch": 0.1520580267412392,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999528646545074,
      "loss": 6.5627,
      "mean_token_accuracy": 0.1212423361837864,
      "num_tokens": 3539099.0,
      "step": 1740
    },
    {
      "entropy": 6.744331502914429,
      "epoch": 0.15249497509394389,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999522246933678,
      "loss": 6.5672,
      "mean_token_accuracy": 0.11636029034852982,
      "num_tokens": 3549819.0,
      "step": 1745
    },
    {
      "entropy": 6.822514200210572,
      "epoch": 0.1529319234466486,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999515804175268,
      "loss": 6.582,
      "mean_token_accuracy": 0.11982138305902482,
      "num_tokens": 3560022.0,
      "step": 1750
    },
    {
      "entropy": 6.689289999008179,
      "epoch": 0.15336887179935332,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999509318269964,
      "loss": 6.5051,
      "mean_token_accuracy": 0.120124302059412,
      "num_tokens": 3569681.0,
      "step": 1755
    },
    {
      "entropy": 6.7860630512237545,
      "epoch": 0.15380582015205801,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999502789217894,
      "loss": 6.5971,
      "mean_token_accuracy": 0.12079088762402534,
      "num_tokens": 3580731.0,
      "step": 1760
    },
    {
      "entropy": 6.683336973190308,
      "epoch": 0.15424276850476273,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999496217019181,
      "loss": 6.5719,
      "mean_token_accuracy": 0.12470733895897865,
      "num_tokens": 3590482.0,
      "step": 1765
    },
    {
      "entropy": 6.740330457687378,
      "epoch": 0.15467971685746745,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999489601673952,
      "loss": 6.55,
      "mean_token_accuracy": 0.11436792835593224,
      "num_tokens": 3600988.0,
      "step": 1770
    },
    {
      "entropy": 6.879163455963135,
      "epoch": 0.15511666521017214,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999482943182334,
      "loss": 6.6141,
      "mean_token_accuracy": 0.11543426737189293,
      "num_tokens": 3611226.0,
      "step": 1775
    },
    {
      "entropy": 6.648011779785156,
      "epoch": 0.15555361356287686,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999476241544454,
      "loss": 6.5402,
      "mean_token_accuracy": 0.12013219818472862,
      "num_tokens": 3621324.0,
      "step": 1780
    },
    {
      "entropy": 6.7309770584106445,
      "epoch": 0.15599056191558158,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999469496760441,
      "loss": 6.4611,
      "mean_token_accuracy": 0.12005406022071838,
      "num_tokens": 3631650.0,
      "step": 1785
    },
    {
      "entropy": 6.717350196838379,
      "epoch": 0.1564275102682863,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999462708830425,
      "loss": 6.4712,
      "mean_token_accuracy": 0.13633796572685242,
      "num_tokens": 3641173.0,
      "step": 1790
    },
    {
      "entropy": 6.669033765792847,
      "epoch": 0.156864458620991,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999455877754535,
      "loss": 6.4952,
      "mean_token_accuracy": 0.12263205051422119,
      "num_tokens": 3651846.0,
      "step": 1795
    },
    {
      "entropy": 6.672937488555908,
      "epoch": 0.1573014069736957,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999449003532903,
      "loss": 6.4924,
      "mean_token_accuracy": 0.12233219742774963,
      "num_tokens": 3661174.0,
      "step": 1800
    },
    {
      "entropy": 6.715725040435791,
      "epoch": 0.15773835532640043,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999442086165661,
      "loss": 6.4768,
      "mean_token_accuracy": 0.1226891666650772,
      "num_tokens": 3671042.0,
      "step": 1805
    },
    {
      "entropy": 6.707070398330688,
      "epoch": 0.15817530367910512,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499943512565294,
      "loss": 6.456,
      "mean_token_accuracy": 0.11890220195055008,
      "num_tokens": 3681206.0,
      "step": 1810
    },
    {
      "entropy": 6.718597173690796,
      "epoch": 0.15861225203180984,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999428121994874,
      "loss": 6.5416,
      "mean_token_accuracy": 0.12524761259555817,
      "num_tokens": 3691748.0,
      "step": 1815
    },
    {
      "entropy": 6.644932508468628,
      "epoch": 0.15904920038451456,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999421075191599,
      "loss": 6.4351,
      "mean_token_accuracy": 0.12655914798378945,
      "num_tokens": 3701730.0,
      "step": 1820
    },
    {
      "entropy": 6.7265056610107425,
      "epoch": 0.15948614873721925,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999413985243249,
      "loss": 6.5473,
      "mean_token_accuracy": 0.12333036363124847,
      "num_tokens": 3712525.0,
      "step": 1825
    },
    {
      "entropy": 6.780407094955445,
      "epoch": 0.15992309708992397,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499940685214996,
      "loss": 6.5108,
      "mean_token_accuracy": 0.12247696071863175,
      "num_tokens": 3722016.0,
      "step": 1830
    },
    {
      "entropy": 6.7228175640106205,
      "epoch": 0.1603600454426287,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999399675911869,
      "loss": 6.4086,
      "mean_token_accuracy": 0.128486368060112,
      "num_tokens": 3731299.0,
      "step": 1835
    },
    {
      "entropy": 6.601348638534546,
      "epoch": 0.16079699379533338,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999392456529113,
      "loss": 6.4831,
      "mean_token_accuracy": 0.12748103588819504,
      "num_tokens": 3740922.0,
      "step": 1840
    },
    {
      "entropy": 6.624817132949829,
      "epoch": 0.1612339421480381,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999385194001831,
      "loss": 6.4599,
      "mean_token_accuracy": 0.12713200226426125,
      "num_tokens": 3750339.0,
      "step": 1845
    },
    {
      "entropy": 6.703272294998169,
      "epoch": 0.16167089050074282,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999377888330164,
      "loss": 6.4525,
      "mean_token_accuracy": 0.12768443152308465,
      "num_tokens": 3761379.0,
      "step": 1850
    },
    {
      "entropy": 6.649005270004272,
      "epoch": 0.1621078388534475,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999370539514249,
      "loss": 6.5191,
      "mean_token_accuracy": 0.12053906321525573,
      "num_tokens": 3771649.0,
      "step": 1855
    },
    {
      "entropy": 6.601364612579346,
      "epoch": 0.16254478720615223,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999363147554229,
      "loss": 6.4493,
      "mean_token_accuracy": 0.12944458052515984,
      "num_tokens": 3782393.0,
      "step": 1860
    },
    {
      "entropy": 6.721639108657837,
      "epoch": 0.16298173555885695,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999355712450244,
      "loss": 6.5061,
      "mean_token_accuracy": 0.12632624581456184,
      "num_tokens": 3792849.0,
      "step": 1865
    },
    {
      "entropy": 6.676987981796264,
      "epoch": 0.16341868391156164,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999348234202438,
      "loss": 6.4481,
      "mean_token_accuracy": 0.1284789800643921,
      "num_tokens": 3803848.0,
      "step": 1870
    },
    {
      "entropy": 6.551715660095215,
      "epoch": 0.16385563226426636,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999340712810956,
      "loss": 6.3909,
      "mean_token_accuracy": 0.13192431330680848,
      "num_tokens": 3814704.0,
      "step": 1875
    },
    {
      "entropy": 6.705342388153076,
      "epoch": 0.16429258061697108,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999333148275939,
      "loss": 6.5028,
      "mean_token_accuracy": 0.11970377936959267,
      "num_tokens": 3824955.0,
      "step": 1880
    },
    {
      "entropy": 6.683031272888184,
      "epoch": 0.1647295289696758,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999325540597535,
      "loss": 6.4761,
      "mean_token_accuracy": 0.12891870439052583,
      "num_tokens": 3836540.0,
      "step": 1885
    },
    {
      "entropy": 6.583834505081176,
      "epoch": 0.1651664773223805,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999317889775887,
      "loss": 6.3804,
      "mean_token_accuracy": 0.1364469476044178,
      "num_tokens": 3846472.0,
      "step": 1890
    },
    {
      "entropy": 6.688028192520141,
      "epoch": 0.1656034256750852,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999310195811143,
      "loss": 6.482,
      "mean_token_accuracy": 0.12633047848939896,
      "num_tokens": 3857187.0,
      "step": 1895
    },
    {
      "entropy": 6.714043521881104,
      "epoch": 0.16604037402778993,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999302458703451,
      "loss": 6.4456,
      "mean_token_accuracy": 0.13051569014787673,
      "num_tokens": 3866259.0,
      "step": 1900
    },
    {
      "entropy": 6.573014926910401,
      "epoch": 0.16647732238049462,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499929467845296,
      "loss": 6.5043,
      "mean_token_accuracy": 0.11940461099147796,
      "num_tokens": 3876869.0,
      "step": 1905
    },
    {
      "entropy": 6.7059375762939455,
      "epoch": 0.16691427073319934,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999286855059818,
      "loss": 6.4832,
      "mean_token_accuracy": 0.12376384809613228,
      "num_tokens": 3888278.0,
      "step": 1910
    },
    {
      "entropy": 6.709025764465332,
      "epoch": 0.16735121908590406,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999278988524177,
      "loss": 6.5398,
      "mean_token_accuracy": 0.12542197927832605,
      "num_tokens": 3898648.0,
      "step": 1915
    },
    {
      "entropy": 6.6800124168396,
      "epoch": 0.16778816743860875,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999271078846184,
      "loss": 6.4707,
      "mean_token_accuracy": 0.12389178350567817,
      "num_tokens": 3908282.0,
      "step": 1920
    },
    {
      "entropy": 6.5305383682250975,
      "epoch": 0.16822511579131347,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999263126025994,
      "loss": 6.4439,
      "mean_token_accuracy": 0.1258118435740471,
      "num_tokens": 3918333.0,
      "step": 1925
    },
    {
      "entropy": 6.743747615814209,
      "epoch": 0.1686620641440182,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999255130063759,
      "loss": 6.451,
      "mean_token_accuracy": 0.12649790495634078,
      "num_tokens": 3927403.0,
      "step": 1930
    },
    {
      "entropy": 6.643372011184693,
      "epoch": 0.16909901249672288,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999247090959633,
      "loss": 6.4763,
      "mean_token_accuracy": 0.1255256250500679,
      "num_tokens": 3937352.0,
      "step": 1935
    },
    {
      "entropy": 6.593483400344849,
      "epoch": 0.1695359608494276,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999239008713767,
      "loss": 6.302,
      "mean_token_accuracy": 0.13238156586885452,
      "num_tokens": 3947164.0,
      "step": 1940
    },
    {
      "entropy": 6.632910919189453,
      "epoch": 0.16997290920213232,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999230883326319,
      "loss": 6.4789,
      "mean_token_accuracy": 0.12670883610844613,
      "num_tokens": 3956887.0,
      "step": 1945
    },
    {
      "entropy": 6.690495347976684,
      "epoch": 0.170409857554837,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999222714797445,
      "loss": 6.4517,
      "mean_token_accuracy": 0.125770603120327,
      "num_tokens": 3966542.0,
      "step": 1950
    },
    {
      "entropy": 6.6743687152862545,
      "epoch": 0.17084680590754173,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00049992145031273,
      "loss": 6.4149,
      "mean_token_accuracy": 0.13651399984955787,
      "num_tokens": 3976127.0,
      "step": 1955
    },
    {
      "entropy": 6.539837646484375,
      "epoch": 0.17128375426024645,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999206248316043,
      "loss": 6.3808,
      "mean_token_accuracy": 0.129552910476923,
      "num_tokens": 3987371.0,
      "step": 1960
    },
    {
      "entropy": 6.5061633586883545,
      "epoch": 0.17172070261295114,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499919795036383,
      "loss": 6.3923,
      "mean_token_accuracy": 0.13139066472649574,
      "num_tokens": 3997924.0,
      "step": 1965
    },
    {
      "entropy": 6.676038503646851,
      "epoch": 0.17215765096565586,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999189609270823,
      "loss": 6.4851,
      "mean_token_accuracy": 0.12391790822148323,
      "num_tokens": 4009150.0,
      "step": 1970
    },
    {
      "entropy": 6.673120498657227,
      "epoch": 0.17259459931836058,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000499918122503718,
      "loss": 6.4923,
      "mean_token_accuracy": 0.1261556513607502,
      "num_tokens": 4018899.0,
      "step": 1975
    },
    {
      "entropy": 6.587709045410156,
      "epoch": 0.17303154767106527,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999172797663064,
      "loss": 6.4382,
      "mean_token_accuracy": 0.12643140107393264,
      "num_tokens": 4028652.0,
      "step": 1980
    },
    {
      "entropy": 6.630083847045898,
      "epoch": 0.17346849602376999,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999164327148634,
      "loss": 6.4762,
      "mean_token_accuracy": 0.12422962710261345,
      "num_tokens": 4039819.0,
      "step": 1985
    },
    {
      "entropy": 6.656488132476807,
      "epoch": 0.1739054443764747,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999155813494053,
      "loss": 6.3795,
      "mean_token_accuracy": 0.13175911754369735,
      "num_tokens": 4050029.0,
      "step": 1990
    },
    {
      "entropy": 6.498795366287231,
      "epoch": 0.17434239272917942,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999147256699487,
      "loss": 6.4081,
      "mean_token_accuracy": 0.13507219925522804,
      "num_tokens": 4060497.0,
      "step": 1995
    },
    {
      "entropy": 6.6314342498779295,
      "epoch": 0.17477934108188412,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999138656765096,
      "loss": 6.3982,
      "mean_token_accuracy": 0.13177526667714118,
      "num_tokens": 4071477.0,
      "step": 2000
    },
    {
      "entropy": 6.677942276000977,
      "epoch": 0.17521628943458883,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999130013691049,
      "loss": 6.5053,
      "mean_token_accuracy": 0.12303248420357704,
      "num_tokens": 4081869.0,
      "step": 2005
    },
    {
      "entropy": 6.6382654190063475,
      "epoch": 0.17565323778729355,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999121327477509,
      "loss": 6.4715,
      "mean_token_accuracy": 0.1286171071231365,
      "num_tokens": 4092797.0,
      "step": 2010
    },
    {
      "entropy": 6.596001386642456,
      "epoch": 0.17609018613999824,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999112598124642,
      "loss": 6.4183,
      "mean_token_accuracy": 0.12707913890480996,
      "num_tokens": 4103688.0,
      "step": 2015
    },
    {
      "entropy": 6.670361614227295,
      "epoch": 0.17652713449270296,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999103825632618,
      "loss": 6.3977,
      "mean_token_accuracy": 0.12762655541300774,
      "num_tokens": 4114037.0,
      "step": 2020
    },
    {
      "entropy": 6.578020477294922,
      "epoch": 0.17696408284540768,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999095010001605,
      "loss": 6.3639,
      "mean_token_accuracy": 0.1284645155072212,
      "num_tokens": 4124164.0,
      "step": 2025
    },
    {
      "entropy": 6.659268522262574,
      "epoch": 0.17740103119811237,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000499908615123177,
      "loss": 6.4891,
      "mean_token_accuracy": 0.12261471077799797,
      "num_tokens": 4134225.0,
      "step": 2030
    },
    {
      "entropy": 6.626478099822998,
      "epoch": 0.1778379795508171,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999077249323284,
      "loss": 6.5314,
      "mean_token_accuracy": 0.12370351999998093,
      "num_tokens": 4145962.0,
      "step": 2035
    },
    {
      "entropy": 6.630055475234985,
      "epoch": 0.1782749279035218,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999068304276319,
      "loss": 6.4111,
      "mean_token_accuracy": 0.12499069198966026,
      "num_tokens": 4156572.0,
      "step": 2040
    },
    {
      "entropy": 6.6319256782531735,
      "epoch": 0.1787118762562265,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999059316091045,
      "loss": 6.4848,
      "mean_token_accuracy": 0.12434430345892906,
      "num_tokens": 4166960.0,
      "step": 2045
    },
    {
      "entropy": 6.6504209518432615,
      "epoch": 0.17914882460893122,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999050284767634,
      "loss": 6.4215,
      "mean_token_accuracy": 0.12848171815276146,
      "num_tokens": 4177833.0,
      "step": 2050
    },
    {
      "entropy": 6.527290678024292,
      "epoch": 0.17958577296163594,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999041210306261,
      "loss": 6.349,
      "mean_token_accuracy": 0.13348185643553734,
      "num_tokens": 4187858.0,
      "step": 2055
    },
    {
      "entropy": 6.70694146156311,
      "epoch": 0.18002272131434063,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999032092707098,
      "loss": 6.4954,
      "mean_token_accuracy": 0.12123825252056122,
      "num_tokens": 4198207.0,
      "step": 2060
    },
    {
      "entropy": 6.589783763885498,
      "epoch": 0.18045966966704535,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999022931970323,
      "loss": 6.3684,
      "mean_token_accuracy": 0.13207549452781678,
      "num_tokens": 4207594.0,
      "step": 2065
    },
    {
      "entropy": 6.5987868309021,
      "epoch": 0.18089661801975007,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999013728096107,
      "loss": 6.3884,
      "mean_token_accuracy": 0.1320921741425991,
      "num_tokens": 4217334.0,
      "step": 2070
    },
    {
      "entropy": 6.566110038757325,
      "epoch": 0.18133356637245476,
      "grad_norm": 1.0,
      "learning_rate": 0.000499900448108463,
      "loss": 6.3573,
      "mean_token_accuracy": 0.13278961703181266,
      "num_tokens": 4226407.0,
      "step": 2075
    },
    {
      "entropy": 6.574927711486817,
      "epoch": 0.18177051472515948,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499899519093607,
      "loss": 6.3179,
      "mean_token_accuracy": 0.13256798908114434,
      "num_tokens": 4237818.0,
      "step": 2080
    },
    {
      "entropy": 6.590792036056518,
      "epoch": 0.1822074630778642,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998985857650603,
      "loss": 6.4259,
      "mean_token_accuracy": 0.12269085347652435,
      "num_tokens": 4247783.0,
      "step": 2085
    },
    {
      "entropy": 6.530824613571167,
      "epoch": 0.18264441143056892,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998976481228409,
      "loss": 6.3893,
      "mean_token_accuracy": 0.127620892226696,
      "num_tokens": 4258049.0,
      "step": 2090
    },
    {
      "entropy": 6.59817476272583,
      "epoch": 0.1830813597832736,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004998967061669668,
      "loss": 6.3786,
      "mean_token_accuracy": 0.12377788498997688,
      "num_tokens": 4269357.0,
      "step": 2095
    },
    {
      "entropy": 6.52078332901001,
      "epoch": 0.18351830813597833,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998957598974559,
      "loss": 6.2953,
      "mean_token_accuracy": 0.14164591878652572,
      "num_tokens": 4279576.0,
      "step": 2100
    },
    {
      "entropy": 6.556330728530884,
      "epoch": 0.18395525648868305,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998948093143266,
      "loss": 6.3485,
      "mean_token_accuracy": 0.1299436144530773,
      "num_tokens": 4289761.0,
      "step": 2105
    },
    {
      "entropy": 6.555707740783691,
      "epoch": 0.18439220484138774,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499893854417597,
      "loss": 6.3027,
      "mean_token_accuracy": 0.13434590697288512,
      "num_tokens": 4299231.0,
      "step": 2110
    },
    {
      "entropy": 6.549562931060791,
      "epoch": 0.18482915319409246,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998928952072854,
      "loss": 6.4692,
      "mean_token_accuracy": 0.12588368579745293,
      "num_tokens": 4308687.0,
      "step": 2115
    },
    {
      "entropy": 6.6162632465362545,
      "epoch": 0.18526610154679718,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998919316834102,
      "loss": 6.4108,
      "mean_token_accuracy": 0.12744370847940445,
      "num_tokens": 4318183.0,
      "step": 2120
    },
    {
      "entropy": 6.607033395767212,
      "epoch": 0.18570304989950187,
      "grad_norm": 1.03125,
      "learning_rate": 0.00049989096384599,
      "loss": 6.4688,
      "mean_token_accuracy": 0.12834835201501846,
      "num_tokens": 4328482.0,
      "step": 2125
    },
    {
      "entropy": 6.611848211288452,
      "epoch": 0.1861399982522066,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004998899916950431,
      "loss": 6.327,
      "mean_token_accuracy": 0.12936775758862495,
      "num_tokens": 4339313.0,
      "step": 2130
    },
    {
      "entropy": 6.535123682022094,
      "epoch": 0.1865769466049113,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998890152305885,
      "loss": 6.448,
      "mean_token_accuracy": 0.12641657814383506,
      "num_tokens": 4350299.0,
      "step": 2135
    },
    {
      "entropy": 6.666938543319702,
      "epoch": 0.187013894957616,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998880344526447,
      "loss": 6.3671,
      "mean_token_accuracy": 0.13759638294577597,
      "num_tokens": 4361424.0,
      "step": 2140
    },
    {
      "entropy": 6.489548110961914,
      "epoch": 0.18745084331032072,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998870493612306,
      "loss": 6.3444,
      "mean_token_accuracy": 0.13067372441291808,
      "num_tokens": 4371480.0,
      "step": 2145
    },
    {
      "entropy": 6.4931432723999025,
      "epoch": 0.18788779166302544,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004998860599563649,
      "loss": 6.3826,
      "mean_token_accuracy": 0.1308943137526512,
      "num_tokens": 4381951.0,
      "step": 2150
    },
    {
      "entropy": 6.611650037765503,
      "epoch": 0.18832474001573013,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004998850662380668,
      "loss": 6.4303,
      "mean_token_accuracy": 0.12348892539739609,
      "num_tokens": 4392290.0,
      "step": 2155
    },
    {
      "entropy": 6.502107381820679,
      "epoch": 0.18876168836843485,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998840682063553,
      "loss": 6.2015,
      "mean_token_accuracy": 0.1464900940656662,
      "num_tokens": 4403350.0,
      "step": 2160
    },
    {
      "entropy": 6.535247755050659,
      "epoch": 0.18919863672113957,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998830658612496,
      "loss": 6.3957,
      "mean_token_accuracy": 0.13261518627405167,
      "num_tokens": 4413171.0,
      "step": 2165
    },
    {
      "entropy": 6.483105993270874,
      "epoch": 0.18963558507384426,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998820592027688,
      "loss": 6.3331,
      "mean_token_accuracy": 0.13156379163265228,
      "num_tokens": 4424371.0,
      "step": 2170
    },
    {
      "entropy": 6.603164386749268,
      "epoch": 0.19007253342654898,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998810482309322,
      "loss": 6.3293,
      "mean_token_accuracy": 0.1259684681892395,
      "num_tokens": 4434040.0,
      "step": 2175
    },
    {
      "entropy": 6.611560964584351,
      "epoch": 0.1905094817792537,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998800329457595,
      "loss": 6.4017,
      "mean_token_accuracy": 0.12301709577441215,
      "num_tokens": 4444663.0,
      "step": 2180
    },
    {
      "entropy": 6.527074813842773,
      "epoch": 0.1909464301319584,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998790133472697,
      "loss": 6.3604,
      "mean_token_accuracy": 0.12777338102459906,
      "num_tokens": 4454733.0,
      "step": 2185
    },
    {
      "entropy": 6.54938063621521,
      "epoch": 0.1913833784846631,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998779894354827,
      "loss": 6.3562,
      "mean_token_accuracy": 0.13192371279001236,
      "num_tokens": 4464286.0,
      "step": 2190
    },
    {
      "entropy": 6.550158071517944,
      "epoch": 0.19182032683736783,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000499876961210418,
      "loss": 6.3274,
      "mean_token_accuracy": 0.13941287770867347,
      "num_tokens": 4475005.0,
      "step": 2195
    },
    {
      "entropy": 6.516703271865845,
      "epoch": 0.19225727519007255,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998759286720953,
      "loss": 6.2863,
      "mean_token_accuracy": 0.1340303435921669,
      "num_tokens": 4484606.0,
      "step": 2200
    },
    {
      "entropy": 6.484053611755371,
      "epoch": 0.19269422354277724,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998748918205345,
      "loss": 6.4027,
      "mean_token_accuracy": 0.12878716513514518,
      "num_tokens": 4494930.0,
      "step": 2205
    },
    {
      "entropy": 6.576354551315307,
      "epoch": 0.19313117189548196,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998738506557554,
      "loss": 6.4254,
      "mean_token_accuracy": 0.12400433123111725,
      "num_tokens": 4505304.0,
      "step": 2210
    },
    {
      "entropy": 6.619117164611817,
      "epoch": 0.19356812024818668,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499872805177778,
      "loss": 6.3368,
      "mean_token_accuracy": 0.1325274594128132,
      "num_tokens": 4514333.0,
      "step": 2215
    },
    {
      "entropy": 6.427423667907715,
      "epoch": 0.19400506860089137,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998717553866225,
      "loss": 6.3963,
      "mean_token_accuracy": 0.1355910688638687,
      "num_tokens": 4525175.0,
      "step": 2220
    },
    {
      "entropy": 6.591354560852051,
      "epoch": 0.1944420169535961,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004998707012823087,
      "loss": 6.2709,
      "mean_token_accuracy": 0.13448979929089547,
      "num_tokens": 4534212.0,
      "step": 2225
    },
    {
      "entropy": 6.457765913009643,
      "epoch": 0.1948789653063008,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004998696428648573,
      "loss": 6.4433,
      "mean_token_accuracy": 0.1265091545879841,
      "num_tokens": 4545141.0,
      "step": 2230
    },
    {
      "entropy": 6.59932131767273,
      "epoch": 0.1953159136590055,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998685801342881,
      "loss": 6.3016,
      "mean_token_accuracy": 0.1358583614230156,
      "num_tokens": 4554410.0,
      "step": 2235
    },
    {
      "entropy": 6.4797813415527346,
      "epoch": 0.19575286201171022,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998675130906217,
      "loss": 6.3079,
      "mean_token_accuracy": 0.13782036304473877,
      "num_tokens": 4564397.0,
      "step": 2240
    },
    {
      "entropy": 6.455375337600708,
      "epoch": 0.19618981036441493,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998664417338787,
      "loss": 6.304,
      "mean_token_accuracy": 0.13762991055846213,
      "num_tokens": 4573986.0,
      "step": 2245
    },
    {
      "entropy": 6.545703554153443,
      "epoch": 0.19662675871711963,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004998653660640794,
      "loss": 6.2973,
      "mean_token_accuracy": 0.13973189294338226,
      "num_tokens": 4583449.0,
      "step": 2250
    },
    {
      "entropy": 6.559102153778076,
      "epoch": 0.19706370706982435,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998642860812447,
      "loss": 6.3459,
      "mean_token_accuracy": 0.12769658789038657,
      "num_tokens": 4593637.0,
      "step": 2255
    },
    {
      "entropy": 6.478931617736817,
      "epoch": 0.19750065542252906,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998632017853951,
      "loss": 6.3123,
      "mean_token_accuracy": 0.13895439505577087,
      "num_tokens": 4603865.0,
      "step": 2260
    },
    {
      "entropy": 6.565149307250977,
      "epoch": 0.19793760377523376,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998621131765514,
      "loss": 6.3555,
      "mean_token_accuracy": 0.13697974607348443,
      "num_tokens": 4613502.0,
      "step": 2265
    },
    {
      "entropy": 6.505105781555176,
      "epoch": 0.19837455212793847,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998610202547345,
      "loss": 6.3232,
      "mean_token_accuracy": 0.13375489190220832,
      "num_tokens": 4623163.0,
      "step": 2270
    },
    {
      "entropy": 6.5777122497558596,
      "epoch": 0.1988115004806432,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004998599230199654,
      "loss": 6.335,
      "mean_token_accuracy": 0.13985270708799363,
      "num_tokens": 4632850.0,
      "step": 2275
    },
    {
      "entropy": 6.4656078815460205,
      "epoch": 0.19924844883334789,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998588214722653,
      "loss": 6.3446,
      "mean_token_accuracy": 0.1339777112007141,
      "num_tokens": 4642858.0,
      "step": 2280
    },
    {
      "entropy": 6.614597415924072,
      "epoch": 0.1996853971860526,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998577156116551,
      "loss": 6.3736,
      "mean_token_accuracy": 0.12828925773501396,
      "num_tokens": 4653083.0,
      "step": 2285
    },
    {
      "entropy": 6.54793062210083,
      "epoch": 0.20012234553875732,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998566054381561,
      "loss": 6.3764,
      "mean_token_accuracy": 0.13061879575252533,
      "num_tokens": 4663353.0,
      "step": 2290
    },
    {
      "entropy": 6.474935913085938,
      "epoch": 0.20055929389146204,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004998554909517895,
      "loss": 6.2265,
      "mean_token_accuracy": 0.13771099895238875,
      "num_tokens": 4673878.0,
      "step": 2295
    },
    {
      "entropy": 6.482282829284668,
      "epoch": 0.20099624224416673,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998543721525769,
      "loss": 6.3145,
      "mean_token_accuracy": 0.13249781727790833,
      "num_tokens": 4683661.0,
      "step": 2300
    },
    {
      "entropy": 6.474229240417481,
      "epoch": 0.20143319059687145,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998532490405397,
      "loss": 6.3564,
      "mean_token_accuracy": 0.1330401547253132,
      "num_tokens": 4693817.0,
      "step": 2305
    },
    {
      "entropy": 6.484963512420654,
      "epoch": 0.20187013894957617,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998521216156991,
      "loss": 6.2864,
      "mean_token_accuracy": 0.13420451059937477,
      "num_tokens": 4703550.0,
      "step": 2310
    },
    {
      "entropy": 6.518954229354859,
      "epoch": 0.20230708730228086,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998509898780771,
      "loss": 6.4029,
      "mean_token_accuracy": 0.12855391204357147,
      "num_tokens": 4714872.0,
      "step": 2315
    },
    {
      "entropy": 6.501462841033936,
      "epoch": 0.20274403565498558,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004998498538276954,
      "loss": 6.29,
      "mean_token_accuracy": 0.13525153920054436,
      "num_tokens": 4725745.0,
      "step": 2320
    },
    {
      "entropy": 6.45821533203125,
      "epoch": 0.2031809840076903,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998487134645755,
      "loss": 6.3736,
      "mean_token_accuracy": 0.13093238174915314,
      "num_tokens": 4735560.0,
      "step": 2325
    },
    {
      "entropy": 6.549660491943359,
      "epoch": 0.203617932360395,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998475687887396,
      "loss": 6.3601,
      "mean_token_accuracy": 0.12788162082433702,
      "num_tokens": 4746304.0,
      "step": 2330
    },
    {
      "entropy": 6.57174506187439,
      "epoch": 0.2040548807130997,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998464198002095,
      "loss": 6.3098,
      "mean_token_accuracy": 0.13458219319581985,
      "num_tokens": 4756200.0,
      "step": 2335
    },
    {
      "entropy": 6.5113461971282955,
      "epoch": 0.20449182906580443,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998452664990072,
      "loss": 6.3378,
      "mean_token_accuracy": 0.12682230174541473,
      "num_tokens": 4766249.0,
      "step": 2340
    },
    {
      "entropy": 6.55569429397583,
      "epoch": 0.20492877741850912,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998441088851548,
      "loss": 6.3573,
      "mean_token_accuracy": 0.13034775629639625,
      "num_tokens": 4776510.0,
      "step": 2345
    },
    {
      "entropy": 6.524681043624878,
      "epoch": 0.20536572577121384,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004998429469586748,
      "loss": 6.3306,
      "mean_token_accuracy": 0.13234503641724588,
      "num_tokens": 4786798.0,
      "step": 2350
    },
    {
      "entropy": 6.471218490600586,
      "epoch": 0.20580267412391856,
      "grad_norm": 0.953125,
      "learning_rate": 0.000499841780719589,
      "loss": 6.27,
      "mean_token_accuracy": 0.12901424914598464,
      "num_tokens": 4796001.0,
      "step": 2355
    },
    {
      "entropy": 6.446377468109131,
      "epoch": 0.20623962247662325,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004998406101679202,
      "loss": 6.2392,
      "mean_token_accuracy": 0.13897980153560638,
      "num_tokens": 4807634.0,
      "step": 2360
    },
    {
      "entropy": 6.482134389877319,
      "epoch": 0.20667657082932797,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998394353036906,
      "loss": 6.2602,
      "mean_token_accuracy": 0.14035317897796631,
      "num_tokens": 4817840.0,
      "step": 2365
    },
    {
      "entropy": 6.448006582260132,
      "epoch": 0.2071135191820327,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998382561269229,
      "loss": 6.2297,
      "mean_token_accuracy": 0.13662357553839682,
      "num_tokens": 4827186.0,
      "step": 2370
    },
    {
      "entropy": 6.418888902664184,
      "epoch": 0.20755046753473738,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004998370726376395,
      "loss": 6.2963,
      "mean_token_accuracy": 0.12804412618279457,
      "num_tokens": 4837140.0,
      "step": 2375
    },
    {
      "entropy": 6.525313377380371,
      "epoch": 0.2079874158874421,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998358848358634,
      "loss": 6.3115,
      "mean_token_accuracy": 0.13363988548517228,
      "num_tokens": 4846907.0,
      "step": 2380
    },
    {
      "entropy": 6.421245479583741,
      "epoch": 0.20842436424014682,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000499834692721617,
      "loss": 6.3246,
      "mean_token_accuracy": 0.13632143288850784,
      "num_tokens": 4856165.0,
      "step": 2385
    },
    {
      "entropy": 6.532106208801269,
      "epoch": 0.2088613125928515,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998334962949237,
      "loss": 6.1893,
      "mean_token_accuracy": 0.14790109992027284,
      "num_tokens": 4865308.0,
      "step": 2390
    },
    {
      "entropy": 6.363882780075073,
      "epoch": 0.20929826094555623,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004998322955558059,
      "loss": 6.2627,
      "mean_token_accuracy": 0.14519340097904204,
      "num_tokens": 4876286.0,
      "step": 2395
    },
    {
      "entropy": 6.495657873153687,
      "epoch": 0.20973520929826095,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004998310905042869,
      "loss": 6.2851,
      "mean_token_accuracy": 0.13914690390229226,
      "num_tokens": 4887227.0,
      "step": 2400
    },
    {
      "entropy": 6.47929368019104,
      "epoch": 0.21017215765096567,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004998298811403898,
      "loss": 6.3224,
      "mean_token_accuracy": 0.13582812398672103,
      "num_tokens": 4896994.0,
      "step": 2405
    },
    {
      "entropy": 6.486909008026123,
      "epoch": 0.21060910600367036,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998286674641378,
      "loss": 6.3204,
      "mean_token_accuracy": 0.13193437755107879,
      "num_tokens": 4907623.0,
      "step": 2410
    },
    {
      "entropy": 6.420040416717529,
      "epoch": 0.21104605435637508,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499827449475554,
      "loss": 6.213,
      "mean_token_accuracy": 0.1360305741429329,
      "num_tokens": 4917040.0,
      "step": 2415
    },
    {
      "entropy": 6.561966419219971,
      "epoch": 0.2114830027090798,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499826227174662,
      "loss": 6.3483,
      "mean_token_accuracy": 0.13053651601076127,
      "num_tokens": 4928390.0,
      "step": 2420
    },
    {
      "entropy": 6.418824338912964,
      "epoch": 0.2119199510617845,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004998250005614852,
      "loss": 6.3213,
      "mean_token_accuracy": 0.13254272788763047,
      "num_tokens": 4939040.0,
      "step": 2425
    },
    {
      "entropy": 6.432943868637085,
      "epoch": 0.2123568994144892,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000499823769636047,
      "loss": 6.3587,
      "mean_token_accuracy": 0.13161036968231202,
      "num_tokens": 4949915.0,
      "step": 2430
    },
    {
      "entropy": 6.4896979331970215,
      "epoch": 0.21279384776719393,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004998225343983712,
      "loss": 6.2475,
      "mean_token_accuracy": 0.13825453370809554,
      "num_tokens": 4960127.0,
      "step": 2435
    },
    {
      "entropy": 6.534148836135865,
      "epoch": 0.21323079611989862,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998212948484814,
      "loss": 6.3152,
      "mean_token_accuracy": 0.13082096055150033,
      "num_tokens": 4970152.0,
      "step": 2440
    },
    {
      "entropy": 6.409393072128296,
      "epoch": 0.21366774447260334,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004998200509864012,
      "loss": 6.2538,
      "mean_token_accuracy": 0.13555221557617186,
      "num_tokens": 4979745.0,
      "step": 2445
    },
    {
      "entropy": 6.533963584899903,
      "epoch": 0.21410469282530806,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998188028121547,
      "loss": 6.276,
      "mean_token_accuracy": 0.1419918142259121,
      "num_tokens": 4989924.0,
      "step": 2450
    },
    {
      "entropy": 6.393548154830933,
      "epoch": 0.21454164117801275,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998175503257658,
      "loss": 6.2304,
      "mean_token_accuracy": 0.14116282463073732,
      "num_tokens": 4999698.0,
      "step": 2455
    },
    {
      "entropy": 6.470434617996216,
      "epoch": 0.21497858953071747,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004998162935272585,
      "loss": 6.3248,
      "mean_token_accuracy": 0.13296620473265647,
      "num_tokens": 5010625.0,
      "step": 2460
    },
    {
      "entropy": 6.392181825637818,
      "epoch": 0.2154155378834222,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004998150324166568,
      "loss": 6.2241,
      "mean_token_accuracy": 0.14569925218820573,
      "num_tokens": 5019740.0,
      "step": 2465
    },
    {
      "entropy": 6.5314465999603275,
      "epoch": 0.21585248623612688,
      "grad_norm": 0.8203125,
      "learning_rate": 0.000499813766993985,
      "loss": 6.336,
      "mean_token_accuracy": 0.13134812116622924,
      "num_tokens": 5030601.0,
      "step": 2470
    },
    {
      "entropy": 6.560648441314697,
      "epoch": 0.2162894345888316,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998124972592674,
      "loss": 6.2817,
      "mean_token_accuracy": 0.1379701852798462,
      "num_tokens": 5040567.0,
      "step": 2475
    },
    {
      "entropy": 6.4075439453125,
      "epoch": 0.21672638294153632,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998112232125283,
      "loss": 6.1694,
      "mean_token_accuracy": 0.140249789506197,
      "num_tokens": 5049512.0,
      "step": 2480
    },
    {
      "entropy": 6.388670444488525,
      "epoch": 0.217163331294241,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000499809944853792,
      "loss": 6.359,
      "mean_token_accuracy": 0.12962737679481506,
      "num_tokens": 5060986.0,
      "step": 2485
    },
    {
      "entropy": 6.469041967391968,
      "epoch": 0.21760027964694573,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998086621830833,
      "loss": 6.2681,
      "mean_token_accuracy": 0.14133114963769913,
      "num_tokens": 5071492.0,
      "step": 2490
    },
    {
      "entropy": 6.473101425170898,
      "epoch": 0.21803722799965045,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998073752004267,
      "loss": 6.2934,
      "mean_token_accuracy": 0.13215823620557784,
      "num_tokens": 5081722.0,
      "step": 2495
    },
    {
      "entropy": 6.478217506408692,
      "epoch": 0.21847417635235516,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998060839058467,
      "loss": 6.2577,
      "mean_token_accuracy": 0.13338202238082886,
      "num_tokens": 5091906.0,
      "step": 2500
    },
    {
      "entropy": 6.408456039428711,
      "epoch": 0.21891112470505986,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998047882993683,
      "loss": 6.2437,
      "mean_token_accuracy": 0.1377291664481163,
      "num_tokens": 5102025.0,
      "step": 2505
    },
    {
      "entropy": 6.419159936904907,
      "epoch": 0.21934807305776458,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004998034883810162,
      "loss": 6.2568,
      "mean_token_accuracy": 0.13685641512274743,
      "num_tokens": 5112354.0,
      "step": 2510
    },
    {
      "entropy": 6.460919380187988,
      "epoch": 0.2197850214104693,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004998021841508155,
      "loss": 6.1919,
      "mean_token_accuracy": 0.14290908649563788,
      "num_tokens": 5121695.0,
      "step": 2515
    },
    {
      "entropy": 6.459331607818603,
      "epoch": 0.22022196976317399,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998008756087911,
      "loss": 6.3082,
      "mean_token_accuracy": 0.1325561910867691,
      "num_tokens": 5132561.0,
      "step": 2520
    },
    {
      "entropy": 6.490088653564453,
      "epoch": 0.2206589181158787,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499799562754968,
      "loss": 6.3357,
      "mean_token_accuracy": 0.1301969990134239,
      "num_tokens": 5141981.0,
      "step": 2525
    },
    {
      "entropy": 6.422775936126709,
      "epoch": 0.22109586646858342,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997982455893716,
      "loss": 6.2075,
      "mean_token_accuracy": 0.13869994953274728,
      "num_tokens": 5151550.0,
      "step": 2530
    },
    {
      "entropy": 6.518757390975952,
      "epoch": 0.22153281482128812,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997969241120271,
      "loss": 6.271,
      "mean_token_accuracy": 0.1292865790426731,
      "num_tokens": 5161679.0,
      "step": 2535
    },
    {
      "entropy": 6.383720397949219,
      "epoch": 0.22196976317399283,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997955983229597,
      "loss": 6.3304,
      "mean_token_accuracy": 0.13600479513406755,
      "num_tokens": 5171575.0,
      "step": 2540
    },
    {
      "entropy": 6.425492429733277,
      "epoch": 0.22240671152669755,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499794268222195,
      "loss": 6.1906,
      "mean_token_accuracy": 0.1381123498082161,
      "num_tokens": 5180152.0,
      "step": 2545
    },
    {
      "entropy": 6.384013509750366,
      "epoch": 0.22284365987940224,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997929338097583,
      "loss": 6.2734,
      "mean_token_accuracy": 0.133567613363266,
      "num_tokens": 5190518.0,
      "step": 2550
    },
    {
      "entropy": 6.496050786972046,
      "epoch": 0.22328060823210696,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004997915950856755,
      "loss": 6.3065,
      "mean_token_accuracy": 0.12552305087447166,
      "num_tokens": 5201729.0,
      "step": 2555
    },
    {
      "entropy": 6.451396703720093,
      "epoch": 0.22371755658481168,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997902520499721,
      "loss": 6.2854,
      "mean_token_accuracy": 0.1372365988790989,
      "num_tokens": 5211940.0,
      "step": 2560
    },
    {
      "entropy": 6.411359596252441,
      "epoch": 0.22415450493751637,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997889047026738,
      "loss": 6.2217,
      "mean_token_accuracy": 0.1385316163301468,
      "num_tokens": 5221028.0,
      "step": 2565
    },
    {
      "entropy": 6.40747299194336,
      "epoch": 0.2245914532902211,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997875530438066,
      "loss": 6.2034,
      "mean_token_accuracy": 0.1430547997355461,
      "num_tokens": 5230498.0,
      "step": 2570
    },
    {
      "entropy": 6.414899444580078,
      "epoch": 0.2250284016429258,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004997861970733963,
      "loss": 6.2426,
      "mean_token_accuracy": 0.1374136283993721,
      "num_tokens": 5240112.0,
      "step": 2575
    },
    {
      "entropy": 6.420394611358643,
      "epoch": 0.2254653499956305,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499784836791469,
      "loss": 6.213,
      "mean_token_accuracy": 0.14016053378582,
      "num_tokens": 5249201.0,
      "step": 2580
    },
    {
      "entropy": 6.429572916030883,
      "epoch": 0.22590229834833522,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004997834721980507,
      "loss": 6.2215,
      "mean_token_accuracy": 0.13842421472072602,
      "num_tokens": 5258541.0,
      "step": 2585
    },
    {
      "entropy": 6.5185013771057125,
      "epoch": 0.22633924670103994,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004997821032931676,
      "loss": 6.3291,
      "mean_token_accuracy": 0.1385103665292263,
      "num_tokens": 5268524.0,
      "step": 2590
    },
    {
      "entropy": 6.363296604156494,
      "epoch": 0.22677619505374463,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000499780730076846,
      "loss": 6.2425,
      "mean_token_accuracy": 0.14057053327560426,
      "num_tokens": 5280702.0,
      "step": 2595
    },
    {
      "entropy": 6.452259969711304,
      "epoch": 0.22721314340644935,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004997793525491122,
      "loss": 6.2208,
      "mean_token_accuracy": 0.1316538669168949,
      "num_tokens": 5291214.0,
      "step": 2600
    },
    {
      "entropy": 6.3743082046508786,
      "epoch": 0.22765009175915407,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997779707099926,
      "loss": 6.1768,
      "mean_token_accuracy": 0.14723527133464814,
      "num_tokens": 5300729.0,
      "step": 2605
    },
    {
      "entropy": 6.421396255493164,
      "epoch": 0.2280870401118588,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004997765845595138,
      "loss": 6.3155,
      "mean_token_accuracy": 0.13397107347846032,
      "num_tokens": 5310551.0,
      "step": 2610
    },
    {
      "entropy": 6.42731294631958,
      "epoch": 0.22852398846456348,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997751940977024,
      "loss": 6.185,
      "mean_token_accuracy": 0.14047921672463418,
      "num_tokens": 5320038.0,
      "step": 2615
    },
    {
      "entropy": 6.375219440460205,
      "epoch": 0.2289609368172682,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997737993245848,
      "loss": 6.2425,
      "mean_token_accuracy": 0.137275530397892,
      "num_tokens": 5329753.0,
      "step": 2620
    },
    {
      "entropy": 6.359013986587525,
      "epoch": 0.22939788516997292,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997724002401881,
      "loss": 6.2234,
      "mean_token_accuracy": 0.14074066802859306,
      "num_tokens": 5340546.0,
      "step": 2625
    },
    {
      "entropy": 6.377727031707764,
      "epoch": 0.2298348335226776,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997709968445391,
      "loss": 6.2944,
      "mean_token_accuracy": 0.1307218536734581,
      "num_tokens": 5352097.0,
      "step": 2630
    },
    {
      "entropy": 6.516751003265381,
      "epoch": 0.23027178187538233,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004997695891376644,
      "loss": 6.2301,
      "mean_token_accuracy": 0.13981532603502272,
      "num_tokens": 5361646.0,
      "step": 2635
    },
    {
      "entropy": 6.378463172912598,
      "epoch": 0.23070873022808705,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997681771195912,
      "loss": 6.2131,
      "mean_token_accuracy": 0.14355284944176674,
      "num_tokens": 5371695.0,
      "step": 2640
    },
    {
      "entropy": 6.387977123260498,
      "epoch": 0.23114567858079174,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997667607903466,
      "loss": 6.2837,
      "mean_token_accuracy": 0.13416371122002602,
      "num_tokens": 5381978.0,
      "step": 2645
    },
    {
      "entropy": 6.501465606689453,
      "epoch": 0.23158262693349646,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004997653401499578,
      "loss": 6.286,
      "mean_token_accuracy": 0.14000471904873849,
      "num_tokens": 5391891.0,
      "step": 2650
    },
    {
      "entropy": 6.419810676574707,
      "epoch": 0.23201957528620118,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000499763915198452,
      "loss": 6.2509,
      "mean_token_accuracy": 0.13402704000473023,
      "num_tokens": 5403172.0,
      "step": 2655
    },
    {
      "entropy": 6.39618444442749,
      "epoch": 0.23245652363890587,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004997624859358565,
      "loss": 6.2691,
      "mean_token_accuracy": 0.13881371021270753,
      "num_tokens": 5413130.0,
      "step": 2660
    },
    {
      "entropy": 6.355148601531982,
      "epoch": 0.2328934719916106,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997610523621988,
      "loss": 6.2918,
      "mean_token_accuracy": 0.13786400184035302,
      "num_tokens": 5423510.0,
      "step": 2665
    },
    {
      "entropy": 6.4528110980987545,
      "epoch": 0.2333304203443153,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997596144775063,
      "loss": 6.1061,
      "mean_token_accuracy": 0.1457971803843975,
      "num_tokens": 5433707.0,
      "step": 2670
    },
    {
      "entropy": 6.2398120880126955,
      "epoch": 0.23376736869702,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004997581722818067,
      "loss": 6.2262,
      "mean_token_accuracy": 0.1393609881401062,
      "num_tokens": 5444381.0,
      "step": 2675
    },
    {
      "entropy": 6.543811321258545,
      "epoch": 0.23420431704972472,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004997567257751275,
      "loss": 6.175,
      "mean_token_accuracy": 0.1374103344976902,
      "num_tokens": 5453905.0,
      "step": 2680
    },
    {
      "entropy": 6.345109462738037,
      "epoch": 0.23464126540242944,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004997552749574965,
      "loss": 6.2029,
      "mean_token_accuracy": 0.13802675306797027,
      "num_tokens": 5464839.0,
      "step": 2685
    },
    {
      "entropy": 6.386873865127564,
      "epoch": 0.23507821375513413,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997538198289416,
      "loss": 6.2275,
      "mean_token_accuracy": 0.13631488382816315,
      "num_tokens": 5475459.0,
      "step": 2690
    },
    {
      "entropy": 6.363434362411499,
      "epoch": 0.23551516210783885,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997523603894907,
      "loss": 6.2474,
      "mean_token_accuracy": 0.14079780876636505,
      "num_tokens": 5484886.0,
      "step": 2695
    },
    {
      "entropy": 6.415025997161865,
      "epoch": 0.23595211046054357,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004997508966391717,
      "loss": 6.2246,
      "mean_token_accuracy": 0.1376207336783409,
      "num_tokens": 5494803.0,
      "step": 2700
    },
    {
      "entropy": 6.335872220993042,
      "epoch": 0.2363890588132483,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004997494285780129,
      "loss": 6.1795,
      "mean_token_accuracy": 0.13543259873986244,
      "num_tokens": 5504997.0,
      "step": 2705
    },
    {
      "entropy": 6.385657262802124,
      "epoch": 0.23682600716595298,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997479562060421,
      "loss": 6.1725,
      "mean_token_accuracy": 0.14626708552241324,
      "num_tokens": 5514492.0,
      "step": 2710
    },
    {
      "entropy": 6.409390449523926,
      "epoch": 0.2372629555186577,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004997464795232877,
      "loss": 6.2201,
      "mean_token_accuracy": 0.13575974926352502,
      "num_tokens": 5524419.0,
      "step": 2715
    },
    {
      "entropy": 6.417884826660156,
      "epoch": 0.23769990387136242,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004997449985297782,
      "loss": 6.2417,
      "mean_token_accuracy": 0.13905856385827065,
      "num_tokens": 5534289.0,
      "step": 2720
    },
    {
      "entropy": 6.332362174987793,
      "epoch": 0.2381368522240671,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997435132255418,
      "loss": 6.1802,
      "mean_token_accuracy": 0.1442882761359215,
      "num_tokens": 5543034.0,
      "step": 2725
    },
    {
      "entropy": 6.458979845046997,
      "epoch": 0.23857380057677183,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004997420236106071,
      "loss": 6.2378,
      "mean_token_accuracy": 0.1366375796496868,
      "num_tokens": 5553291.0,
      "step": 2730
    },
    {
      "entropy": 6.407902002334595,
      "epoch": 0.23901074892947655,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004997405296850026,
      "loss": 6.2287,
      "mean_token_accuracy": 0.13833343163132666,
      "num_tokens": 5563772.0,
      "step": 2735
    },
    {
      "entropy": 6.36975212097168,
      "epoch": 0.23944769728218124,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997390314487569,
      "loss": 6.2664,
      "mean_token_accuracy": 0.13527763187885283,
      "num_tokens": 5574348.0,
      "step": 2740
    },
    {
      "entropy": 6.397701358795166,
      "epoch": 0.23988464563488596,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004997375289018989,
      "loss": 6.12,
      "mean_token_accuracy": 0.13704326897859573,
      "num_tokens": 5583491.0,
      "step": 2745
    },
    {
      "entropy": 6.280293273925781,
      "epoch": 0.24032159398759068,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997360220444574,
      "loss": 6.2807,
      "mean_token_accuracy": 0.13119822591543198,
      "num_tokens": 5594273.0,
      "step": 2750
    },
    {
      "entropy": 6.4317409038543705,
      "epoch": 0.24075854234029537,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997345108764611,
      "loss": 6.1803,
      "mean_token_accuracy": 0.1414405457675457,
      "num_tokens": 5604529.0,
      "step": 2755
    },
    {
      "entropy": 6.3399059772491455,
      "epoch": 0.2411954906930001,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997329953979391,
      "loss": 6.1615,
      "mean_token_accuracy": 0.1501401461660862,
      "num_tokens": 5614036.0,
      "step": 2760
    },
    {
      "entropy": 6.434519290924072,
      "epoch": 0.2416324390457048,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997314756089206,
      "loss": 6.2155,
      "mean_token_accuracy": 0.1381111390888691,
      "num_tokens": 5624036.0,
      "step": 2765
    },
    {
      "entropy": 6.3883912563323975,
      "epoch": 0.2420693873984095,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997299515094347,
      "loss": 6.2076,
      "mean_token_accuracy": 0.14164475947618485,
      "num_tokens": 5633987.0,
      "step": 2770
    },
    {
      "entropy": 6.329027700424194,
      "epoch": 0.24250633575111422,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997284230995105,
      "loss": 6.0789,
      "mean_token_accuracy": 0.15508053749799727,
      "num_tokens": 5643905.0,
      "step": 2775
    },
    {
      "entropy": 6.36656379699707,
      "epoch": 0.24294328410381894,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004997268903791773,
      "loss": 6.2894,
      "mean_token_accuracy": 0.13616579845547677,
      "num_tokens": 5654922.0,
      "step": 2780
    },
    {
      "entropy": 6.366892242431641,
      "epoch": 0.24338023245652363,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004997253533484647,
      "loss": 6.1128,
      "mean_token_accuracy": 0.14602213948965073,
      "num_tokens": 5664407.0,
      "step": 2785
    },
    {
      "entropy": 6.357431697845459,
      "epoch": 0.24381718080922835,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997238120074021,
      "loss": 6.1859,
      "mean_token_accuracy": 0.13616643026471137,
      "num_tokens": 5674524.0,
      "step": 2790
    },
    {
      "entropy": 6.405643796920776,
      "epoch": 0.24425412916193306,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004997222663560188,
      "loss": 6.228,
      "mean_token_accuracy": 0.13521140292286873,
      "num_tokens": 5684691.0,
      "step": 2795
    },
    {
      "entropy": 6.368612289428711,
      "epoch": 0.24469107751463778,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997207163943449,
      "loss": 6.1395,
      "mean_token_accuracy": 0.1438075728714466,
      "num_tokens": 5693683.0,
      "step": 2800
    },
    {
      "entropy": 6.34757628440857,
      "epoch": 0.24512802586734248,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004997191621224098,
      "loss": 6.2258,
      "mean_token_accuracy": 0.13985715731978415,
      "num_tokens": 5704394.0,
      "step": 2805
    },
    {
      "entropy": 6.330892181396484,
      "epoch": 0.2455649742200472,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004997176035402435,
      "loss": 6.139,
      "mean_token_accuracy": 0.14314275234937668,
      "num_tokens": 5715480.0,
      "step": 2810
    },
    {
      "entropy": 6.420187997817993,
      "epoch": 0.2460019225727519,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004997160406478758,
      "loss": 6.1379,
      "mean_token_accuracy": 0.1458885796368122,
      "num_tokens": 5725973.0,
      "step": 2815
    },
    {
      "entropy": 6.216833543777466,
      "epoch": 0.2464388709254566,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997144734453367,
      "loss": 6.1469,
      "mean_token_accuracy": 0.14866761416196822,
      "num_tokens": 5735798.0,
      "step": 2820
    },
    {
      "entropy": 6.44745454788208,
      "epoch": 0.24687581927816132,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004997129019326562,
      "loss": 6.2351,
      "mean_token_accuracy": 0.13342514783143997,
      "num_tokens": 5746623.0,
      "step": 2825
    },
    {
      "entropy": 6.438055801391601,
      "epoch": 0.24731276763086604,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997113261098645,
      "loss": 6.179,
      "mean_token_accuracy": 0.14569296687841415,
      "num_tokens": 5756526.0,
      "step": 2830
    },
    {
      "entropy": 6.3516847610473635,
      "epoch": 0.24774971598357073,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004997097459769919,
      "loss": 6.1627,
      "mean_token_accuracy": 0.14307619631290436,
      "num_tokens": 5766670.0,
      "step": 2835
    },
    {
      "entropy": 6.425496816635132,
      "epoch": 0.24818666433627545,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004997081615340685,
      "loss": 6.2188,
      "mean_token_accuracy": 0.1372871719300747,
      "num_tokens": 5776473.0,
      "step": 2840
    },
    {
      "entropy": 6.430360794067383,
      "epoch": 0.24862361268898017,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997065727811249,
      "loss": 6.2628,
      "mean_token_accuracy": 0.13984661772847176,
      "num_tokens": 5787059.0,
      "step": 2845
    },
    {
      "entropy": 6.357260465621948,
      "epoch": 0.24906056104168486,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004997049797181915,
      "loss": 6.2442,
      "mean_token_accuracy": 0.13402743339538575,
      "num_tokens": 5797367.0,
      "step": 2850
    },
    {
      "entropy": 6.306977415084839,
      "epoch": 0.24949750939438958,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997033823452988,
      "loss": 6.128,
      "mean_token_accuracy": 0.1462322786450386,
      "num_tokens": 5806704.0,
      "step": 2855
    },
    {
      "entropy": 6.339516878128052,
      "epoch": 0.2499344577470943,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004997017806624774,
      "loss": 6.2057,
      "mean_token_accuracy": 0.1426796108484268,
      "num_tokens": 5817797.0,
      "step": 2860
    },
    {
      "entropy": 6.462503385543823,
      "epoch": 0.250371406099799,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004997001746697582,
      "loss": 6.229,
      "mean_token_accuracy": 0.14381124824285507,
      "num_tokens": 5829270.0,
      "step": 2865
    },
    {
      "entropy": 6.333521842956543,
      "epoch": 0.2508083544525037,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004996985643671719,
      "loss": 6.1819,
      "mean_token_accuracy": 0.1438910335302353,
      "num_tokens": 5839214.0,
      "step": 2870
    },
    {
      "entropy": 6.331501054763794,
      "epoch": 0.2512453028052084,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004996969497547495,
      "loss": 6.1694,
      "mean_token_accuracy": 0.13977854549884797,
      "num_tokens": 5848171.0,
      "step": 2875
    },
    {
      "entropy": 6.337689781188965,
      "epoch": 0.25168225115791315,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004996953308325217,
      "loss": 6.0864,
      "mean_token_accuracy": 0.14665822237730025,
      "num_tokens": 5858399.0,
      "step": 2880
    },
    {
      "entropy": 6.259796857833862,
      "epoch": 0.25211919951061784,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996937076005198,
      "loss": 6.0955,
      "mean_token_accuracy": 0.1473238319158554,
      "num_tokens": 5868212.0,
      "step": 2885
    },
    {
      "entropy": 6.379367685317993,
      "epoch": 0.25255614786332253,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004996920800587749,
      "loss": 6.1579,
      "mean_token_accuracy": 0.1477441743016243,
      "num_tokens": 5877883.0,
      "step": 2890
    },
    {
      "entropy": 6.401600027084351,
      "epoch": 0.2529930962160273,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004996904482073181,
      "loss": 6.1745,
      "mean_token_accuracy": 0.1420045681297779,
      "num_tokens": 5887090.0,
      "step": 2895
    },
    {
      "entropy": 6.228640031814575,
      "epoch": 0.25343004456873197,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004996888120461808,
      "loss": 6.2163,
      "mean_token_accuracy": 0.13503760173916818,
      "num_tokens": 5897491.0,
      "step": 2900
    },
    {
      "entropy": 6.423250532150268,
      "epoch": 0.25386699292143666,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996871715753943,
      "loss": 6.2207,
      "mean_token_accuracy": 0.14232490286231042,
      "num_tokens": 5908494.0,
      "step": 2905
    },
    {
      "entropy": 6.44255633354187,
      "epoch": 0.2543039412741414,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00049968552679499,
      "loss": 6.2737,
      "mean_token_accuracy": 0.12691745460033416,
      "num_tokens": 5919708.0,
      "step": 2910
    },
    {
      "entropy": 6.353352165222168,
      "epoch": 0.2547408896268461,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996838777049998,
      "loss": 6.1368,
      "mean_token_accuracy": 0.14446713924407958,
      "num_tokens": 5930122.0,
      "step": 2915
    },
    {
      "entropy": 6.272373819351197,
      "epoch": 0.2551778379795508,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000499682224305455,
      "loss": 6.1361,
      "mean_token_accuracy": 0.13961574882268907,
      "num_tokens": 5940033.0,
      "step": 2920
    },
    {
      "entropy": 6.319667816162109,
      "epoch": 0.25561478633225554,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004996805665963874,
      "loss": 6.1342,
      "mean_token_accuracy": 0.141078320145607,
      "num_tokens": 5949335.0,
      "step": 2925
    },
    {
      "entropy": 6.390504455566406,
      "epoch": 0.25605173468496023,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996789045778287,
      "loss": 6.1704,
      "mean_token_accuracy": 0.14173810854554175,
      "num_tokens": 5959299.0,
      "step": 2930
    },
    {
      "entropy": 6.357183980941772,
      "epoch": 0.2564886830376649,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004996772382498111,
      "loss": 6.1365,
      "mean_token_accuracy": 0.13954563736915587,
      "num_tokens": 5970114.0,
      "step": 2935
    },
    {
      "entropy": 6.311671018600464,
      "epoch": 0.25692563139036967,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004996755676123662,
      "loss": 6.2433,
      "mean_token_accuracy": 0.1331164039671421,
      "num_tokens": 5979630.0,
      "step": 2940
    },
    {
      "entropy": 6.456780195236206,
      "epoch": 0.25736257974307436,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996738926655262,
      "loss": 6.2217,
      "mean_token_accuracy": 0.13437832593917848,
      "num_tokens": 5990190.0,
      "step": 2945
    },
    {
      "entropy": 6.370059871673584,
      "epoch": 0.25779952809577905,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004996722134093231,
      "loss": 6.2331,
      "mean_token_accuracy": 0.13279297351837158,
      "num_tokens": 6000706.0,
      "step": 2950
    },
    {
      "entropy": 6.242281103134156,
      "epoch": 0.2582364764484838,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996705298437894,
      "loss": 6.2095,
      "mean_token_accuracy": 0.14207181483507156,
      "num_tokens": 6011274.0,
      "step": 2955
    },
    {
      "entropy": 6.422606420516968,
      "epoch": 0.2586734248011885,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004996688419689572,
      "loss": 6.1129,
      "mean_token_accuracy": 0.14407445564866067,
      "num_tokens": 6021481.0,
      "step": 2960
    },
    {
      "entropy": 6.295702743530273,
      "epoch": 0.25911037315389324,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004996671497848589,
      "loss": 6.2045,
      "mean_token_accuracy": 0.14156209751963617,
      "num_tokens": 6032091.0,
      "step": 2965
    },
    {
      "entropy": 6.362182235717773,
      "epoch": 0.25954732150659793,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004996654532915269,
      "loss": 6.1971,
      "mean_token_accuracy": 0.13414182811975478,
      "num_tokens": 6043371.0,
      "step": 2970
    },
    {
      "entropy": 6.409013032913208,
      "epoch": 0.2599842698593026,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996637524889937,
      "loss": 6.136,
      "mean_token_accuracy": 0.1435550257563591,
      "num_tokens": 6052812.0,
      "step": 2975
    },
    {
      "entropy": 6.265077257156372,
      "epoch": 0.26042121821200737,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004996620473772921,
      "loss": 6.1223,
      "mean_token_accuracy": 0.14507505744695665,
      "num_tokens": 6064033.0,
      "step": 2980
    },
    {
      "entropy": 6.355175495147705,
      "epoch": 0.26085816656471206,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004996603379564546,
      "loss": 6.255,
      "mean_token_accuracy": 0.13743579238653184,
      "num_tokens": 6075837.0,
      "step": 2985
    },
    {
      "entropy": 6.47666974067688,
      "epoch": 0.26129511491741675,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004996586242265142,
      "loss": 6.1525,
      "mean_token_accuracy": 0.14620224833488465,
      "num_tokens": 6085514.0,
      "step": 2990
    },
    {
      "entropy": 6.294611263275146,
      "epoch": 0.2617320632701215,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004996569061875036,
      "loss": 6.0654,
      "mean_token_accuracy": 0.1461657240986824,
      "num_tokens": 6094992.0,
      "step": 2995
    },
    {
      "entropy": 6.339985752105713,
      "epoch": 0.2621690116228262,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004996551838394559,
      "loss": 6.1403,
      "mean_token_accuracy": 0.14486877620220184,
      "num_tokens": 6104841.0,
      "step": 3000
    },
    {
      "epoch": 0.2621690116228262,
      "eval_entropy": 6.096288530878612,
      "eval_loss": 6.168129920959473,
      "eval_mean_token_accuracy": 0.14990296079035464,
      "eval_num_tokens": 6104841.0,
      "eval_runtime": 21.503,
      "eval_samples_per_second": 1721.34,
      "eval_steps_per_second": 215.179,
      "step": 3000
    },
    {
      "entropy": 6.224702310562134,
      "epoch": 0.2626059599755309,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000499653457182404,
      "loss": 6.0341,
      "mean_token_accuracy": 0.14952232241630553,
      "num_tokens": 6114830.0,
      "step": 3005
    },
    {
      "entropy": 6.294047880172729,
      "epoch": 0.2630429083282356,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004996517262163811,
      "loss": 6.165,
      "mean_token_accuracy": 0.1426911659538746,
      "num_tokens": 6125165.0,
      "step": 3010
    },
    {
      "entropy": 6.358997631072998,
      "epoch": 0.2634798566809403,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996499909414204,
      "loss": 6.104,
      "mean_token_accuracy": 0.14181375056505202,
      "num_tokens": 6135670.0,
      "step": 3015
    },
    {
      "entropy": 6.169686126708984,
      "epoch": 0.263916805033645,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004996482513575551,
      "loss": 6.09,
      "mean_token_accuracy": 0.15284790247678756,
      "num_tokens": 6146055.0,
      "step": 3020
    },
    {
      "entropy": 6.235915470123291,
      "epoch": 0.26435375338634975,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996465074648186,
      "loss": 6.1109,
      "mean_token_accuracy": 0.1489691525697708,
      "num_tokens": 6155941.0,
      "step": 3025
    },
    {
      "entropy": 6.397075176239014,
      "epoch": 0.26479070173905445,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996447592632444,
      "loss": 6.1931,
      "mean_token_accuracy": 0.1449305720627308,
      "num_tokens": 6166291.0,
      "step": 3030
    },
    {
      "entropy": 6.3069055557250975,
      "epoch": 0.26522765009175914,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996430067528661,
      "loss": 6.1609,
      "mean_token_accuracy": 0.14132383316755295,
      "num_tokens": 6175294.0,
      "step": 3035
    },
    {
      "entropy": 6.38663682937622,
      "epoch": 0.2656645984444639,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004996412499337172,
      "loss": 6.155,
      "mean_token_accuracy": 0.14668313413858414,
      "num_tokens": 6185273.0,
      "step": 3040
    },
    {
      "entropy": 6.209409904479981,
      "epoch": 0.2661015467971686,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004996394888058314,
      "loss": 6.1117,
      "mean_token_accuracy": 0.15102686285972594,
      "num_tokens": 6195075.0,
      "step": 3045
    },
    {
      "entropy": 6.2139918327331545,
      "epoch": 0.26653849514987327,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004996377233692425,
      "loss": 5.9907,
      "mean_token_accuracy": 0.15785050243139268,
      "num_tokens": 6206018.0,
      "step": 3050
    },
    {
      "entropy": 6.305926752090454,
      "epoch": 0.266975443502578,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004996359536239843,
      "loss": 6.1272,
      "mean_token_accuracy": 0.1415325164794922,
      "num_tokens": 6216577.0,
      "step": 3055
    },
    {
      "entropy": 6.370986652374268,
      "epoch": 0.2674123918552827,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996341795700908,
      "loss": 6.1581,
      "mean_token_accuracy": 0.1422954112291336,
      "num_tokens": 6226521.0,
      "step": 3060
    },
    {
      "entropy": 6.281765413284302,
      "epoch": 0.2678493402079874,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004996324012075961,
      "loss": 6.1444,
      "mean_token_accuracy": 0.1462144985795021,
      "num_tokens": 6236509.0,
      "step": 3065
    },
    {
      "entropy": 6.328299856185913,
      "epoch": 0.26828628856069214,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996306185365341,
      "loss": 6.118,
      "mean_token_accuracy": 0.1480477288365364,
      "num_tokens": 6246584.0,
      "step": 3070
    },
    {
      "entropy": 6.341874074935913,
      "epoch": 0.26872323691339683,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004996288315569393,
      "loss": 6.1382,
      "mean_token_accuracy": 0.140323144197464,
      "num_tokens": 6256881.0,
      "step": 3075
    },
    {
      "entropy": 6.351254653930664,
      "epoch": 0.2691601852661015,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004996270402688457,
      "loss": 6.1153,
      "mean_token_accuracy": 0.14938713908195494,
      "num_tokens": 6266380.0,
      "step": 3080
    },
    {
      "entropy": 6.268832874298096,
      "epoch": 0.2695971336188063,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996252446722877,
      "loss": 6.1585,
      "mean_token_accuracy": 0.14577163010835648,
      "num_tokens": 6274872.0,
      "step": 3085
    },
    {
      "entropy": 6.341122007369995,
      "epoch": 0.27003408197151096,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004996234447673,
      "loss": 6.1637,
      "mean_token_accuracy": 0.144112665951252,
      "num_tokens": 6286167.0,
      "step": 3090
    },
    {
      "entropy": 6.3181863784790036,
      "epoch": 0.27047103032421566,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004996216405539169,
      "loss": 6.12,
      "mean_token_accuracy": 0.14441015496850013,
      "num_tokens": 6295898.0,
      "step": 3095
    },
    {
      "entropy": 6.299571943283081,
      "epoch": 0.2709079786769204,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004996198320321728,
      "loss": 6.1443,
      "mean_token_accuracy": 0.14308567419648172,
      "num_tokens": 6306543.0,
      "step": 3100
    },
    {
      "entropy": 6.355363225936889,
      "epoch": 0.2713449270296251,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996180192021029,
      "loss": 6.1353,
      "mean_token_accuracy": 0.1403010219335556,
      "num_tokens": 6317709.0,
      "step": 3105
    },
    {
      "entropy": 6.2888991355896,
      "epoch": 0.2717818753823298,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996162020637415,
      "loss": 6.0907,
      "mean_token_accuracy": 0.14399196952581406,
      "num_tokens": 6326874.0,
      "step": 3110
    },
    {
      "entropy": 6.313392066955567,
      "epoch": 0.27221882373503453,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004996143806171238,
      "loss": 6.1879,
      "mean_token_accuracy": 0.1392950028181076,
      "num_tokens": 6337943.0,
      "step": 3115
    },
    {
      "entropy": 6.301876354217529,
      "epoch": 0.2726557720877392,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996125548622845,
      "loss": 6.2328,
      "mean_token_accuracy": 0.14419435262680053,
      "num_tokens": 6348442.0,
      "step": 3120
    },
    {
      "entropy": 6.296331119537354,
      "epoch": 0.2730927204404439,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004996107247992587,
      "loss": 6.1294,
      "mean_token_accuracy": 0.14568759128451347,
      "num_tokens": 6358675.0,
      "step": 3125
    },
    {
      "entropy": 6.355655097961426,
      "epoch": 0.27352966879314866,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004996088904280815,
      "loss": 6.1787,
      "mean_token_accuracy": 0.14147821068763733,
      "num_tokens": 6368631.0,
      "step": 3130
    },
    {
      "entropy": 6.336705732345581,
      "epoch": 0.27396661714585335,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004996070517487881,
      "loss": 6.0336,
      "mean_token_accuracy": 0.14569838643074035,
      "num_tokens": 6378709.0,
      "step": 3135
    },
    {
      "entropy": 6.250777339935302,
      "epoch": 0.27440356549855804,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004996052087614138,
      "loss": 6.1928,
      "mean_token_accuracy": 0.13466008007526398,
      "num_tokens": 6389686.0,
      "step": 3140
    },
    {
      "entropy": 6.319320440292358,
      "epoch": 0.2748405138512628,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996033614659938,
      "loss": 6.1576,
      "mean_token_accuracy": 0.14170488566160203,
      "num_tokens": 6399447.0,
      "step": 3145
    },
    {
      "entropy": 6.327855443954467,
      "epoch": 0.2752774622039675,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996015098625638,
      "loss": 6.1275,
      "mean_token_accuracy": 0.1448117695748806,
      "num_tokens": 6408891.0,
      "step": 3150
    },
    {
      "entropy": 6.264188194274903,
      "epoch": 0.2757144105566722,
      "grad_norm": 0.953125,
      "learning_rate": 0.000499599653951159,
      "loss": 6.1027,
      "mean_token_accuracy": 0.14686870127916335,
      "num_tokens": 6419471.0,
      "step": 3155
    },
    {
      "entropy": 6.289378261566162,
      "epoch": 0.2761513589093769,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004995977937318153,
      "loss": 6.0785,
      "mean_token_accuracy": 0.14780449122190475,
      "num_tokens": 6430445.0,
      "step": 3160
    },
    {
      "entropy": 6.35026593208313,
      "epoch": 0.2765883072620816,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000499595929204568,
      "loss": 6.1274,
      "mean_token_accuracy": 0.14415236487984656,
      "num_tokens": 6440821.0,
      "step": 3165
    },
    {
      "entropy": 6.256247234344483,
      "epoch": 0.27702525561478636,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995940603694533,
      "loss": 6.0543,
      "mean_token_accuracy": 0.14589458256959914,
      "num_tokens": 6450958.0,
      "step": 3170
    },
    {
      "entropy": 6.242272472381591,
      "epoch": 0.27746220396749105,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004995921872265067,
      "loss": 6.1482,
      "mean_token_accuracy": 0.14461207389831543,
      "num_tokens": 6462106.0,
      "step": 3175
    },
    {
      "entropy": 6.291719722747803,
      "epoch": 0.27789915232019574,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004995903097757643,
      "loss": 6.028,
      "mean_token_accuracy": 0.1520974174141884,
      "num_tokens": 6471890.0,
      "step": 3180
    },
    {
      "entropy": 6.233562469482422,
      "epoch": 0.2783361006729005,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995884280172622,
      "loss": 6.0575,
      "mean_token_accuracy": 0.14728479087352753,
      "num_tokens": 6481959.0,
      "step": 3185
    },
    {
      "entropy": 6.305636978149414,
      "epoch": 0.2787730490256052,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995865419510362,
      "loss": 6.0889,
      "mean_token_accuracy": 0.15161275789141654,
      "num_tokens": 6492086.0,
      "step": 3190
    },
    {
      "entropy": 6.191137504577637,
      "epoch": 0.27920999737830987,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995846515771227,
      "loss": 6.1256,
      "mean_token_accuracy": 0.1421547256410122,
      "num_tokens": 6502530.0,
      "step": 3195
    },
    {
      "entropy": 6.367141771316528,
      "epoch": 0.2796469457310146,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499582756895558,
      "loss": 6.1238,
      "mean_token_accuracy": 0.14860651567578315,
      "num_tokens": 6511600.0,
      "step": 3200
    },
    {
      "entropy": 6.251360607147217,
      "epoch": 0.2800838940837193,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004995808579063782,
      "loss": 6.1039,
      "mean_token_accuracy": 0.14189327657222747,
      "num_tokens": 6522712.0,
      "step": 3205
    },
    {
      "entropy": 6.074013710021973,
      "epoch": 0.280520842436424,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00049957895460962,
      "loss": 5.9661,
      "mean_token_accuracy": 0.16473694145679474,
      "num_tokens": 6532472.0,
      "step": 3210
    },
    {
      "entropy": 6.2806854248046875,
      "epoch": 0.28095779078912875,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004995770470053197,
      "loss": 6.013,
      "mean_token_accuracy": 0.16250228583812715,
      "num_tokens": 6541778.0,
      "step": 3215
    },
    {
      "entropy": 6.311449766159058,
      "epoch": 0.28139473914183344,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995751350935139,
      "loss": 6.0319,
      "mean_token_accuracy": 0.14264715015888213,
      "num_tokens": 6552368.0,
      "step": 3220
    },
    {
      "entropy": 6.21621880531311,
      "epoch": 0.28183168749453813,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004995732188742395,
      "loss": 6.0506,
      "mean_token_accuracy": 0.1555971994996071,
      "num_tokens": 6562593.0,
      "step": 3225
    },
    {
      "entropy": 6.237673807144165,
      "epoch": 0.2822686358472429,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004995712983475329,
      "loss": 6.1087,
      "mean_token_accuracy": 0.14431281089782716,
      "num_tokens": 6572695.0,
      "step": 3230
    },
    {
      "entropy": 6.238740015029907,
      "epoch": 0.28270558419994757,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995693735134313,
      "loss": 6.078,
      "mean_token_accuracy": 0.14644903242588042,
      "num_tokens": 6581671.0,
      "step": 3235
    },
    {
      "entropy": 6.246381616592407,
      "epoch": 0.28314253255265226,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004995674443719713,
      "loss": 6.0879,
      "mean_token_accuracy": 0.14340732991695404,
      "num_tokens": 6590460.0,
      "step": 3240
    },
    {
      "entropy": 6.257592296600341,
      "epoch": 0.283579480905357,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00049956551092319,
      "loss": 6.0881,
      "mean_token_accuracy": 0.14692215472459794,
      "num_tokens": 6599892.0,
      "step": 3245
    },
    {
      "entropy": 6.30704779624939,
      "epoch": 0.2840164292580617,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004995635731671247,
      "loss": 6.1701,
      "mean_token_accuracy": 0.13780031576752663,
      "num_tokens": 6610530.0,
      "step": 3250
    },
    {
      "entropy": 6.256830310821533,
      "epoch": 0.2844533776107664,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004995616311038124,
      "loss": 6.0783,
      "mean_token_accuracy": 0.14544694274663925,
      "num_tokens": 6621117.0,
      "step": 3255
    },
    {
      "entropy": 6.314076137542725,
      "epoch": 0.28489032596347114,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995596847332901,
      "loss": 6.109,
      "mean_token_accuracy": 0.14120630994439126,
      "num_tokens": 6631521.0,
      "step": 3260
    },
    {
      "entropy": 6.395283794403076,
      "epoch": 0.2853272743161758,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004995577340555956,
      "loss": 6.2116,
      "mean_token_accuracy": 0.13755208626389503,
      "num_tokens": 6642180.0,
      "step": 3265
    },
    {
      "entropy": 6.298916625976562,
      "epoch": 0.2857642226688805,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004995557790707661,
      "loss": 6.1346,
      "mean_token_accuracy": 0.14626011922955512,
      "num_tokens": 6652735.0,
      "step": 3270
    },
    {
      "entropy": 6.270556211471558,
      "epoch": 0.28620117102158527,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000499553819778839,
      "loss": 6.1224,
      "mean_token_accuracy": 0.14742177501320838,
      "num_tokens": 6663135.0,
      "step": 3275
    },
    {
      "entropy": 6.241147041320801,
      "epoch": 0.28663811937428996,
      "grad_norm": 1.0,
      "learning_rate": 0.000499551856179852,
      "loss": 6.115,
      "mean_token_accuracy": 0.14415645897388457,
      "num_tokens": 6673375.0,
      "step": 3280
    },
    {
      "entropy": 6.314465284347534,
      "epoch": 0.28707506772699465,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004995498882738428,
      "loss": 6.0788,
      "mean_token_accuracy": 0.14837748557329178,
      "num_tokens": 6683101.0,
      "step": 3285
    },
    {
      "entropy": 6.279365348815918,
      "epoch": 0.2875120160796994,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995479160608491,
      "loss": 6.1345,
      "mean_token_accuracy": 0.1512179344892502,
      "num_tokens": 6692763.0,
      "step": 3290
    },
    {
      "entropy": 6.303160667419434,
      "epoch": 0.2879489644324041,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995459395409085,
      "loss": 6.1101,
      "mean_token_accuracy": 0.14191073030233384,
      "num_tokens": 6702833.0,
      "step": 3295
    },
    {
      "entropy": 6.320908260345459,
      "epoch": 0.2883859127851088,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004995439587140593,
      "loss": 6.1138,
      "mean_token_accuracy": 0.1417900577187538,
      "num_tokens": 6713083.0,
      "step": 3300
    },
    {
      "entropy": 6.267672061920166,
      "epoch": 0.2888228611378135,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995419735803392,
      "loss": 6.0581,
      "mean_token_accuracy": 0.14316033944487572,
      "num_tokens": 6722501.0,
      "step": 3305
    },
    {
      "entropy": 6.268769693374634,
      "epoch": 0.2892598094905182,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004995399841397865,
      "loss": 6.0976,
      "mean_token_accuracy": 0.14535312205553055,
      "num_tokens": 6731961.0,
      "step": 3310
    },
    {
      "entropy": 6.246152687072754,
      "epoch": 0.2896967578432229,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004995379903924392,
      "loss": 6.0869,
      "mean_token_accuracy": 0.14219653755426406,
      "num_tokens": 6742576.0,
      "step": 3315
    },
    {
      "entropy": 6.298043632507325,
      "epoch": 0.29013370619592765,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004995359923383356,
      "loss": 6.0837,
      "mean_token_accuracy": 0.15008871704339982,
      "num_tokens": 6753066.0,
      "step": 3320
    },
    {
      "entropy": 6.248604440689087,
      "epoch": 0.29057065454863235,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000499533989977514,
      "loss": 6.2209,
      "mean_token_accuracy": 0.14179124012589456,
      "num_tokens": 6764455.0,
      "step": 3325
    },
    {
      "entropy": 6.316181373596192,
      "epoch": 0.29100760290133704,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004995319833100128,
      "loss": 6.0912,
      "mean_token_accuracy": 0.14703367799520492,
      "num_tokens": 6774655.0,
      "step": 3330
    },
    {
      "entropy": 6.225716686248779,
      "epoch": 0.2914445512540418,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004995299723358705,
      "loss": 6.0651,
      "mean_token_accuracy": 0.14900127649307252,
      "num_tokens": 6785942.0,
      "step": 3335
    },
    {
      "entropy": 6.234990215301513,
      "epoch": 0.2918814996067465,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004995279570551257,
      "loss": 6.0212,
      "mean_token_accuracy": 0.15330268144607545,
      "num_tokens": 6795355.0,
      "step": 3340
    },
    {
      "entropy": 6.214534568786621,
      "epoch": 0.29231844795945117,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000499525937467817,
      "loss": 6.0948,
      "mean_token_accuracy": 0.1484983503818512,
      "num_tokens": 6805579.0,
      "step": 3345
    },
    {
      "entropy": 6.2813645362854,
      "epoch": 0.2927553963121559,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004995239135739833,
      "loss": 6.1201,
      "mean_token_accuracy": 0.14244698882102966,
      "num_tokens": 6814878.0,
      "step": 3350
    },
    {
      "entropy": 6.326752996444702,
      "epoch": 0.2931923446648606,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004995218853736631,
      "loss": 6.0159,
      "mean_token_accuracy": 0.1553651809692383,
      "num_tokens": 6824533.0,
      "step": 3355
    },
    {
      "entropy": 6.077170658111572,
      "epoch": 0.29362929301756535,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004995198528668956,
      "loss": 5.9362,
      "mean_token_accuracy": 0.16292518228292466,
      "num_tokens": 6835529.0,
      "step": 3360
    },
    {
      "entropy": 6.242065668106079,
      "epoch": 0.29406624137027004,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004995178160537197,
      "loss": 6.1229,
      "mean_token_accuracy": 0.1476541742682457,
      "num_tokens": 6845679.0,
      "step": 3365
    },
    {
      "entropy": 6.325022602081299,
      "epoch": 0.29450318972297473,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004995157749341744,
      "loss": 6.1052,
      "mean_token_accuracy": 0.1412689656019211,
      "num_tokens": 6855876.0,
      "step": 3370
    },
    {
      "entropy": 6.243781900405883,
      "epoch": 0.2949401380756795,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995137295082989,
      "loss": 6.1566,
      "mean_token_accuracy": 0.14050692021846772,
      "num_tokens": 6866200.0,
      "step": 3375
    },
    {
      "entropy": 6.319889640808105,
      "epoch": 0.2953770864283842,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004995116797761324,
      "loss": 6.0424,
      "mean_token_accuracy": 0.14720476195216178,
      "num_tokens": 6875191.0,
      "step": 3380
    },
    {
      "entropy": 6.209931945800781,
      "epoch": 0.29581403478108886,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004995096257377143,
      "loss": 6.0153,
      "mean_token_accuracy": 0.15194161385297775,
      "num_tokens": 6885352.0,
      "step": 3385
    },
    {
      "entropy": 6.254030990600586,
      "epoch": 0.2962509831337936,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004995075673930839,
      "loss": 6.0845,
      "mean_token_accuracy": 0.14548156261444092,
      "num_tokens": 6895585.0,
      "step": 3390
    },
    {
      "entropy": 6.353338003158569,
      "epoch": 0.2966879314864983,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004995055047422808,
      "loss": 6.0656,
      "mean_token_accuracy": 0.14423816353082658,
      "num_tokens": 6905532.0,
      "step": 3395
    },
    {
      "entropy": 6.091818475723267,
      "epoch": 0.297124879839203,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004995034377853443,
      "loss": 6.0672,
      "mean_token_accuracy": 0.1501524642109871,
      "num_tokens": 6915835.0,
      "step": 3400
    },
    {
      "entropy": 6.255818319320679,
      "epoch": 0.29756182819190774,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004995013665223143,
      "loss": 6.1157,
      "mean_token_accuracy": 0.14370892941951752,
      "num_tokens": 6925964.0,
      "step": 3405
    },
    {
      "entropy": 6.2515569686889645,
      "epoch": 0.29799877654461243,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004994992909532305,
      "loss": 6.0548,
      "mean_token_accuracy": 0.14685370326042174,
      "num_tokens": 6936145.0,
      "step": 3410
    },
    {
      "entropy": 6.216746711730957,
      "epoch": 0.2984357248973171,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004994972110781326,
      "loss": 6.0295,
      "mean_token_accuracy": 0.1441150814294815,
      "num_tokens": 6945557.0,
      "step": 3415
    },
    {
      "entropy": 6.2465471744537355,
      "epoch": 0.29887267325002187,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994951268970606,
      "loss": 6.0671,
      "mean_token_accuracy": 0.14873078614473342,
      "num_tokens": 6955147.0,
      "step": 3420
    },
    {
      "entropy": 6.225246143341065,
      "epoch": 0.29930962160272656,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994930384100545,
      "loss": 6.0234,
      "mean_token_accuracy": 0.1503932483494282,
      "num_tokens": 6964024.0,
      "step": 3425
    },
    {
      "entropy": 6.2366087436676025,
      "epoch": 0.29974656995543125,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004994909456171541,
      "loss": 6.0832,
      "mean_token_accuracy": 0.1524054527282715,
      "num_tokens": 6974399.0,
      "step": 3430
    },
    {
      "entropy": 6.215136432647705,
      "epoch": 0.300183518308136,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004994888485183998,
      "loss": 6.0477,
      "mean_token_accuracy": 0.1459931790828705,
      "num_tokens": 6984324.0,
      "step": 3435
    },
    {
      "entropy": 6.234580993652344,
      "epoch": 0.3006204666608407,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499486747113832,
      "loss": 6.0401,
      "mean_token_accuracy": 0.15106956139206887,
      "num_tokens": 6995659.0,
      "step": 3440
    },
    {
      "entropy": 6.270761299133301,
      "epoch": 0.3010574150135454,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004994846414034905,
      "loss": 6.1333,
      "mean_token_accuracy": 0.142132718116045,
      "num_tokens": 7006463.0,
      "step": 3445
    },
    {
      "entropy": 6.280776119232177,
      "epoch": 0.30149436336625013,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499482531387416,
      "loss": 6.0374,
      "mean_token_accuracy": 0.14821409657597542,
      "num_tokens": 7016188.0,
      "step": 3450
    },
    {
      "entropy": 6.219879007339477,
      "epoch": 0.3019313117189548,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004994804170656489,
      "loss": 6.1041,
      "mean_token_accuracy": 0.14598944559693336,
      "num_tokens": 7028435.0,
      "step": 3455
    },
    {
      "entropy": 6.348649549484253,
      "epoch": 0.3023682600716595,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994782984382299,
      "loss": 6.0851,
      "mean_token_accuracy": 0.14334651231765747,
      "num_tokens": 7037992.0,
      "step": 3460
    },
    {
      "entropy": 6.21468677520752,
      "epoch": 0.30280520842436426,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004994761755051995,
      "loss": 6.0747,
      "mean_token_accuracy": 0.14589606001973152,
      "num_tokens": 7048743.0,
      "step": 3465
    },
    {
      "entropy": 6.230590534210205,
      "epoch": 0.30324215677706895,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004994740482665982,
      "loss": 6.052,
      "mean_token_accuracy": 0.1487992025911808,
      "num_tokens": 7059524.0,
      "step": 3470
    },
    {
      "entropy": 6.272332000732422,
      "epoch": 0.30367910512977364,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004994719167224672,
      "loss": 6.0623,
      "mean_token_accuracy": 0.14609384387731553,
      "num_tokens": 7069742.0,
      "step": 3475
    },
    {
      "entropy": 6.196537876129151,
      "epoch": 0.3041160534824784,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004994697808728472,
      "loss": 6.0927,
      "mean_token_accuracy": 0.15183692127466203,
      "num_tokens": 7079909.0,
      "step": 3480
    },
    {
      "entropy": 6.294587516784668,
      "epoch": 0.3045530018351831,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994676407177791,
      "loss": 6.1102,
      "mean_token_accuracy": 0.14163097068667413,
      "num_tokens": 7089136.0,
      "step": 3485
    },
    {
      "entropy": 6.25298924446106,
      "epoch": 0.30498995018788777,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004994654962573043,
      "loss": 6.1229,
      "mean_token_accuracy": 0.14757803454995155,
      "num_tokens": 7099621.0,
      "step": 3490
    },
    {
      "entropy": 6.26275053024292,
      "epoch": 0.3054268985405925,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004994633474914634,
      "loss": 5.9655,
      "mean_token_accuracy": 0.15322791337966918,
      "num_tokens": 7109599.0,
      "step": 3495
    },
    {
      "entropy": 6.184872579574585,
      "epoch": 0.3058638468932972,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004994611944202979,
      "loss": 6.0161,
      "mean_token_accuracy": 0.15335869640111924,
      "num_tokens": 7120733.0,
      "step": 3500
    },
    {
      "entropy": 6.243970537185669,
      "epoch": 0.3063007952460019,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994590370438492,
      "loss": 6.1016,
      "mean_token_accuracy": 0.14324494004249572,
      "num_tokens": 7129871.0,
      "step": 3505
    },
    {
      "entropy": 6.237084102630615,
      "epoch": 0.30673774359870665,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004994568753621585,
      "loss": 6.0239,
      "mean_token_accuracy": 0.14570069462060928,
      "num_tokens": 7139167.0,
      "step": 3510
    },
    {
      "entropy": 6.1348083972930905,
      "epoch": 0.30717469195141134,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004994547093752673,
      "loss": 6.0913,
      "mean_token_accuracy": 0.1418399527668953,
      "num_tokens": 7150534.0,
      "step": 3515
    },
    {
      "entropy": 6.273181390762329,
      "epoch": 0.30761164030411603,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004994525390832171,
      "loss": 6.0426,
      "mean_token_accuracy": 0.1464878052473068,
      "num_tokens": 7160472.0,
      "step": 3520
    },
    {
      "entropy": 6.2735999584197994,
      "epoch": 0.3080485886568208,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004994503644860497,
      "loss": 6.0574,
      "mean_token_accuracy": 0.14532073736190795,
      "num_tokens": 7171636.0,
      "step": 3525
    },
    {
      "entropy": 6.190199089050293,
      "epoch": 0.30848553700952547,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004994481855838067,
      "loss": 6.0059,
      "mean_token_accuracy": 0.15013523250818253,
      "num_tokens": 7182163.0,
      "step": 3530
    },
    {
      "entropy": 6.160256385803223,
      "epoch": 0.30892248536223016,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994460023765298,
      "loss": 6.0506,
      "mean_token_accuracy": 0.15310651659965516,
      "num_tokens": 7191694.0,
      "step": 3535
    },
    {
      "entropy": 6.231096267700195,
      "epoch": 0.3093594337149349,
      "grad_norm": 0.80859375,
      "learning_rate": 0.000499443814864261,
      "loss": 6.0758,
      "mean_token_accuracy": 0.15152389258146287,
      "num_tokens": 7202813.0,
      "step": 3540
    },
    {
      "entropy": 6.256965589523316,
      "epoch": 0.3097963820676396,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004994416230470422,
      "loss": 6.0596,
      "mean_token_accuracy": 0.15137551873922347,
      "num_tokens": 7213226.0,
      "step": 3545
    },
    {
      "entropy": 6.257351398468018,
      "epoch": 0.3102333304203443,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004994394269249155,
      "loss": 6.1036,
      "mean_token_accuracy": 0.14474155008792877,
      "num_tokens": 7223520.0,
      "step": 3550
    },
    {
      "entropy": 6.179442739486694,
      "epoch": 0.31067027877304904,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994372264979231,
      "loss": 5.9969,
      "mean_token_accuracy": 0.15374979078769685,
      "num_tokens": 7232651.0,
      "step": 3555
    },
    {
      "entropy": 6.286839246749878,
      "epoch": 0.3111072271257537,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004994350217661069,
      "loss": 6.0884,
      "mean_token_accuracy": 0.14863540530204772,
      "num_tokens": 7242064.0,
      "step": 3560
    },
    {
      "entropy": 6.264086580276489,
      "epoch": 0.3115441754784585,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004994328127295095,
      "loss": 6.0946,
      "mean_token_accuracy": 0.14491878896951677,
      "num_tokens": 7253048.0,
      "step": 3565
    },
    {
      "entropy": 6.206921911239624,
      "epoch": 0.31198112383116317,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994305993881731,
      "loss": 6.0382,
      "mean_token_accuracy": 0.15330977588891984,
      "num_tokens": 7262516.0,
      "step": 3570
    },
    {
      "entropy": 6.172793292999268,
      "epoch": 0.31241807218386786,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004994283817421402,
      "loss": 5.9715,
      "mean_token_accuracy": 0.15392128378152847,
      "num_tokens": 7272079.0,
      "step": 3575
    },
    {
      "entropy": 6.193983745574951,
      "epoch": 0.3128550205365726,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004994261597914534,
      "loss": 5.9526,
      "mean_token_accuracy": 0.1570977583527565,
      "num_tokens": 7282150.0,
      "step": 3580
    },
    {
      "entropy": 6.17117919921875,
      "epoch": 0.3132919688892773,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004994239335361552,
      "loss": 6.0641,
      "mean_token_accuracy": 0.14663082212209702,
      "num_tokens": 7291782.0,
      "step": 3585
    },
    {
      "entropy": 6.24909029006958,
      "epoch": 0.313728917241982,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004994217029762884,
      "loss": 6.0253,
      "mean_token_accuracy": 0.14997395873069763,
      "num_tokens": 7301472.0,
      "step": 3590
    },
    {
      "entropy": 6.206488752365113,
      "epoch": 0.31416586559468673,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004994194681118957,
      "loss": 5.9959,
      "mean_token_accuracy": 0.15701221525669098,
      "num_tokens": 7311456.0,
      "step": 3595
    },
    {
      "entropy": 6.25444860458374,
      "epoch": 0.3146028139473914,
      "grad_norm": 1.0,
      "learning_rate": 0.00049941722894302,
      "loss": 6.0853,
      "mean_token_accuracy": 0.1438571959733963,
      "num_tokens": 7321106.0,
      "step": 3600
    },
    {
      "entropy": 6.223976039886475,
      "epoch": 0.3150397623000961,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004994149854697042,
      "loss": 6.0828,
      "mean_token_accuracy": 0.1484446257352829,
      "num_tokens": 7331232.0,
      "step": 3605
    },
    {
      "entropy": 6.260455989837647,
      "epoch": 0.31547671065280086,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004994127376919916,
      "loss": 6.0635,
      "mean_token_accuracy": 0.14232216030359268,
      "num_tokens": 7341560.0,
      "step": 3610
    },
    {
      "entropy": 6.302635765075683,
      "epoch": 0.31591365900550555,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004994104856099249,
      "loss": 6.0585,
      "mean_token_accuracy": 0.1474137783050537,
      "num_tokens": 7350923.0,
      "step": 3615
    },
    {
      "entropy": 6.134376907348633,
      "epoch": 0.31635060735821025,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004994082292235476,
      "loss": 5.957,
      "mean_token_accuracy": 0.1528281092643738,
      "num_tokens": 7360121.0,
      "step": 3620
    },
    {
      "entropy": 6.140493297576905,
      "epoch": 0.316787555710915,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004994059685329027,
      "loss": 6.051,
      "mean_token_accuracy": 0.1463004857301712,
      "num_tokens": 7370823.0,
      "step": 3625
    },
    {
      "entropy": 6.238322973251343,
      "epoch": 0.3172245040636197,
      "grad_norm": 0.90625,
      "learning_rate": 0.000499403703538034,
      "loss": 6.0187,
      "mean_token_accuracy": 0.14543321505188941,
      "num_tokens": 7380102.0,
      "step": 3630
    },
    {
      "entropy": 6.188670015335083,
      "epoch": 0.3176614524163244,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004994014342389845,
      "loss": 5.9876,
      "mean_token_accuracy": 0.15175077617168425,
      "num_tokens": 7389909.0,
      "step": 3635
    },
    {
      "entropy": 6.230720472335816,
      "epoch": 0.3180984007690291,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004993991606357979,
      "loss": 6.0753,
      "mean_token_accuracy": 0.1465983882546425,
      "num_tokens": 7401092.0,
      "step": 3640
    },
    {
      "entropy": 6.141590261459351,
      "epoch": 0.3185353491217338,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004993968827285179,
      "loss": 5.9987,
      "mean_token_accuracy": 0.14415375292301177,
      "num_tokens": 7412155.0,
      "step": 3645
    },
    {
      "entropy": 6.137734603881836,
      "epoch": 0.3189722974744385,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004993946005171881,
      "loss": 5.9474,
      "mean_token_accuracy": 0.1587568461894989,
      "num_tokens": 7421981.0,
      "step": 3650
    },
    {
      "entropy": 6.249944257736206,
      "epoch": 0.31940924582714325,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993923140018523,
      "loss": 5.9741,
      "mean_token_accuracy": 0.15291838943958283,
      "num_tokens": 7432298.0,
      "step": 3655
    },
    {
      "entropy": 6.140120506286621,
      "epoch": 0.31984619417984794,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004993900231825542,
      "loss": 6.0239,
      "mean_token_accuracy": 0.15544686540961267,
      "num_tokens": 7442575.0,
      "step": 3660
    },
    {
      "entropy": 6.179750919342041,
      "epoch": 0.32028314253255263,
      "grad_norm": 0.875,
      "learning_rate": 0.0004993877280593381,
      "loss": 5.9891,
      "mean_token_accuracy": 0.16168759018182755,
      "num_tokens": 7453072.0,
      "step": 3665
    },
    {
      "entropy": 6.238027095794678,
      "epoch": 0.3207200908852574,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004993854286322477,
      "loss": 5.9635,
      "mean_token_accuracy": 0.1507369577884674,
      "num_tokens": 7463173.0,
      "step": 3670
    },
    {
      "entropy": 6.17151026725769,
      "epoch": 0.32115703923796207,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004993831249013272,
      "loss": 6.0739,
      "mean_token_accuracy": 0.14511893540620804,
      "num_tokens": 7473751.0,
      "step": 3675
    },
    {
      "entropy": 6.193846559524536,
      "epoch": 0.32159398759066676,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004993808168666207,
      "loss": 5.8687,
      "mean_token_accuracy": 0.16345068216323852,
      "num_tokens": 7483233.0,
      "step": 3680
    },
    {
      "entropy": 6.176863431930542,
      "epoch": 0.3220309359433715,
      "grad_norm": 0.875,
      "learning_rate": 0.0004993785045281726,
      "loss": 6.0164,
      "mean_token_accuracy": 0.14087698087096215,
      "num_tokens": 7493152.0,
      "step": 3685
    },
    {
      "entropy": 6.2610138893127445,
      "epoch": 0.3224678842960762,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004993761878860272,
      "loss": 5.9783,
      "mean_token_accuracy": 0.1512422576546669,
      "num_tokens": 7502463.0,
      "step": 3690
    },
    {
      "entropy": 6.228349208831787,
      "epoch": 0.3229048326487809,
      "grad_norm": 0.953125,
      "learning_rate": 0.000499373866940229,
      "loss": 6.0266,
      "mean_token_accuracy": 0.15055646747350693,
      "num_tokens": 7512301.0,
      "step": 3695
    },
    {
      "entropy": 6.1074363708496096,
      "epoch": 0.32334178100148564,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004993715416908225,
      "loss": 6.0667,
      "mean_token_accuracy": 0.1498277962207794,
      "num_tokens": 7522591.0,
      "step": 3700
    },
    {
      "entropy": 6.256154251098633,
      "epoch": 0.32377872935419033,
      "grad_norm": 1.0,
      "learning_rate": 0.0004993692121378521,
      "loss": 6.0528,
      "mean_token_accuracy": 0.14736614972352982,
      "num_tokens": 7532400.0,
      "step": 3705
    },
    {
      "entropy": 6.208133506774902,
      "epoch": 0.324215677706895,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004993668782813629,
      "loss": 5.9598,
      "mean_token_accuracy": 0.1492796003818512,
      "num_tokens": 7542163.0,
      "step": 3710
    },
    {
      "entropy": 6.163573503494263,
      "epoch": 0.32465262605959977,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004993645401213992,
      "loss": 5.952,
      "mean_token_accuracy": 0.15682226419448853,
      "num_tokens": 7551638.0,
      "step": 3715
    },
    {
      "entropy": 6.1557341575622555,
      "epoch": 0.32508957441230446,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004993621976580061,
      "loss": 6.04,
      "mean_token_accuracy": 0.15109157115221022,
      "num_tokens": 7561230.0,
      "step": 3720
    },
    {
      "entropy": 6.209139966964722,
      "epoch": 0.32552652276500915,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004993598508912285,
      "loss": 5.9201,
      "mean_token_accuracy": 0.15766422152519227,
      "num_tokens": 7570962.0,
      "step": 3725
    },
    {
      "entropy": 6.169910144805908,
      "epoch": 0.3259634711177139,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004993574998211114,
      "loss": 6.0946,
      "mean_token_accuracy": 0.14971239268779754,
      "num_tokens": 7581454.0,
      "step": 3730
    },
    {
      "entropy": 6.1962926387786865,
      "epoch": 0.3264004194704186,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004993551444476998,
      "loss": 5.9727,
      "mean_token_accuracy": 0.15400165021419526,
      "num_tokens": 7591497.0,
      "step": 3735
    },
    {
      "entropy": 6.216365623474121,
      "epoch": 0.3268373678231233,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004993527847710392,
      "loss": 5.9757,
      "mean_token_accuracy": 0.15458315759897232,
      "num_tokens": 7600719.0,
      "step": 3740
    },
    {
      "entropy": 6.148260402679443,
      "epoch": 0.32727431617582803,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004993504207911744,
      "loss": 6.0118,
      "mean_token_accuracy": 0.15330495089292526,
      "num_tokens": 7610550.0,
      "step": 3745
    },
    {
      "entropy": 6.262146615982056,
      "epoch": 0.3277112645285327,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004993480525081511,
      "loss": 6.0193,
      "mean_token_accuracy": 0.15175966769456864,
      "num_tokens": 7620472.0,
      "step": 3750
    },
    {
      "entropy": 6.238400411605835,
      "epoch": 0.3281482128812374,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004993456799220145,
      "loss": 5.963,
      "mean_token_accuracy": 0.1483220010995865,
      "num_tokens": 7630354.0,
      "step": 3755
    },
    {
      "entropy": 6.153315877914428,
      "epoch": 0.32858516123394216,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004993433030328103,
      "loss": 5.9908,
      "mean_token_accuracy": 0.1513964429497719,
      "num_tokens": 7639689.0,
      "step": 3760
    },
    {
      "entropy": 6.203240442276001,
      "epoch": 0.32902210958664685,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004993409218405839,
      "loss": 6.0312,
      "mean_token_accuracy": 0.14446706250309943,
      "num_tokens": 7649319.0,
      "step": 3765
    },
    {
      "entropy": 6.291809225082398,
      "epoch": 0.3294590579393516,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004993385363453812,
      "loss": 6.0175,
      "mean_token_accuracy": 0.1495820552110672,
      "num_tokens": 7659008.0,
      "step": 3770
    },
    {
      "entropy": 6.177696800231933,
      "epoch": 0.3298960062920563,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004993361465472477,
      "loss": 5.9629,
      "mean_token_accuracy": 0.16159392148256302,
      "num_tokens": 7669064.0,
      "step": 3775
    },
    {
      "entropy": 6.148918151855469,
      "epoch": 0.330332954644761,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004993337524462295,
      "loss": 6.0369,
      "mean_token_accuracy": 0.15342760980129241,
      "num_tokens": 7679122.0,
      "step": 3780
    },
    {
      "entropy": 6.131776571273804,
      "epoch": 0.3307699029974657,
      "grad_norm": 1.0,
      "learning_rate": 0.0004993313540423722,
      "loss": 6.0269,
      "mean_token_accuracy": 0.1543085664510727,
      "num_tokens": 7688373.0,
      "step": 3785
    },
    {
      "entropy": 6.233128356933594,
      "epoch": 0.3312068513501704,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004993289513357221,
      "loss": 5.972,
      "mean_token_accuracy": 0.14988435804843903,
      "num_tokens": 7697879.0,
      "step": 3790
    },
    {
      "entropy": 6.217293548583984,
      "epoch": 0.3316437997028751,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004993265443263251,
      "loss": 6.0166,
      "mean_token_accuracy": 0.14570185914635658,
      "num_tokens": 7708104.0,
      "step": 3795
    },
    {
      "entropy": 6.206920194625854,
      "epoch": 0.33208074805557986,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004993241330142274,
      "loss": 5.985,
      "mean_token_accuracy": 0.1601417511701584,
      "num_tokens": 7717626.0,
      "step": 3800
    },
    {
      "entropy": 6.140085458755493,
      "epoch": 0.33251769640828455,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004993217173994754,
      "loss": 6.0343,
      "mean_token_accuracy": 0.1543280601501465,
      "num_tokens": 7728221.0,
      "step": 3805
    },
    {
      "entropy": 6.244375848770142,
      "epoch": 0.33295464476098924,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004993192974821154,
      "loss": 6.0181,
      "mean_token_accuracy": 0.15254275649785995,
      "num_tokens": 7739083.0,
      "step": 3810
    },
    {
      "entropy": 6.206749439239502,
      "epoch": 0.333391593113694,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993168732621935,
      "loss": 6.0605,
      "mean_token_accuracy": 0.1473893404006958,
      "num_tokens": 7748573.0,
      "step": 3815
    },
    {
      "entropy": 6.19917893409729,
      "epoch": 0.3338285414663987,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004993144447397565,
      "loss": 6.017,
      "mean_token_accuracy": 0.15601289570331572,
      "num_tokens": 7758741.0,
      "step": 3820
    },
    {
      "entropy": 6.18432650566101,
      "epoch": 0.33426548981910337,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499312011914851,
      "loss": 6.0359,
      "mean_token_accuracy": 0.144418466091156,
      "num_tokens": 7768504.0,
      "step": 3825
    },
    {
      "entropy": 6.202671051025391,
      "epoch": 0.3347024381718081,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004993095747875234,
      "loss": 6.1692,
      "mean_token_accuracy": 0.1434830419719219,
      "num_tokens": 7779017.0,
      "step": 3830
    },
    {
      "entropy": 6.207543897628784,
      "epoch": 0.3351393865245128,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004993071333578207,
      "loss": 5.9559,
      "mean_token_accuracy": 0.15900591462850572,
      "num_tokens": 7787630.0,
      "step": 3835
    },
    {
      "entropy": 6.146871519088745,
      "epoch": 0.3355763348772175,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004993046876257898,
      "loss": 5.9722,
      "mean_token_accuracy": 0.15409253016114235,
      "num_tokens": 7797105.0,
      "step": 3840
    },
    {
      "entropy": 6.0951550006866455,
      "epoch": 0.33601328322992224,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004993022375914773,
      "loss": 5.9709,
      "mean_token_accuracy": 0.15621946305036544,
      "num_tokens": 7807070.0,
      "step": 3845
    },
    {
      "entropy": 6.2579302310943605,
      "epoch": 0.33645023158262694,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004992997832549303,
      "loss": 6.0018,
      "mean_token_accuracy": 0.1488547667860985,
      "num_tokens": 7818090.0,
      "step": 3850
    },
    {
      "entropy": 6.14893159866333,
      "epoch": 0.3368871799353316,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004992973246161961,
      "loss": 5.9375,
      "mean_token_accuracy": 0.15769325643777848,
      "num_tokens": 7827562.0,
      "step": 3855
    },
    {
      "entropy": 6.155753087997437,
      "epoch": 0.3373241282880364,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004992948616753215,
      "loss": 5.9997,
      "mean_token_accuracy": 0.14654372110962868,
      "num_tokens": 7837581.0,
      "step": 3860
    },
    {
      "entropy": 6.18585319519043,
      "epoch": 0.33776107664074106,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004992923944323541,
      "loss": 6.0643,
      "mean_token_accuracy": 0.1491297662258148,
      "num_tokens": 7848729.0,
      "step": 3865
    },
    {
      "entropy": 6.220846557617188,
      "epoch": 0.33819802499344576,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499289922887341,
      "loss": 5.9768,
      "mean_token_accuracy": 0.15103816837072373,
      "num_tokens": 7857379.0,
      "step": 3870
    },
    {
      "entropy": 6.15864930152893,
      "epoch": 0.3386349733461505,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004992874470403296,
      "loss": 5.963,
      "mean_token_accuracy": 0.15526015162467957,
      "num_tokens": 7867200.0,
      "step": 3875
    },
    {
      "entropy": 6.176467037200927,
      "epoch": 0.3390719216988552,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004992849668913674,
      "loss": 5.9603,
      "mean_token_accuracy": 0.15273297876119613,
      "num_tokens": 7876709.0,
      "step": 3880
    },
    {
      "entropy": 6.17171196937561,
      "epoch": 0.3395088700515599,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004992824824405021,
      "loss": 6.0568,
      "mean_token_accuracy": 0.1515241652727127,
      "num_tokens": 7886256.0,
      "step": 3885
    },
    {
      "entropy": 6.1952050685882565,
      "epoch": 0.33994581840426463,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004992799936877813,
      "loss": 5.9975,
      "mean_token_accuracy": 0.15033403038978577,
      "num_tokens": 7896497.0,
      "step": 3890
    },
    {
      "entropy": 6.195372438430786,
      "epoch": 0.3403827667569693,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004992775006332526,
      "loss": 5.9839,
      "mean_token_accuracy": 0.15072593092918396,
      "num_tokens": 7906608.0,
      "step": 3895
    },
    {
      "entropy": 6.200692939758301,
      "epoch": 0.340819715109674,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499275003276964,
      "loss": 6.0623,
      "mean_token_accuracy": 0.15134186148643494,
      "num_tokens": 7916359.0,
      "step": 3900
    },
    {
      "entropy": 6.193755006790161,
      "epoch": 0.34125666346237876,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992725016189632,
      "loss": 5.9485,
      "mean_token_accuracy": 0.15418373346328734,
      "num_tokens": 7925800.0,
      "step": 3905
    },
    {
      "entropy": 6.21275691986084,
      "epoch": 0.34169361181508345,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004992699956592984,
      "loss": 6.0341,
      "mean_token_accuracy": 0.1482062131166458,
      "num_tokens": 7935424.0,
      "step": 3910
    },
    {
      "entropy": 6.221978807449341,
      "epoch": 0.34213056016778814,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004992674853980175,
      "loss": 6.0119,
      "mean_token_accuracy": 0.15201026052236558,
      "num_tokens": 7945766.0,
      "step": 3915
    },
    {
      "entropy": 6.206089782714844,
      "epoch": 0.3425675085204929,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004992649708351689,
      "loss": 6.0287,
      "mean_token_accuracy": 0.14456811025738717,
      "num_tokens": 7955622.0,
      "step": 3920
    },
    {
      "entropy": 6.217914152145386,
      "epoch": 0.3430044568731976,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004992624519708004,
      "loss": 6.0923,
      "mean_token_accuracy": 0.14374652057886123,
      "num_tokens": 7965750.0,
      "step": 3925
    },
    {
      "entropy": 6.215335035324097,
      "epoch": 0.3434414052259023,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004992599288049607,
      "loss": 6.0217,
      "mean_token_accuracy": 0.1541246086359024,
      "num_tokens": 7977568.0,
      "step": 3930
    },
    {
      "entropy": 6.187791252136231,
      "epoch": 0.343878353578607,
      "grad_norm": 1.0,
      "learning_rate": 0.000499257401337698,
      "loss": 6.0751,
      "mean_token_accuracy": 0.14888253659009934,
      "num_tokens": 7988411.0,
      "step": 3935
    },
    {
      "entropy": 6.137399530410766,
      "epoch": 0.3443153019313117,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004992548695690608,
      "loss": 5.9446,
      "mean_token_accuracy": 0.15488209053874016,
      "num_tokens": 7998315.0,
      "step": 3940
    },
    {
      "entropy": 6.151676082611084,
      "epoch": 0.3447522502840164,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004992523334990977,
      "loss": 5.9891,
      "mean_token_accuracy": 0.15101110190153122,
      "num_tokens": 8007133.0,
      "step": 3945
    },
    {
      "entropy": 6.160750484466552,
      "epoch": 0.34518919863672115,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004992497931278574,
      "loss": 5.9285,
      "mean_token_accuracy": 0.15777034908533097,
      "num_tokens": 8016882.0,
      "step": 3950
    },
    {
      "entropy": 6.1163647174835205,
      "epoch": 0.34562614698942584,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004992472484553886,
      "loss": 5.9465,
      "mean_token_accuracy": 0.1563786208629608,
      "num_tokens": 8026893.0,
      "step": 3955
    },
    {
      "entropy": 6.1823328018188475,
      "epoch": 0.34606309534213053,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00049924469948174,
      "loss": 5.975,
      "mean_token_accuracy": 0.15222077816724777,
      "num_tokens": 8036240.0,
      "step": 3960
    },
    {
      "entropy": 6.208459329605103,
      "epoch": 0.3465000436948353,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004992421462069606,
      "loss": 5.9519,
      "mean_token_accuracy": 0.14994053691625595,
      "num_tokens": 8046086.0,
      "step": 3965
    },
    {
      "entropy": 6.072393798828125,
      "epoch": 0.34693699204753997,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004992395886310993,
      "loss": 5.9636,
      "mean_token_accuracy": 0.15662624537944794,
      "num_tokens": 8056294.0,
      "step": 3970
    },
    {
      "entropy": 6.21268858909607,
      "epoch": 0.3473739404002447,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004992370267542052,
      "loss": 5.9988,
      "mean_token_accuracy": 0.14892108887434005,
      "num_tokens": 8067329.0,
      "step": 3975
    },
    {
      "entropy": 6.167745923995971,
      "epoch": 0.3478108887529494,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004992344605763275,
      "loss": 5.9626,
      "mean_token_accuracy": 0.15514642894268035,
      "num_tokens": 8077106.0,
      "step": 3980
    },
    {
      "entropy": 6.145469379425049,
      "epoch": 0.3482478371056541,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004992318900975153,
      "loss": 5.8579,
      "mean_token_accuracy": 0.1612818643450737,
      "num_tokens": 8087766.0,
      "step": 3985
    },
    {
      "entropy": 6.053077220916748,
      "epoch": 0.34868478545835885,
      "grad_norm": 1.0,
      "learning_rate": 0.0004992293153178179,
      "loss": 5.9736,
      "mean_token_accuracy": 0.15829520300030708,
      "num_tokens": 8098649.0,
      "step": 3990
    },
    {
      "entropy": 6.190802812576294,
      "epoch": 0.34912173381106354,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004992267362372848,
      "loss": 5.992,
      "mean_token_accuracy": 0.15101442635059356,
      "num_tokens": 8109237.0,
      "step": 3995
    },
    {
      "entropy": 6.143446636199951,
      "epoch": 0.34955868216376823,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004992241528559653,
      "loss": 5.9873,
      "mean_token_accuracy": 0.15395535230636598,
      "num_tokens": 8119731.0,
      "step": 4000
    },
    {
      "entropy": 6.21689805984497,
      "epoch": 0.349995630516473,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004992215651739092,
      "loss": 5.9973,
      "mean_token_accuracy": 0.1451298750936985,
      "num_tokens": 8129448.0,
      "step": 4005
    },
    {
      "entropy": 6.122711992263794,
      "epoch": 0.35043257886917767,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000499218973191166,
      "loss": 5.8952,
      "mean_token_accuracy": 0.1595328703522682,
      "num_tokens": 8138345.0,
      "step": 4010
    },
    {
      "entropy": 6.184520673751831,
      "epoch": 0.35086952722188236,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004992163769077853,
      "loss": 5.9585,
      "mean_token_accuracy": 0.15615929663181305,
      "num_tokens": 8148380.0,
      "step": 4015
    },
    {
      "entropy": 6.1191223621368405,
      "epoch": 0.3513064755745871,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499213776323817,
      "loss": 5.9737,
      "mean_token_accuracy": 0.15443783402442932,
      "num_tokens": 8159359.0,
      "step": 4020
    },
    {
      "entropy": 6.195638132095337,
      "epoch": 0.3517434239272918,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499211171439311,
      "loss": 6.0588,
      "mean_token_accuracy": 0.14493529200553895,
      "num_tokens": 8168586.0,
      "step": 4025
    },
    {
      "entropy": 6.1902841091156,
      "epoch": 0.3521803722799965,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004992085622543172,
      "loss": 5.9584,
      "mean_token_accuracy": 0.1562234103679657,
      "num_tokens": 8178836.0,
      "step": 4030
    },
    {
      "entropy": 6.117028665542603,
      "epoch": 0.35261732063270124,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004992059487688859,
      "loss": 5.9694,
      "mean_token_accuracy": 0.15520091205835343,
      "num_tokens": 8188933.0,
      "step": 4035
    },
    {
      "entropy": 6.099272203445435,
      "epoch": 0.35305426898540593,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992033309830668,
      "loss": 5.9606,
      "mean_token_accuracy": 0.16192768961191178,
      "num_tokens": 8199366.0,
      "step": 4040
    },
    {
      "entropy": 6.247333765029907,
      "epoch": 0.3534912173381106,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004992007088969105,
      "loss": 5.9244,
      "mean_token_accuracy": 0.15734761506319045,
      "num_tokens": 8209806.0,
      "step": 4045
    },
    {
      "entropy": 6.188394832611084,
      "epoch": 0.35392816569081537,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499198082510467,
      "loss": 6.0445,
      "mean_token_accuracy": 0.14461320415139198,
      "num_tokens": 8219367.0,
      "step": 4050
    },
    {
      "entropy": 6.108756065368652,
      "epoch": 0.35436511404352006,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004991954518237868,
      "loss": 5.9144,
      "mean_token_accuracy": 0.15329063534736634,
      "num_tokens": 8228918.0,
      "step": 4055
    },
    {
      "entropy": 6.12732892036438,
      "epoch": 0.35480206239622475,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004991928168369205,
      "loss": 6.0076,
      "mean_token_accuracy": 0.15693602114915847,
      "num_tokens": 8239380.0,
      "step": 4060
    },
    {
      "entropy": 6.216499376296997,
      "epoch": 0.3552390107489295,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004991901775499183,
      "loss": 6.0538,
      "mean_token_accuracy": 0.15166293531656266,
      "num_tokens": 8249131.0,
      "step": 4065
    },
    {
      "entropy": 6.237508010864258,
      "epoch": 0.3556759591016342,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004991875339628311,
      "loss": 5.956,
      "mean_token_accuracy": 0.1515304461121559,
      "num_tokens": 8258550.0,
      "step": 4070
    },
    {
      "entropy": 6.120978879928589,
      "epoch": 0.3561129074543389,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004991848860757095,
      "loss": 5.9777,
      "mean_token_accuracy": 0.15791711956262589,
      "num_tokens": 8269089.0,
      "step": 4075
    },
    {
      "entropy": 6.12967529296875,
      "epoch": 0.3565498558070436,
      "grad_norm": 0.875,
      "learning_rate": 0.0004991822338886044,
      "loss": 5.9603,
      "mean_token_accuracy": 0.1507464587688446,
      "num_tokens": 8279735.0,
      "step": 4080
    },
    {
      "entropy": 6.155176258087158,
      "epoch": 0.3569868041597483,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004991795774015665,
      "loss": 5.929,
      "mean_token_accuracy": 0.1593891128897667,
      "num_tokens": 8289693.0,
      "step": 4085
    },
    {
      "entropy": 6.16059627532959,
      "epoch": 0.357423752512453,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004991769166146469,
      "loss": 6.0007,
      "mean_token_accuracy": 0.15330177992582322,
      "num_tokens": 8300909.0,
      "step": 4090
    },
    {
      "entropy": 6.152325582504273,
      "epoch": 0.35786070086515775,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004991742515278965,
      "loss": 5.9659,
      "mean_token_accuracy": 0.1568295031785965,
      "num_tokens": 8311479.0,
      "step": 4095
    },
    {
      "entropy": 6.193072032928467,
      "epoch": 0.35829764921786245,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004991715821413665,
      "loss": 5.9698,
      "mean_token_accuracy": 0.15319570004940034,
      "num_tokens": 8321267.0,
      "step": 4100
    },
    {
      "entropy": 6.1143323421478275,
      "epoch": 0.35873459757056714,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004991689084551082,
      "loss": 5.9265,
      "mean_token_accuracy": 0.1565576121211052,
      "num_tokens": 8331445.0,
      "step": 4105
    },
    {
      "entropy": 6.134524297714234,
      "epoch": 0.3591715459232719,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004991662304691727,
      "loss": 6.0089,
      "mean_token_accuracy": 0.150065641105175,
      "num_tokens": 8340519.0,
      "step": 4110
    },
    {
      "entropy": 6.160966110229492,
      "epoch": 0.3596084942759766,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004991635481836114,
      "loss": 5.9819,
      "mean_token_accuracy": 0.14954216852784158,
      "num_tokens": 8350467.0,
      "step": 4115
    },
    {
      "entropy": 6.1642660140991214,
      "epoch": 0.36004544262868127,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004991608615984757,
      "loss": 5.9362,
      "mean_token_accuracy": 0.14789557456970215,
      "num_tokens": 8360067.0,
      "step": 4120
    },
    {
      "entropy": 6.176195859909058,
      "epoch": 0.360482390981386,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004991581707138175,
      "loss": 6.0121,
      "mean_token_accuracy": 0.15181783437728882,
      "num_tokens": 8370509.0,
      "step": 4125
    },
    {
      "entropy": 6.1493510723114015,
      "epoch": 0.3609193393340907,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004991554755296878,
      "loss": 5.9477,
      "mean_token_accuracy": 0.15763404667377473,
      "num_tokens": 8381489.0,
      "step": 4130
    },
    {
      "entropy": 6.074535083770752,
      "epoch": 0.3613562876867954,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004991527760461389,
      "loss": 5.9799,
      "mean_token_accuracy": 0.15404071509838105,
      "num_tokens": 8392287.0,
      "step": 4135
    },
    {
      "entropy": 6.141118144989013,
      "epoch": 0.36179323603950014,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004991500722632223,
      "loss": 5.9545,
      "mean_token_accuracy": 0.15313552618026732,
      "num_tokens": 8402433.0,
      "step": 4140
    },
    {
      "entropy": 6.152544593811035,
      "epoch": 0.36223018439220483,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004991473641809896,
      "loss": 5.98,
      "mean_token_accuracy": 0.15163057744503022,
      "num_tokens": 8412944.0,
      "step": 4145
    },
    {
      "entropy": 6.1541948318481445,
      "epoch": 0.3626671327449095,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004991446517994932,
      "loss": 5.8907,
      "mean_token_accuracy": 0.15631764084100724,
      "num_tokens": 8422511.0,
      "step": 4150
    },
    {
      "entropy": 6.119439697265625,
      "epoch": 0.3631040810976143,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004991419351187849,
      "loss": 5.9412,
      "mean_token_accuracy": 0.16003672629594803,
      "num_tokens": 8433094.0,
      "step": 4155
    },
    {
      "entropy": 6.054240942001343,
      "epoch": 0.36354102945031896,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004991392141389169,
      "loss": 6.0368,
      "mean_token_accuracy": 0.15489192605018615,
      "num_tokens": 8444430.0,
      "step": 4160
    },
    {
      "entropy": 6.220261430740356,
      "epoch": 0.36397797780302366,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004991364888599412,
      "loss": 6.0836,
      "mean_token_accuracy": 0.1493389882147312,
      "num_tokens": 8455171.0,
      "step": 4165
    },
    {
      "entropy": 6.1484057903289795,
      "epoch": 0.3644149261557284,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991337592819103,
      "loss": 5.8743,
      "mean_token_accuracy": 0.16199919134378432,
      "num_tokens": 8464724.0,
      "step": 4170
    },
    {
      "entropy": 6.1508533477783205,
      "epoch": 0.3648518745084331,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004991310254048763,
      "loss": 5.9875,
      "mean_token_accuracy": 0.15001552179455757,
      "num_tokens": 8474732.0,
      "step": 4175
    },
    {
      "entropy": 6.128949928283691,
      "epoch": 0.36528882286113784,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004991282872288919,
      "loss": 5.9384,
      "mean_token_accuracy": 0.15761642754077912,
      "num_tokens": 8484731.0,
      "step": 4180
    },
    {
      "entropy": 6.077857780456543,
      "epoch": 0.36572577121384253,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004991255447540095,
      "loss": 5.9345,
      "mean_token_accuracy": 0.15963230580091475,
      "num_tokens": 8494509.0,
      "step": 4185
    },
    {
      "entropy": 6.178862714767456,
      "epoch": 0.3661627195665472,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991227979802817,
      "loss": 6.0058,
      "mean_token_accuracy": 0.15825022608041764,
      "num_tokens": 8504260.0,
      "step": 4190
    },
    {
      "entropy": 6.16565055847168,
      "epoch": 0.36659966791925197,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004991200469077612,
      "loss": 6.0338,
      "mean_token_accuracy": 0.14986559748649597,
      "num_tokens": 8514222.0,
      "step": 4195
    },
    {
      "entropy": 6.189094352722168,
      "epoch": 0.36703661627195666,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004991172915365008,
      "loss": 5.9729,
      "mean_token_accuracy": 0.1543964222073555,
      "num_tokens": 8523979.0,
      "step": 4200
    },
    {
      "entropy": 6.12542724609375,
      "epoch": 0.36747356462466135,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004991145318665532,
      "loss": 5.9273,
      "mean_token_accuracy": 0.1569794923067093,
      "num_tokens": 8533516.0,
      "step": 4205
    },
    {
      "entropy": 6.112308406829834,
      "epoch": 0.3679105129773661,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004991117678979716,
      "loss": 6.0151,
      "mean_token_accuracy": 0.14600328952074051,
      "num_tokens": 8544444.0,
      "step": 4210
    },
    {
      "entropy": 6.181897735595703,
      "epoch": 0.3683474613300708,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004991089996308089,
      "loss": 5.9489,
      "mean_token_accuracy": 0.15276936441659927,
      "num_tokens": 8555571.0,
      "step": 4215
    },
    {
      "entropy": 6.1980299949646,
      "epoch": 0.3687844096827755,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004991062270651179,
      "loss": 6.0112,
      "mean_token_accuracy": 0.14825379848480225,
      "num_tokens": 8566486.0,
      "step": 4220
    },
    {
      "entropy": 6.206994295120239,
      "epoch": 0.36922135803548023,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991034502009522,
      "loss": 5.9307,
      "mean_token_accuracy": 0.1546737790107727,
      "num_tokens": 8575324.0,
      "step": 4225
    },
    {
      "entropy": 6.098765707015991,
      "epoch": 0.3696583063881849,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004991006690383649,
      "loss": 5.9191,
      "mean_token_accuracy": 0.15529632717370986,
      "num_tokens": 8586015.0,
      "step": 4230
    },
    {
      "entropy": 6.092961359024048,
      "epoch": 0.3700952547408896,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004990978835774093,
      "loss": 5.9106,
      "mean_token_accuracy": 0.15810996443033218,
      "num_tokens": 8594748.0,
      "step": 4235
    },
    {
      "entropy": 6.091306877136231,
      "epoch": 0.37053220309359436,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004990950938181389,
      "loss": 5.9714,
      "mean_token_accuracy": 0.14972231686115264,
      "num_tokens": 8605805.0,
      "step": 4240
    },
    {
      "entropy": 6.125380420684815,
      "epoch": 0.37096915144629905,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990922997606071,
      "loss": 5.9376,
      "mean_token_accuracy": 0.15340033024549485,
      "num_tokens": 8616591.0,
      "step": 4245
    },
    {
      "entropy": 6.15188455581665,
      "epoch": 0.37140609979900374,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004990895014048677,
      "loss": 5.9621,
      "mean_token_accuracy": 0.15207428634166717,
      "num_tokens": 8627063.0,
      "step": 4250
    },
    {
      "entropy": 6.2279682636260985,
      "epoch": 0.3718430481517085,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990866987509741,
      "loss": 6.0331,
      "mean_token_accuracy": 0.14305689334869384,
      "num_tokens": 8637041.0,
      "step": 4255
    },
    {
      "entropy": 6.183949565887451,
      "epoch": 0.3722799965044132,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004990838917989802,
      "loss": 5.9519,
      "mean_token_accuracy": 0.16125427335500717,
      "num_tokens": 8647485.0,
      "step": 4260
    },
    {
      "entropy": 6.138697481155395,
      "epoch": 0.37271694485711787,
      "grad_norm": 0.984375,
      "learning_rate": 0.00049908108054894,
      "loss": 5.9254,
      "mean_token_accuracy": 0.15362533926963806,
      "num_tokens": 8657185.0,
      "step": 4265
    },
    {
      "entropy": 6.145811891555786,
      "epoch": 0.3731538932098226,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004990782650009071,
      "loss": 5.9201,
      "mean_token_accuracy": 0.15877077728509903,
      "num_tokens": 8667639.0,
      "step": 4270
    },
    {
      "entropy": 6.135325765609741,
      "epoch": 0.3735908415625273,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004990754451549358,
      "loss": 5.9558,
      "mean_token_accuracy": 0.1507587969303131,
      "num_tokens": 8678003.0,
      "step": 4275
    },
    {
      "entropy": 6.184204053878784,
      "epoch": 0.374027789915232,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004990726210110798,
      "loss": 6.0093,
      "mean_token_accuracy": 0.15558700412511825,
      "num_tokens": 8688837.0,
      "step": 4280
    },
    {
      "entropy": 6.104807376861572,
      "epoch": 0.37446473826793675,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004990697925693938,
      "loss": 5.7952,
      "mean_token_accuracy": 0.1686731457710266,
      "num_tokens": 8698395.0,
      "step": 4285
    },
    {
      "entropy": 6.087707424163819,
      "epoch": 0.37490168662064144,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004990669598299316,
      "loss": 5.9355,
      "mean_token_accuracy": 0.1568262904882431,
      "num_tokens": 8708031.0,
      "step": 4290
    },
    {
      "entropy": 6.092851352691651,
      "epoch": 0.37533863497334613,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004990641227927478,
      "loss": 5.8974,
      "mean_token_accuracy": 0.1581382170319557,
      "num_tokens": 8717825.0,
      "step": 4295
    },
    {
      "entropy": 6.223221063613892,
      "epoch": 0.3757755833260509,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004990612814578967,
      "loss": 5.9887,
      "mean_token_accuracy": 0.1510862223803997,
      "num_tokens": 8728873.0,
      "step": 4300
    },
    {
      "entropy": 6.136455965042114,
      "epoch": 0.37621253167875557,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004990584358254327,
      "loss": 5.9253,
      "mean_token_accuracy": 0.1545376054942608,
      "num_tokens": 8737917.0,
      "step": 4305
    },
    {
      "entropy": 6.057396411895752,
      "epoch": 0.37664948003146026,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004990555858954107,
      "loss": 5.8821,
      "mean_token_accuracy": 0.15671781599521636,
      "num_tokens": 8747466.0,
      "step": 4310
    },
    {
      "entropy": 6.108720541000366,
      "epoch": 0.377086428384165,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000499052731667885,
      "loss": 5.9101,
      "mean_token_accuracy": 0.15692686289548874,
      "num_tokens": 8757037.0,
      "step": 4315
    },
    {
      "entropy": 6.131253337860107,
      "epoch": 0.3775233767368697,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004990498731429105,
      "loss": 5.9464,
      "mean_token_accuracy": 0.1584157355129719,
      "num_tokens": 8766780.0,
      "step": 4320
    },
    {
      "entropy": 6.105612421035767,
      "epoch": 0.3779603250895744,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004990470103205421,
      "loss": 5.9112,
      "mean_token_accuracy": 0.15456206500530242,
      "num_tokens": 8777063.0,
      "step": 4325
    },
    {
      "entropy": 6.0960869789123535,
      "epoch": 0.37839727344227914,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004990441432008347,
      "loss": 5.9589,
      "mean_token_accuracy": 0.15549104362726213,
      "num_tokens": 8786479.0,
      "step": 4330
    },
    {
      "entropy": 6.199940252304077,
      "epoch": 0.3788342217949838,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004990412717838431,
      "loss": 5.9352,
      "mean_token_accuracy": 0.1555895447731018,
      "num_tokens": 8795663.0,
      "step": 4335
    },
    {
      "entropy": 6.084987163543701,
      "epoch": 0.3792711701476885,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004990383960696226,
      "loss": 5.8887,
      "mean_token_accuracy": 0.15801238715648652,
      "num_tokens": 8805854.0,
      "step": 4340
    },
    {
      "entropy": 6.121315145492554,
      "epoch": 0.37970811850039327,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004990355160582283,
      "loss": 5.8664,
      "mean_token_accuracy": 0.15441805124282837,
      "num_tokens": 8816068.0,
      "step": 4345
    },
    {
      "entropy": 6.043383026123047,
      "epoch": 0.38014506685309796,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004990326317497153,
      "loss": 5.875,
      "mean_token_accuracy": 0.16005214005708696,
      "num_tokens": 8826301.0,
      "step": 4350
    },
    {
      "entropy": 6.138577938079834,
      "epoch": 0.38058201520580265,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004990297431441391,
      "loss": 5.9253,
      "mean_token_accuracy": 0.16221683025360106,
      "num_tokens": 8837482.0,
      "step": 4355
    },
    {
      "entropy": 6.142789077758789,
      "epoch": 0.3810189635585074,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000499026850241555,
      "loss": 5.9515,
      "mean_token_accuracy": 0.1573454514145851,
      "num_tokens": 8847041.0,
      "step": 4360
    },
    {
      "entropy": 6.1118231296539305,
      "epoch": 0.3814559119112121,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004990239530420185,
      "loss": 5.9318,
      "mean_token_accuracy": 0.15020157843828202,
      "num_tokens": 8856963.0,
      "step": 4365
    },
    {
      "entropy": 6.132876014709472,
      "epoch": 0.3818928602639168,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004990210515455853,
      "loss": 5.9068,
      "mean_token_accuracy": 0.1551741138100624,
      "num_tokens": 8867318.0,
      "step": 4370
    },
    {
      "entropy": 6.0603608131408695,
      "epoch": 0.3823298086166215,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004990181457523109,
      "loss": 5.9282,
      "mean_token_accuracy": 0.15901948213577272,
      "num_tokens": 8877634.0,
      "step": 4375
    },
    {
      "entropy": 6.092250156402588,
      "epoch": 0.3827667569693262,
      "grad_norm": 0.9375,
      "learning_rate": 0.000499015235662251,
      "loss": 5.9642,
      "mean_token_accuracy": 0.15656676292419433,
      "num_tokens": 8887708.0,
      "step": 4380
    },
    {
      "entropy": 6.118471765518189,
      "epoch": 0.38320370532203096,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004990123212754616,
      "loss": 5.8851,
      "mean_token_accuracy": 0.15717434734106064,
      "num_tokens": 8897510.0,
      "step": 4385
    },
    {
      "entropy": 6.077293872833252,
      "epoch": 0.38364065367473565,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004990094025919985,
      "loss": 5.9302,
      "mean_token_accuracy": 0.15012910664081575,
      "num_tokens": 8908282.0,
      "step": 4390
    },
    {
      "entropy": 6.206435680389404,
      "epoch": 0.38407760202744035,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004990064796119175,
      "loss": 5.9597,
      "mean_token_accuracy": 0.14675080999732018,
      "num_tokens": 8918332.0,
      "step": 4395
    },
    {
      "entropy": 6.1221198558807375,
      "epoch": 0.3845145503801451,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004990035523352751,
      "loss": 5.9158,
      "mean_token_accuracy": 0.1555359274148941,
      "num_tokens": 8929138.0,
      "step": 4400
    },
    {
      "entropy": 6.064434623718261,
      "epoch": 0.3849514987328498,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000499000620762127,
      "loss": 5.9581,
      "mean_token_accuracy": 0.15178584903478623,
      "num_tokens": 8940871.0,
      "step": 4405
    },
    {
      "entropy": 6.202463293075562,
      "epoch": 0.3853884470855545,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004989976848925297,
      "loss": 5.9143,
      "mean_token_accuracy": 0.1592526078224182,
      "num_tokens": 8951199.0,
      "step": 4410
    },
    {
      "entropy": 6.120063018798828,
      "epoch": 0.3858253954382592,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004989947447265394,
      "loss": 6.0198,
      "mean_token_accuracy": 0.1502373956143856,
      "num_tokens": 8961482.0,
      "step": 4415
    },
    {
      "entropy": 6.195671129226684,
      "epoch": 0.3862623437909639,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004989918002642126,
      "loss": 5.9322,
      "mean_token_accuracy": 0.15296921283006668,
      "num_tokens": 8971884.0,
      "step": 4420
    },
    {
      "entropy": 6.093695402145386,
      "epoch": 0.3866992921436686,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004989888515056057,
      "loss": 5.8872,
      "mean_token_accuracy": 0.1532720759510994,
      "num_tokens": 8982086.0,
      "step": 4425
    },
    {
      "entropy": 6.05369725227356,
      "epoch": 0.38713624049637335,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004989858984507752,
      "loss": 5.9571,
      "mean_token_accuracy": 0.15088309347629547,
      "num_tokens": 8991555.0,
      "step": 4430
    },
    {
      "entropy": 6.169110059738159,
      "epoch": 0.38757318884907804,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004989829410997778,
      "loss": 5.931,
      "mean_token_accuracy": 0.15364627912640572,
      "num_tokens": 9002459.0,
      "step": 4435
    },
    {
      "entropy": 6.101735258102417,
      "epoch": 0.38801013720178273,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004989799794526702,
      "loss": 5.9861,
      "mean_token_accuracy": 0.1510950393974781,
      "num_tokens": 9012514.0,
      "step": 4440
    },
    {
      "entropy": 6.0932494640350345,
      "epoch": 0.3884470855544875,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004989770135095093,
      "loss": 5.8925,
      "mean_token_accuracy": 0.15784662067890168,
      "num_tokens": 9022636.0,
      "step": 4445
    },
    {
      "entropy": 6.1229030132293705,
      "epoch": 0.3888840339071922,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004989740432703519,
      "loss": 5.9111,
      "mean_token_accuracy": 0.15686276108026503,
      "num_tokens": 9032868.0,
      "step": 4450
    },
    {
      "entropy": 6.088069057464599,
      "epoch": 0.38932098225989686,
      "grad_norm": 0.9375,
      "learning_rate": 0.000498971068735255,
      "loss": 5.9011,
      "mean_token_accuracy": 0.15465139150619506,
      "num_tokens": 9043057.0,
      "step": 4455
    },
    {
      "entropy": 6.06488528251648,
      "epoch": 0.3897579306126016,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004989680899042756,
      "loss": 5.9202,
      "mean_token_accuracy": 0.15525414496660234,
      "num_tokens": 9053179.0,
      "step": 4460
    },
    {
      "entropy": 6.085233926773071,
      "epoch": 0.3901948789653063,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004989651067774708,
      "loss": 5.9286,
      "mean_token_accuracy": 0.15788234919309616,
      "num_tokens": 9063112.0,
      "step": 4465
    },
    {
      "entropy": 6.141564130783081,
      "epoch": 0.390631827318011,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004989621193548981,
      "loss": 5.9428,
      "mean_token_accuracy": 0.1567154720425606,
      "num_tokens": 9073512.0,
      "step": 4470
    },
    {
      "entropy": 6.141236734390259,
      "epoch": 0.39106877567071574,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004989591276366145,
      "loss": 5.9611,
      "mean_token_accuracy": 0.1534711793065071,
      "num_tokens": 9083369.0,
      "step": 4475
    },
    {
      "entropy": 6.076042938232422,
      "epoch": 0.39150572402342043,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004989561316226774,
      "loss": 5.9336,
      "mean_token_accuracy": 0.153956837952137,
      "num_tokens": 9092814.0,
      "step": 4480
    },
    {
      "entropy": 6.006694555282593,
      "epoch": 0.3919426723761251,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004989531313131444,
      "loss": 5.848,
      "mean_token_accuracy": 0.16450390815734864,
      "num_tokens": 9102816.0,
      "step": 4485
    },
    {
      "entropy": 6.126395225524902,
      "epoch": 0.39237962072882987,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000498950126708073,
      "loss": 5.9741,
      "mean_token_accuracy": 0.14918582811951636,
      "num_tokens": 9114470.0,
      "step": 4490
    },
    {
      "entropy": 6.1541100025177,
      "epoch": 0.39281656908153456,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004989471178075208,
      "loss": 5.9613,
      "mean_token_accuracy": 0.15156185925006865,
      "num_tokens": 9125604.0,
      "step": 4495
    },
    {
      "entropy": 6.114582586288452,
      "epoch": 0.39325351743423925,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004989441046115456,
      "loss": 5.9074,
      "mean_token_accuracy": 0.16039225608110427,
      "num_tokens": 9135088.0,
      "step": 4500
    },
    {
      "entropy": 6.1255189895629885,
      "epoch": 0.393690465786944,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004989410871202051,
      "loss": 5.9564,
      "mean_token_accuracy": 0.15606121867895126,
      "num_tokens": 9145302.0,
      "step": 4505
    },
    {
      "entropy": 6.039809894561768,
      "epoch": 0.3941274141396487,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004989380653335571,
      "loss": 5.8806,
      "mean_token_accuracy": 0.15818860679864882,
      "num_tokens": 9155104.0,
      "step": 4510
    },
    {
      "entropy": 6.101123809814453,
      "epoch": 0.3945643624923534,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004989350392516598,
      "loss": 5.8256,
      "mean_token_accuracy": 0.1587645962834358,
      "num_tokens": 9165055.0,
      "step": 4515
    },
    {
      "entropy": 6.094510173797607,
      "epoch": 0.39500131084505813,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004989320088745711,
      "loss": 5.8437,
      "mean_token_accuracy": 0.16077055037021637,
      "num_tokens": 9175305.0,
      "step": 4520
    },
    {
      "entropy": 6.060928440093994,
      "epoch": 0.3954382591977628,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000498928974202349,
      "loss": 5.9472,
      "mean_token_accuracy": 0.1519213378429413,
      "num_tokens": 9185153.0,
      "step": 4525
    },
    {
      "entropy": 6.1433248043060305,
      "epoch": 0.3958752075504675,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000498925935235052,
      "loss": 5.8951,
      "mean_token_accuracy": 0.1532336466014385,
      "num_tokens": 9195385.0,
      "step": 4530
    },
    {
      "entropy": 6.031657457351685,
      "epoch": 0.39631215590317226,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004989228919727382,
      "loss": 5.8505,
      "mean_token_accuracy": 0.15958960354328156,
      "num_tokens": 9205009.0,
      "step": 4535
    },
    {
      "entropy": 6.065906381607055,
      "epoch": 0.39674910425587695,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000498919844415466,
      "loss": 5.9115,
      "mean_token_accuracy": 0.15849722176790237,
      "num_tokens": 9214684.0,
      "step": 4540
    },
    {
      "entropy": 6.069202661514282,
      "epoch": 0.39718605260858164,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004989167925632937,
      "loss": 5.847,
      "mean_token_accuracy": 0.15935237258672713,
      "num_tokens": 9225721.0,
      "step": 4545
    },
    {
      "entropy": 6.085740470886231,
      "epoch": 0.3976230009612864,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004989137364162801,
      "loss": 5.9324,
      "mean_token_accuracy": 0.15576011836528778,
      "num_tokens": 9236167.0,
      "step": 4550
    },
    {
      "entropy": 6.031605005264282,
      "epoch": 0.3980599493139911,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004989106759744837,
      "loss": 5.8814,
      "mean_token_accuracy": 0.1644463434815407,
      "num_tokens": 9245716.0,
      "step": 4555
    },
    {
      "entropy": 6.093547725677491,
      "epoch": 0.39849689766669577,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004989076112379633,
      "loss": 5.9667,
      "mean_token_accuracy": 0.1497325897216797,
      "num_tokens": 9256897.0,
      "step": 4560
    },
    {
      "entropy": 6.188295364379883,
      "epoch": 0.3989338460194005,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004989045422067775,
      "loss": 5.9659,
      "mean_token_accuracy": 0.14358162581920625,
      "num_tokens": 9266080.0,
      "step": 4565
    },
    {
      "entropy": 6.057019186019898,
      "epoch": 0.3993707943721052,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004989014688809853,
      "loss": 5.8883,
      "mean_token_accuracy": 0.1564632087945938,
      "num_tokens": 9276815.0,
      "step": 4570
    },
    {
      "entropy": 6.082181167602539,
      "epoch": 0.3998077427248099,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004988983912606454,
      "loss": 5.9499,
      "mean_token_accuracy": 0.15490922778844834,
      "num_tokens": 9287339.0,
      "step": 4575
    },
    {
      "entropy": 6.0535625457763675,
      "epoch": 0.40024469107751465,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004988953093458173,
      "loss": 5.8278,
      "mean_token_accuracy": 0.1621466264128685,
      "num_tokens": 9298002.0,
      "step": 4580
    },
    {
      "entropy": 6.098121309280396,
      "epoch": 0.40068163943021934,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004988922231365597,
      "loss": 5.9566,
      "mean_token_accuracy": 0.15271753072738647,
      "num_tokens": 9309008.0,
      "step": 4585
    },
    {
      "entropy": 6.107387590408325,
      "epoch": 0.4011185877829241,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000498889132632932,
      "loss": 5.9616,
      "mean_token_accuracy": 0.14986975491046906,
      "num_tokens": 9319391.0,
      "step": 4590
    },
    {
      "entropy": 6.009187507629394,
      "epoch": 0.4015555361356288,
      "grad_norm": 0.875,
      "learning_rate": 0.0004988860378349934,
      "loss": 5.8026,
      "mean_token_accuracy": 0.16928617507219315,
      "num_tokens": 9329808.0,
      "step": 4595
    },
    {
      "entropy": 6.0693920135498045,
      "epoch": 0.40199248448833347,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004988829387428033,
      "loss": 5.872,
      "mean_token_accuracy": 0.15713104009628295,
      "num_tokens": 9339580.0,
      "step": 4600
    },
    {
      "entropy": 6.1516563415527346,
      "epoch": 0.4024294328410382,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004988798353564211,
      "loss": 5.8827,
      "mean_token_accuracy": 0.16016404181718827,
      "num_tokens": 9350008.0,
      "step": 4605
    },
    {
      "entropy": 6.102530384063721,
      "epoch": 0.4028663811937429,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004988767276759064,
      "loss": 5.8891,
      "mean_token_accuracy": 0.15995637625455855,
      "num_tokens": 9359026.0,
      "step": 4610
    },
    {
      "entropy": 6.08635778427124,
      "epoch": 0.4033033295464476,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004988736157013186,
      "loss": 5.8541,
      "mean_token_accuracy": 0.15809393376111985,
      "num_tokens": 9369084.0,
      "step": 4615
    },
    {
      "entropy": 6.056956958770752,
      "epoch": 0.40374027789915234,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004988704994327178,
      "loss": 5.8385,
      "mean_token_accuracy": 0.16860908269882202,
      "num_tokens": 9378273.0,
      "step": 4620
    },
    {
      "entropy": 6.043905210494995,
      "epoch": 0.40417722625185704,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004988673788701633,
      "loss": 5.8712,
      "mean_token_accuracy": 0.1564175680279732,
      "num_tokens": 9388434.0,
      "step": 4625
    },
    {
      "entropy": 6.096612930297852,
      "epoch": 0.4046141746045617,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004988642540137152,
      "loss": 5.9615,
      "mean_token_accuracy": 0.15139524936676024,
      "num_tokens": 9398484.0,
      "step": 4630
    },
    {
      "entropy": 6.041618919372558,
      "epoch": 0.4050511229572665,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004988611248634335,
      "loss": 5.8087,
      "mean_token_accuracy": 0.16204500794410706,
      "num_tokens": 9409544.0,
      "step": 4635
    },
    {
      "entropy": 6.105325174331665,
      "epoch": 0.40548807130997117,
      "grad_norm": 0.9375,
      "learning_rate": 0.000498857991419378,
      "loss": 5.8812,
      "mean_token_accuracy": 0.1569657102227211,
      "num_tokens": 9419391.0,
      "step": 4640
    },
    {
      "entropy": 6.105036211013794,
      "epoch": 0.40592501966267586,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004988548536816091,
      "loss": 5.8483,
      "mean_token_accuracy": 0.16299069076776504,
      "num_tokens": 9429342.0,
      "step": 4645
    },
    {
      "entropy": 6.0683198928833,
      "epoch": 0.4063619680153806,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004988517116501867,
      "loss": 5.9651,
      "mean_token_accuracy": 0.14840319603681565,
      "num_tokens": 9440643.0,
      "step": 4650
    },
    {
      "entropy": 6.162385559082031,
      "epoch": 0.4067989163680853,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004988485653251711,
      "loss": 5.9167,
      "mean_token_accuracy": 0.15395443737506867,
      "num_tokens": 9450674.0,
      "step": 4655
    },
    {
      "entropy": 6.10659613609314,
      "epoch": 0.40723586472079,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004988454147066228,
      "loss": 5.8772,
      "mean_token_accuracy": 0.1581096902489662,
      "num_tokens": 9460075.0,
      "step": 4660
    },
    {
      "entropy": 6.058187532424927,
      "epoch": 0.40767281307349473,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004988422597946022,
      "loss": 5.8647,
      "mean_token_accuracy": 0.16078261584043502,
      "num_tokens": 9469327.0,
      "step": 4665
    },
    {
      "entropy": 6.107449769973755,
      "epoch": 0.4081097614261994,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004988391005891698,
      "loss": 5.9111,
      "mean_token_accuracy": 0.15315258949995042,
      "num_tokens": 9478304.0,
      "step": 4670
    },
    {
      "entropy": 6.054425191879273,
      "epoch": 0.4085467097789041,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000498835937090386,
      "loss": 5.8413,
      "mean_token_accuracy": 0.15889210253953934,
      "num_tokens": 9487453.0,
      "step": 4675
    },
    {
      "entropy": 6.043490552902222,
      "epoch": 0.40898365813160886,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004988327692983116,
      "loss": 5.8039,
      "mean_token_accuracy": 0.16317633986473085,
      "num_tokens": 9496623.0,
      "step": 4680
    },
    {
      "entropy": 6.053945922851563,
      "epoch": 0.40942060648431355,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004988295972130076,
      "loss": 5.9474,
      "mean_token_accuracy": 0.15022542774677278,
      "num_tokens": 9506293.0,
      "step": 4685
    },
    {
      "entropy": 6.093422698974609,
      "epoch": 0.40985755483701825,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004988264208345346,
      "loss": 5.8414,
      "mean_token_accuracy": 0.15870769023895265,
      "num_tokens": 9516127.0,
      "step": 4690
    },
    {
      "entropy": 6.027067184448242,
      "epoch": 0.410294503189723,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004988232401629535,
      "loss": 5.8373,
      "mean_token_accuracy": 0.15987443625926973,
      "num_tokens": 9525597.0,
      "step": 4695
    },
    {
      "entropy": 6.062117099761963,
      "epoch": 0.4107314515424277,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004988200551983254,
      "loss": 5.8184,
      "mean_token_accuracy": 0.15929939597845078,
      "num_tokens": 9535208.0,
      "step": 4700
    },
    {
      "entropy": 6.003118085861206,
      "epoch": 0.4111683998951324,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004988168659407113,
      "loss": 5.8633,
      "mean_token_accuracy": 0.1684591829776764,
      "num_tokens": 9545078.0,
      "step": 4705
    },
    {
      "entropy": 6.125294303894043,
      "epoch": 0.4116053482478371,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004988136723901725,
      "loss": 5.8786,
      "mean_token_accuracy": 0.150190307199955,
      "num_tokens": 9555074.0,
      "step": 4710
    },
    {
      "entropy": 6.101122856140137,
      "epoch": 0.4120422966005418,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004988104745467702,
      "loss": 5.8709,
      "mean_token_accuracy": 0.1622257947921753,
      "num_tokens": 9564777.0,
      "step": 4715
    },
    {
      "entropy": 6.001462745666504,
      "epoch": 0.4124792449532465,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004988072724105657,
      "loss": 5.8175,
      "mean_token_accuracy": 0.15790566354990004,
      "num_tokens": 9574551.0,
      "step": 4720
    },
    {
      "entropy": 6.049485874176026,
      "epoch": 0.41291619330595125,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004988040659816203,
      "loss": 5.9372,
      "mean_token_accuracy": 0.1452365405857563,
      "num_tokens": 9585162.0,
      "step": 4725
    },
    {
      "entropy": 6.094488048553467,
      "epoch": 0.41335314165865594,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004988008552599958,
      "loss": 5.8564,
      "mean_token_accuracy": 0.16154664307832717,
      "num_tokens": 9595422.0,
      "step": 4730
    },
    {
      "entropy": 6.05279803276062,
      "epoch": 0.41379009001136063,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004987976402457536,
      "loss": 5.8596,
      "mean_token_accuracy": 0.16663300693035127,
      "num_tokens": 9604315.0,
      "step": 4735
    },
    {
      "entropy": 6.052405071258545,
      "epoch": 0.4142270383640654,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004987944209389555,
      "loss": 5.8717,
      "mean_token_accuracy": 0.15581987276673318,
      "num_tokens": 9614369.0,
      "step": 4740
    },
    {
      "entropy": 6.124220561981201,
      "epoch": 0.41466398671677007,
      "grad_norm": 1.015625,
      "learning_rate": 0.000498791197339663,
      "loss": 5.918,
      "mean_token_accuracy": 0.15587083548307418,
      "num_tokens": 9624154.0,
      "step": 4745
    },
    {
      "entropy": 6.1386953830719,
      "epoch": 0.41510093506947476,
      "grad_norm": 1.046875,
      "learning_rate": 0.000498787969447938,
      "loss": 5.9248,
      "mean_token_accuracy": 0.15510949790477752,
      "num_tokens": 9634903.0,
      "step": 4750
    },
    {
      "entropy": 6.0277105331420895,
      "epoch": 0.4155378834221795,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004987847372638427,
      "loss": 5.9297,
      "mean_token_accuracy": 0.16144349426031113,
      "num_tokens": 9644873.0,
      "step": 4755
    },
    {
      "entropy": 6.073593044281006,
      "epoch": 0.4159748317748842,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004987815007874388,
      "loss": 5.8361,
      "mean_token_accuracy": 0.1696905180811882,
      "num_tokens": 9654661.0,
      "step": 4760
    },
    {
      "entropy": 6.0954677104949955,
      "epoch": 0.4164117801275889,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004987782600187884,
      "loss": 5.8489,
      "mean_token_accuracy": 0.16493165791034697,
      "num_tokens": 9664980.0,
      "step": 4765
    },
    {
      "entropy": 6.0639159202575685,
      "epoch": 0.41684872848029364,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004987750149579537,
      "loss": 5.8448,
      "mean_token_accuracy": 0.16241007000207902,
      "num_tokens": 9675117.0,
      "step": 4770
    },
    {
      "entropy": 6.075041389465332,
      "epoch": 0.41728567683299833,
      "grad_norm": 0.953125,
      "learning_rate": 0.000498771765604997,
      "loss": 5.8926,
      "mean_token_accuracy": 0.157795350253582,
      "num_tokens": 9684900.0,
      "step": 4775
    },
    {
      "entropy": 6.112214040756226,
      "epoch": 0.417722625185703,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004987685119599806,
      "loss": 5.8786,
      "mean_token_accuracy": 0.15652080923318862,
      "num_tokens": 9694586.0,
      "step": 4780
    },
    {
      "entropy": 6.060893487930298,
      "epoch": 0.41815957353840777,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004987652540229668,
      "loss": 5.9003,
      "mean_token_accuracy": 0.15676239281892776,
      "num_tokens": 9705676.0,
      "step": 4785
    },
    {
      "entropy": 6.09125394821167,
      "epoch": 0.41859652189111246,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004987619917940183,
      "loss": 5.8955,
      "mean_token_accuracy": 0.1618335574865341,
      "num_tokens": 9715442.0,
      "step": 4790
    },
    {
      "entropy": 6.161313915252686,
      "epoch": 0.4190334702438172,
      "grad_norm": 1.125,
      "learning_rate": 0.0004987587252731975,
      "loss": 5.9818,
      "mean_token_accuracy": 0.15474387481808663,
      "num_tokens": 9726621.0,
      "step": 4795
    },
    {
      "entropy": 6.088565111160278,
      "epoch": 0.4194704185965219,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987554544605671,
      "loss": 5.9995,
      "mean_token_accuracy": 0.14664513021707534,
      "num_tokens": 9736688.0,
      "step": 4800
    },
    {
      "entropy": 6.105941343307495,
      "epoch": 0.4199073669492266,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004987521793561898,
      "loss": 5.867,
      "mean_token_accuracy": 0.16337641477584838,
      "num_tokens": 9746478.0,
      "step": 4805
    },
    {
      "entropy": 6.060980653762817,
      "epoch": 0.42034431530193134,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004987488999601285,
      "loss": 5.895,
      "mean_token_accuracy": 0.16131228506565093,
      "num_tokens": 9756370.0,
      "step": 4810
    },
    {
      "entropy": 6.111278676986695,
      "epoch": 0.42078126365463603,
      "grad_norm": 0.96875,
      "learning_rate": 0.000498745616272446,
      "loss": 5.9062,
      "mean_token_accuracy": 0.15403577089309692,
      "num_tokens": 9766293.0,
      "step": 4815
    },
    {
      "entropy": 6.081045818328858,
      "epoch": 0.4212182120073407,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004987423282932054,
      "loss": 5.8926,
      "mean_token_accuracy": 0.15888576656579972,
      "num_tokens": 9777172.0,
      "step": 4820
    },
    {
      "entropy": 6.017177772521973,
      "epoch": 0.42165516036004547,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004987390360224697,
      "loss": 5.8654,
      "mean_token_accuracy": 0.15845351219177245,
      "num_tokens": 9787194.0,
      "step": 4825
    },
    {
      "entropy": 6.054066228866577,
      "epoch": 0.42209210871275016,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000498735739460302,
      "loss": 5.8879,
      "mean_token_accuracy": 0.15876548290252684,
      "num_tokens": 9797528.0,
      "step": 4830
    },
    {
      "entropy": 6.0701648712158205,
      "epoch": 0.42252905706545485,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004987324386067656,
      "loss": 5.9126,
      "mean_token_accuracy": 0.15464605987071992,
      "num_tokens": 9808412.0,
      "step": 4835
    },
    {
      "entropy": 6.138042879104614,
      "epoch": 0.4229660054181596,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004987291334619239,
      "loss": 6.0017,
      "mean_token_accuracy": 0.15031925290822984,
      "num_tokens": 9820047.0,
      "step": 4840
    },
    {
      "entropy": 6.090927648544311,
      "epoch": 0.4234029537708643,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00049872582402584,
      "loss": 5.8142,
      "mean_token_accuracy": 0.16150591224431993,
      "num_tokens": 9828812.0,
      "step": 4845
    },
    {
      "entropy": 6.036190605163574,
      "epoch": 0.423839902123569,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004987225102985778,
      "loss": 5.844,
      "mean_token_accuracy": 0.1648171916604042,
      "num_tokens": 9838288.0,
      "step": 4850
    },
    {
      "entropy": 6.139537572860718,
      "epoch": 0.4242768504762737,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004987191922802003,
      "loss": 5.8656,
      "mean_token_accuracy": 0.16157304048538207,
      "num_tokens": 9848536.0,
      "step": 4855
    },
    {
      "entropy": 5.993442964553833,
      "epoch": 0.4247137988289784,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004987158699707717,
      "loss": 5.9164,
      "mean_token_accuracy": 0.15896894112229348,
      "num_tokens": 9859741.0,
      "step": 4860
    },
    {
      "entropy": 6.098559093475342,
      "epoch": 0.4251507471816831,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004987125433703554,
      "loss": 5.8614,
      "mean_token_accuracy": 0.15820819586515428,
      "num_tokens": 9869173.0,
      "step": 4865
    },
    {
      "entropy": 6.124248886108399,
      "epoch": 0.42558769553438786,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004987092124790155,
      "loss": 5.8817,
      "mean_token_accuracy": 0.15666389614343643,
      "num_tokens": 9879619.0,
      "step": 4870
    },
    {
      "entropy": 6.026104688644409,
      "epoch": 0.42602464388709255,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004987058772968154,
      "loss": 5.8127,
      "mean_token_accuracy": 0.17008768022060394,
      "num_tokens": 9890214.0,
      "step": 4875
    },
    {
      "entropy": 6.005315113067627,
      "epoch": 0.42646159223979724,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004987025378238195,
      "loss": 5.8413,
      "mean_token_accuracy": 0.15807499885559081,
      "num_tokens": 9900266.0,
      "step": 4880
    },
    {
      "entropy": 6.104454135894775,
      "epoch": 0.426898540592502,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004986991940600917,
      "loss": 5.8823,
      "mean_token_accuracy": 0.15803514122962953,
      "num_tokens": 9910790.0,
      "step": 4885
    },
    {
      "entropy": 6.109101057052612,
      "epoch": 0.4273354889452067,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004986958460056961,
      "loss": 5.8886,
      "mean_token_accuracy": 0.15273160636425018,
      "num_tokens": 9921167.0,
      "step": 4890
    },
    {
      "entropy": 6.047287511825561,
      "epoch": 0.42777243729791137,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000498692493660697,
      "loss": 5.8354,
      "mean_token_accuracy": 0.1613122433423996,
      "num_tokens": 9931674.0,
      "step": 4895
    },
    {
      "entropy": 6.061746644973755,
      "epoch": 0.4282093856506161,
      "grad_norm": 1.0,
      "learning_rate": 0.0004986891370251587,
      "loss": 5.797,
      "mean_token_accuracy": 0.1594436228275299,
      "num_tokens": 9941484.0,
      "step": 4900
    },
    {
      "entropy": 6.105138444900513,
      "epoch": 0.4286463340033208,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004986857760991454,
      "loss": 5.9363,
      "mean_token_accuracy": 0.1525562062859535,
      "num_tokens": 9951885.0,
      "step": 4905
    },
    {
      "entropy": 6.06098484992981,
      "epoch": 0.4290832823560255,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004986824108827218,
      "loss": 5.7706,
      "mean_token_accuracy": 0.1636825129389763,
      "num_tokens": 9961658.0,
      "step": 4910
    },
    {
      "entropy": 5.964638996124267,
      "epoch": 0.42952023070873024,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004986790413759523,
      "loss": 5.8825,
      "mean_token_accuracy": 0.15844525247812272,
      "num_tokens": 9971971.0,
      "step": 4915
    },
    {
      "entropy": 6.047934532165527,
      "epoch": 0.42995717906143494,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004986756675789016,
      "loss": 5.781,
      "mean_token_accuracy": 0.15906192660331725,
      "num_tokens": 9982211.0,
      "step": 4920
    },
    {
      "entropy": 6.063864135742188,
      "epoch": 0.4303941274141396,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004986722894916344,
      "loss": 5.829,
      "mean_token_accuracy": 0.16408199965953826,
      "num_tokens": 9991935.0,
      "step": 4925
    },
    {
      "entropy": 6.013537645339966,
      "epoch": 0.4308310757668444,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004986689071142155,
      "loss": 5.8679,
      "mean_token_accuracy": 0.15154636055231094,
      "num_tokens": 10002014.0,
      "step": 4930
    },
    {
      "entropy": 6.070289945602417,
      "epoch": 0.43126802411954906,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004986655204467097,
      "loss": 5.8879,
      "mean_token_accuracy": 0.1567220062017441,
      "num_tokens": 10012030.0,
      "step": 4935
    },
    {
      "entropy": 6.046710205078125,
      "epoch": 0.43170497247225376,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004986621294891822,
      "loss": 5.8541,
      "mean_token_accuracy": 0.16050705462694168,
      "num_tokens": 10022188.0,
      "step": 4940
    },
    {
      "entropy": 6.055511045455932,
      "epoch": 0.4321419208249585,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004986587342416976,
      "loss": 5.7709,
      "mean_token_accuracy": 0.16340430974960327,
      "num_tokens": 10032131.0,
      "step": 4945
    },
    {
      "entropy": 6.071274852752685,
      "epoch": 0.4325788691776632,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004986553347043214,
      "loss": 5.8254,
      "mean_token_accuracy": 0.1580820009112358,
      "num_tokens": 10043204.0,
      "step": 4950
    },
    {
      "entropy": 6.045436668395996,
      "epoch": 0.4330158175303679,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004986519308771188,
      "loss": 5.878,
      "mean_token_accuracy": 0.15784966200590134,
      "num_tokens": 10052863.0,
      "step": 4955
    },
    {
      "entropy": 6.031168270111084,
      "epoch": 0.43345276588307263,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004986485227601548,
      "loss": 5.8628,
      "mean_token_accuracy": 0.1613911807537079,
      "num_tokens": 10063212.0,
      "step": 4960
    },
    {
      "entropy": 6.081945323944092,
      "epoch": 0.4338897142357773,
      "grad_norm": 0.859375,
      "learning_rate": 0.000498645110353495,
      "loss": 5.8857,
      "mean_token_accuracy": 0.15518673211336137,
      "num_tokens": 10074180.0,
      "step": 4965
    },
    {
      "entropy": 5.999382543563843,
      "epoch": 0.434326662588482,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004986416936572049,
      "loss": 5.8379,
      "mean_token_accuracy": 0.16377767473459243,
      "num_tokens": 10084513.0,
      "step": 4970
    },
    {
      "entropy": 6.074960422515869,
      "epoch": 0.43476361094118676,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004986382726713497,
      "loss": 5.8505,
      "mean_token_accuracy": 0.15784327685832977,
      "num_tokens": 10094352.0,
      "step": 4975
    },
    {
      "entropy": 5.977413606643677,
      "epoch": 0.43520055929389145,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004986348473959955,
      "loss": 5.7162,
      "mean_token_accuracy": 0.17215230762958528,
      "num_tokens": 10103677.0,
      "step": 4980
    },
    {
      "entropy": 6.051468229293823,
      "epoch": 0.43563750764659614,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004986314178312076,
      "loss": 5.9413,
      "mean_token_accuracy": 0.1489313840866089,
      "num_tokens": 10113479.0,
      "step": 4985
    },
    {
      "entropy": 6.044587564468384,
      "epoch": 0.4360744559993009,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004986279839770519,
      "loss": 5.8317,
      "mean_token_accuracy": 0.15717597603797911,
      "num_tokens": 10123526.0,
      "step": 4990
    },
    {
      "entropy": 6.066796016693115,
      "epoch": 0.4365114043520056,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004986245458335944,
      "loss": 5.8376,
      "mean_token_accuracy": 0.16081769466400148,
      "num_tokens": 10134176.0,
      "step": 4995
    },
    {
      "entropy": 6.068060350418091,
      "epoch": 0.43694835270471033,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004986211034009008,
      "loss": 5.8144,
      "mean_token_accuracy": 0.16548700779676437,
      "num_tokens": 10144763.0,
      "step": 5000
    },
    {
      "entropy": 5.991634321212769,
      "epoch": 0.437385301057415,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004986176566790375,
      "loss": 5.7855,
      "mean_token_accuracy": 0.168276646733284,
      "num_tokens": 10155011.0,
      "step": 5005
    },
    {
      "entropy": 6.06312198638916,
      "epoch": 0.4378222494101197,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004986142056680701,
      "loss": 5.8676,
      "mean_token_accuracy": 0.16136534363031388,
      "num_tokens": 10165474.0,
      "step": 5010
    },
    {
      "entropy": 6.026047563552856,
      "epoch": 0.43825919776282446,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004986107503680652,
      "loss": 5.8982,
      "mean_token_accuracy": 0.16189261078834533,
      "num_tokens": 10175137.0,
      "step": 5015
    },
    {
      "entropy": 6.047072029113769,
      "epoch": 0.43869614611552915,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004986072907790889,
      "loss": 5.8974,
      "mean_token_accuracy": 0.15533759593963622,
      "num_tokens": 10185623.0,
      "step": 5020
    },
    {
      "entropy": 6.128658676147461,
      "epoch": 0.43913309446823384,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004986038269012076,
      "loss": 5.8535,
      "mean_token_accuracy": 0.16214749664068223,
      "num_tokens": 10195201.0,
      "step": 5025
    },
    {
      "entropy": 6.117952632904053,
      "epoch": 0.4395700428209386,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004986003587344877,
      "loss": 5.8909,
      "mean_token_accuracy": 0.1607968732714653,
      "num_tokens": 10205951.0,
      "step": 5030
    },
    {
      "entropy": 6.004427528381347,
      "epoch": 0.4400069911736433,
      "grad_norm": 1.0,
      "learning_rate": 0.0004985968862789957,
      "loss": 5.7733,
      "mean_token_accuracy": 0.1643867775797844,
      "num_tokens": 10215665.0,
      "step": 5035
    },
    {
      "entropy": 6.05168342590332,
      "epoch": 0.44044393952634797,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004985934095347985,
      "loss": 5.8655,
      "mean_token_accuracy": 0.15620046705007554,
      "num_tokens": 10225037.0,
      "step": 5040
    },
    {
      "entropy": 6.043855953216553,
      "epoch": 0.4408808878790527,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004985899285019624,
      "loss": 5.8131,
      "mean_token_accuracy": 0.15734778940677643,
      "num_tokens": 10234581.0,
      "step": 5045
    },
    {
      "entropy": 6.046975469589233,
      "epoch": 0.4413178362317574,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004985864431805543,
      "loss": 5.7304,
      "mean_token_accuracy": 0.1733717739582062,
      "num_tokens": 10243716.0,
      "step": 5050
    },
    {
      "entropy": 5.99574728012085,
      "epoch": 0.4417547845844621,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004985829535706411,
      "loss": 5.8451,
      "mean_token_accuracy": 0.1616597831249237,
      "num_tokens": 10253198.0,
      "step": 5055
    },
    {
      "entropy": 6.027127122879028,
      "epoch": 0.44219173293716685,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004985794596722897,
      "loss": 5.9086,
      "mean_token_accuracy": 0.1528984524309635,
      "num_tokens": 10262912.0,
      "step": 5060
    },
    {
      "entropy": 6.0631575107574465,
      "epoch": 0.44262868128987154,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000498575961485567,
      "loss": 5.8604,
      "mean_token_accuracy": 0.1614742785692215,
      "num_tokens": 10272960.0,
      "step": 5065
    },
    {
      "entropy": 6.086499118804932,
      "epoch": 0.44306562964257623,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004985724590105403,
      "loss": 5.8992,
      "mean_token_accuracy": 0.15877998173236846,
      "num_tokens": 10283034.0,
      "step": 5070
    },
    {
      "entropy": 6.02888765335083,
      "epoch": 0.443502577995281,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004985689522472767,
      "loss": 5.8597,
      "mean_token_accuracy": 0.16141433417797088,
      "num_tokens": 10292553.0,
      "step": 5075
    },
    {
      "entropy": 5.964289808273316,
      "epoch": 0.44393952634798567,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004985654411958434,
      "loss": 5.8195,
      "mean_token_accuracy": 0.16030110716819762,
      "num_tokens": 10302475.0,
      "step": 5080
    },
    {
      "entropy": 6.032690763473511,
      "epoch": 0.44437647470069036,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004985619258563078,
      "loss": 5.836,
      "mean_token_accuracy": 0.16374730169773102,
      "num_tokens": 10311693.0,
      "step": 5085
    },
    {
      "entropy": 6.071450519561767,
      "epoch": 0.4448134230533951,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004985584062287372,
      "loss": 5.8322,
      "mean_token_accuracy": 0.16497460454702378,
      "num_tokens": 10323052.0,
      "step": 5090
    },
    {
      "entropy": 6.017006254196167,
      "epoch": 0.4452503714060998,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004985548823131994,
      "loss": 5.7986,
      "mean_token_accuracy": 0.16336701512336732,
      "num_tokens": 10332997.0,
      "step": 5095
    },
    {
      "entropy": 6.1304384708404545,
      "epoch": 0.4456873197588045,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004985513541097618,
      "loss": 5.9213,
      "mean_token_accuracy": 0.1565610408782959,
      "num_tokens": 10343298.0,
      "step": 5100
    },
    {
      "entropy": 6.117843437194824,
      "epoch": 0.44612426811150924,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000498547821618492,
      "loss": 5.8299,
      "mean_token_accuracy": 0.1553485795855522,
      "num_tokens": 10352485.0,
      "step": 5105
    },
    {
      "entropy": 5.946697759628296,
      "epoch": 0.44656121646421393,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004985442848394579,
      "loss": 5.8862,
      "mean_token_accuracy": 0.15780410319566726,
      "num_tokens": 10362810.0,
      "step": 5110
    },
    {
      "entropy": 6.060162973403931,
      "epoch": 0.4469981648169186,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004985407437727273,
      "loss": 5.813,
      "mean_token_accuracy": 0.16509923189878464,
      "num_tokens": 10372817.0,
      "step": 5115
    },
    {
      "entropy": 6.0719382762908936,
      "epoch": 0.44743511316962337,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004985371984183681,
      "loss": 5.841,
      "mean_token_accuracy": 0.16420152634382248,
      "num_tokens": 10383046.0,
      "step": 5120
    },
    {
      "entropy": 6.040710592269898,
      "epoch": 0.44787206152232806,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004985336487764483,
      "loss": 5.8936,
      "mean_token_accuracy": 0.15618296563625336,
      "num_tokens": 10393498.0,
      "step": 5125
    },
    {
      "entropy": 6.090593862533569,
      "epoch": 0.44830900987503275,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004985300948470359,
      "loss": 5.8998,
      "mean_token_accuracy": 0.1548370435833931,
      "num_tokens": 10403595.0,
      "step": 5130
    },
    {
      "entropy": 6.074947261810303,
      "epoch": 0.4487459582277375,
      "grad_norm": 0.875,
      "learning_rate": 0.0004985265366301992,
      "loss": 5.8293,
      "mean_token_accuracy": 0.1571945935487747,
      "num_tokens": 10413475.0,
      "step": 5135
    },
    {
      "entropy": 5.995162725448608,
      "epoch": 0.4491829065804422,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004985229741260065,
      "loss": 5.8509,
      "mean_token_accuracy": 0.16885699033737184,
      "num_tokens": 10423512.0,
      "step": 5140
    },
    {
      "entropy": 6.03759446144104,
      "epoch": 0.4496198549331469,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004985194073345259,
      "loss": 5.8711,
      "mean_token_accuracy": 0.1559654250741005,
      "num_tokens": 10434198.0,
      "step": 5145
    },
    {
      "entropy": 6.108378076553345,
      "epoch": 0.4500568032858516,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004985158362558261,
      "loss": 5.7852,
      "mean_token_accuracy": 0.16432345807552337,
      "num_tokens": 10444077.0,
      "step": 5150
    },
    {
      "entropy": 6.030838918685913,
      "epoch": 0.4504937516385563,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004985122608899752,
      "loss": 5.8514,
      "mean_token_accuracy": 0.1590752437710762,
      "num_tokens": 10454481.0,
      "step": 5155
    },
    {
      "entropy": 6.05402102470398,
      "epoch": 0.450930699991261,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004985086812370422,
      "loss": 5.865,
      "mean_token_accuracy": 0.1630973607301712,
      "num_tokens": 10464453.0,
      "step": 5160
    },
    {
      "entropy": 6.0818284034729,
      "epoch": 0.45136764834396575,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004985050972970956,
      "loss": 5.9347,
      "mean_token_accuracy": 0.15671077519655227,
      "num_tokens": 10474932.0,
      "step": 5165
    },
    {
      "entropy": 6.126150608062744,
      "epoch": 0.45180459669667045,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000498501509070204,
      "loss": 5.9369,
      "mean_token_accuracy": 0.15456461161375046,
      "num_tokens": 10484719.0,
      "step": 5170
    },
    {
      "entropy": 6.088868951797485,
      "epoch": 0.45224154504937514,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004984979165564364,
      "loss": 5.9693,
      "mean_token_accuracy": 0.15047609955072402,
      "num_tokens": 10495808.0,
      "step": 5175
    },
    {
      "entropy": 6.116308403015137,
      "epoch": 0.4526784934020799,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004984943197558616,
      "loss": 5.872,
      "mean_token_accuracy": 0.15841442346572876,
      "num_tokens": 10506528.0,
      "step": 5180
    },
    {
      "entropy": 6.0591692447662355,
      "epoch": 0.4531154417547846,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004984907186685487,
      "loss": 5.8393,
      "mean_token_accuracy": 0.16865510791540145,
      "num_tokens": 10517117.0,
      "step": 5185
    },
    {
      "entropy": 6.150075531005859,
      "epoch": 0.45355239010748927,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004984871132945667,
      "loss": 5.8452,
      "mean_token_accuracy": 0.15962809175252915,
      "num_tokens": 10527394.0,
      "step": 5190
    },
    {
      "entropy": 6.046996831893921,
      "epoch": 0.453989338460194,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004984835036339848,
      "loss": 5.8493,
      "mean_token_accuracy": 0.15912731736898422,
      "num_tokens": 10536779.0,
      "step": 5195
    },
    {
      "entropy": 6.051007223129273,
      "epoch": 0.4544262868128987,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004984798896868722,
      "loss": 5.9198,
      "mean_token_accuracy": 0.1543247103691101,
      "num_tokens": 10547116.0,
      "step": 5200
    },
    {
      "entropy": 6.046619653701782,
      "epoch": 0.45486323516560345,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004984762714532981,
      "loss": 5.8006,
      "mean_token_accuracy": 0.1617807999253273,
      "num_tokens": 10556396.0,
      "step": 5205
    },
    {
      "entropy": 6.009440135955811,
      "epoch": 0.45530018351830814,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000498472648933332,
      "loss": 5.9813,
      "mean_token_accuracy": 0.15642509534955024,
      "num_tokens": 10568127.0,
      "step": 5210
    },
    {
      "entropy": 6.121495199203491,
      "epoch": 0.45573713187101283,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004984690221270436,
      "loss": 5.8095,
      "mean_token_accuracy": 0.16523925662040712,
      "num_tokens": 10578565.0,
      "step": 5215
    },
    {
      "entropy": 6.056030941009522,
      "epoch": 0.4561740802237176,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004984653910345021,
      "loss": 5.841,
      "mean_token_accuracy": 0.1518455281853676,
      "num_tokens": 10588633.0,
      "step": 5220
    },
    {
      "entropy": 6.040780162811279,
      "epoch": 0.4566110285764223,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004984617556557774,
      "loss": 5.8001,
      "mean_token_accuracy": 0.1605903573334217,
      "num_tokens": 10597917.0,
      "step": 5225
    },
    {
      "entropy": 6.0840130805969235,
      "epoch": 0.45704797692912696,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000498458115990939,
      "loss": 5.8423,
      "mean_token_accuracy": 0.1549955815076828,
      "num_tokens": 10608396.0,
      "step": 5230
    },
    {
      "entropy": 6.020921754837036,
      "epoch": 0.4574849252818317,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004984544720400568,
      "loss": 5.8634,
      "mean_token_accuracy": 0.165152008831501,
      "num_tokens": 10618191.0,
      "step": 5235
    },
    {
      "entropy": 6.053033781051636,
      "epoch": 0.4579218736345364,
      "grad_norm": 0.96875,
      "learning_rate": 0.000498450823803201,
      "loss": 5.85,
      "mean_token_accuracy": 0.15579962134361267,
      "num_tokens": 10627594.0,
      "step": 5240
    },
    {
      "entropy": 6.005799579620361,
      "epoch": 0.4583588219872411,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004984471712804412,
      "loss": 5.8278,
      "mean_token_accuracy": 0.1651892602443695,
      "num_tokens": 10637260.0,
      "step": 5245
    },
    {
      "entropy": 6.091345500946045,
      "epoch": 0.45879577033994584,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004984435144718475,
      "loss": 5.8576,
      "mean_token_accuracy": 0.16034567952156067,
      "num_tokens": 10647968.0,
      "step": 5250
    },
    {
      "entropy": 5.984779310226441,
      "epoch": 0.45923271869265053,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004984398533774901,
      "loss": 5.801,
      "mean_token_accuracy": 0.16446565836668015,
      "num_tokens": 10657760.0,
      "step": 5255
    },
    {
      "entropy": 6.0619738578796385,
      "epoch": 0.4596696670453552,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004984361879974393,
      "loss": 5.8208,
      "mean_token_accuracy": 0.1531989634037018,
      "num_tokens": 10667577.0,
      "step": 5260
    },
    {
      "entropy": 6.041530656814575,
      "epoch": 0.46010661539805997,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004984325183317654,
      "loss": 5.7777,
      "mean_token_accuracy": 0.1609012246131897,
      "num_tokens": 10677984.0,
      "step": 5265
    },
    {
      "entropy": 6.006534910202026,
      "epoch": 0.46054356375076466,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004984288443805386,
      "loss": 5.8469,
      "mean_token_accuracy": 0.16630094945430757,
      "num_tokens": 10687700.0,
      "step": 5270
    },
    {
      "entropy": 6.0508713722229,
      "epoch": 0.46098051210346935,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004984251661438295,
      "loss": 5.8154,
      "mean_token_accuracy": 0.16486253142356871,
      "num_tokens": 10697351.0,
      "step": 5275
    },
    {
      "entropy": 6.017149782180786,
      "epoch": 0.4614174604561741,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004984214836217087,
      "loss": 5.8477,
      "mean_token_accuracy": 0.16182752847671508,
      "num_tokens": 10707498.0,
      "step": 5280
    },
    {
      "entropy": 6.051641941070557,
      "epoch": 0.4618544088088788,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004984177968142467,
      "loss": 5.8758,
      "mean_token_accuracy": 0.15616871565580367,
      "num_tokens": 10717724.0,
      "step": 5285
    },
    {
      "entropy": 6.07336106300354,
      "epoch": 0.4622913571615835,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004984141057215143,
      "loss": 5.8272,
      "mean_token_accuracy": 0.1631024494767189,
      "num_tokens": 10727030.0,
      "step": 5290
    },
    {
      "entropy": 6.008185625076294,
      "epoch": 0.46272830551428823,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004984104103435824,
      "loss": 5.714,
      "mean_token_accuracy": 0.17103925347328186,
      "num_tokens": 10736861.0,
      "step": 5295
    },
    {
      "entropy": 6.095879459381104,
      "epoch": 0.4631652538669929,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004984067106805217,
      "loss": 5.9441,
      "mean_token_accuracy": 0.15545603185892104,
      "num_tokens": 10748310.0,
      "step": 5300
    },
    {
      "entropy": 6.018623495101929,
      "epoch": 0.4636022022196976,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004984030067324032,
      "loss": 5.7891,
      "mean_token_accuracy": 0.1631061002612114,
      "num_tokens": 10758110.0,
      "step": 5305
    },
    {
      "entropy": 6.020611906051636,
      "epoch": 0.46403915057240236,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004983992984992979,
      "loss": 5.8225,
      "mean_token_accuracy": 0.16338275372982025,
      "num_tokens": 10767919.0,
      "step": 5310
    },
    {
      "entropy": 6.022727155685425,
      "epoch": 0.46447609892510705,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004983955859812771,
      "loss": 5.7562,
      "mean_token_accuracy": 0.16483165621757506,
      "num_tokens": 10776888.0,
      "step": 5315
    },
    {
      "entropy": 6.052799797058105,
      "epoch": 0.46491304727781174,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004983918691784119,
      "loss": 5.8274,
      "mean_token_accuracy": 0.1591514155268669,
      "num_tokens": 10786806.0,
      "step": 5320
    },
    {
      "entropy": 6.066069602966309,
      "epoch": 0.4653499956305165,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004983881480907736,
      "loss": 5.8131,
      "mean_token_accuracy": 0.16172914505004882,
      "num_tokens": 10796452.0,
      "step": 5325
    },
    {
      "entropy": 6.039560317993164,
      "epoch": 0.4657869439832212,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004983844227184334,
      "loss": 5.7836,
      "mean_token_accuracy": 0.15788405239582062,
      "num_tokens": 10806718.0,
      "step": 5330
    },
    {
      "entropy": 6.05526909828186,
      "epoch": 0.46622389233592587,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004983806930614631,
      "loss": 5.8807,
      "mean_token_accuracy": 0.15862741619348525,
      "num_tokens": 10817978.0,
      "step": 5335
    },
    {
      "entropy": 5.987653541564941,
      "epoch": 0.4666608406886306,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000498376959119934,
      "loss": 5.752,
      "mean_token_accuracy": 0.16792583614587783,
      "num_tokens": 10827818.0,
      "step": 5340
    },
    {
      "entropy": 6.068410873413086,
      "epoch": 0.4670977890413353,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004983732208939178,
      "loss": 5.9019,
      "mean_token_accuracy": 0.15744334906339646,
      "num_tokens": 10838492.0,
      "step": 5345
    },
    {
      "entropy": 6.038463068008423,
      "epoch": 0.46753473739404,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004983694783834863,
      "loss": 5.7859,
      "mean_token_accuracy": 0.16228380501270295,
      "num_tokens": 10847807.0,
      "step": 5350
    },
    {
      "entropy": 6.011783218383789,
      "epoch": 0.46797168574674475,
      "grad_norm": 0.875,
      "learning_rate": 0.0004983657315887109,
      "loss": 5.8675,
      "mean_token_accuracy": 0.1569453775882721,
      "num_tokens": 10858637.0,
      "step": 5355
    },
    {
      "entropy": 6.085317325592041,
      "epoch": 0.46840863409944944,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004983619805096639,
      "loss": 5.7959,
      "mean_token_accuracy": 0.16722953766584397,
      "num_tokens": 10868042.0,
      "step": 5360
    },
    {
      "entropy": 6.020178699493409,
      "epoch": 0.46884558245215413,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004983582251464171,
      "loss": 5.8142,
      "mean_token_accuracy": 0.1631641462445259,
      "num_tokens": 10877171.0,
      "step": 5365
    },
    {
      "entropy": 6.010836172103882,
      "epoch": 0.4692825308048589,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004983544654990423,
      "loss": 5.8283,
      "mean_token_accuracy": 0.15954248309135438,
      "num_tokens": 10888434.0,
      "step": 5370
    },
    {
      "entropy": 6.0488402366638185,
      "epoch": 0.46971947915756357,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000498350701567612,
      "loss": 5.8438,
      "mean_token_accuracy": 0.1614022135734558,
      "num_tokens": 10898467.0,
      "step": 5375
    },
    {
      "entropy": 5.982282066345215,
      "epoch": 0.47015642751026826,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004983469333521983,
      "loss": 5.8233,
      "mean_token_accuracy": 0.16743752509355544,
      "num_tokens": 10908178.0,
      "step": 5380
    },
    {
      "entropy": 6.041627407073975,
      "epoch": 0.470593375862973,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004983431608528732,
      "loss": 5.8917,
      "mean_token_accuracy": 0.15152108669281006,
      "num_tokens": 10919957.0,
      "step": 5385
    },
    {
      "entropy": 6.1094520568847654,
      "epoch": 0.4710303242156777,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004983393840697092,
      "loss": 5.8584,
      "mean_token_accuracy": 0.15698882341384887,
      "num_tokens": 10929728.0,
      "step": 5390
    },
    {
      "entropy": 6.029668283462525,
      "epoch": 0.4714672725683824,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004983356030027789,
      "loss": 5.8847,
      "mean_token_accuracy": 0.15731559246778487,
      "num_tokens": 10940427.0,
      "step": 5395
    },
    {
      "entropy": 5.976920127868652,
      "epoch": 0.47190422092108714,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004983318176521546,
      "loss": 5.8508,
      "mean_token_accuracy": 0.16184997111558913,
      "num_tokens": 10950641.0,
      "step": 5400
    },
    {
      "entropy": 6.120208978652954,
      "epoch": 0.4723411692737918,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004983280280179092,
      "loss": 5.7857,
      "mean_token_accuracy": 0.16230563670396805,
      "num_tokens": 10959341.0,
      "step": 5405
    },
    {
      "entropy": 6.034132814407348,
      "epoch": 0.4727781176264966,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000498324234100115,
      "loss": 5.8173,
      "mean_token_accuracy": 0.1641610085964203,
      "num_tokens": 10967745.0,
      "step": 5410
    },
    {
      "entropy": 5.9795153617858885,
      "epoch": 0.47321506597920127,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004983204358988451,
      "loss": 5.798,
      "mean_token_accuracy": 0.16295356452465057,
      "num_tokens": 10978345.0,
      "step": 5415
    },
    {
      "entropy": 6.076958560943604,
      "epoch": 0.47365201433190596,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004983166334141722,
      "loss": 5.8467,
      "mean_token_accuracy": 0.16045865043997765,
      "num_tokens": 10988030.0,
      "step": 5420
    },
    {
      "entropy": 6.091433525085449,
      "epoch": 0.4740889626846107,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004983128266461693,
      "loss": 5.8152,
      "mean_token_accuracy": 0.1592065677046776,
      "num_tokens": 10999169.0,
      "step": 5425
    },
    {
      "entropy": 6.029356813430786,
      "epoch": 0.4745259110373154,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004983090155949093,
      "loss": 5.905,
      "mean_token_accuracy": 0.15437909364700317,
      "num_tokens": 11009608.0,
      "step": 5430
    },
    {
      "entropy": 6.098097276687622,
      "epoch": 0.4749628593900201,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004983052002604653,
      "loss": 5.8334,
      "mean_token_accuracy": 0.15981026887893676,
      "num_tokens": 11020187.0,
      "step": 5435
    },
    {
      "entropy": 6.060362100601196,
      "epoch": 0.47539980774272483,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004983013806429107,
      "loss": 5.807,
      "mean_token_accuracy": 0.16479516327381133,
      "num_tokens": 11030459.0,
      "step": 5440
    },
    {
      "entropy": 5.956500768661499,
      "epoch": 0.4758367560954295,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004982975567423185,
      "loss": 5.7689,
      "mean_token_accuracy": 0.16472658514976501,
      "num_tokens": 11040880.0,
      "step": 5445
    },
    {
      "entropy": 6.010826110839844,
      "epoch": 0.4762737044481342,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004982937285587623,
      "loss": 5.7813,
      "mean_token_accuracy": 0.16700105369091034,
      "num_tokens": 11050901.0,
      "step": 5450
    },
    {
      "entropy": 6.093382978439331,
      "epoch": 0.47671065280083896,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004982898960923152,
      "loss": 5.9504,
      "mean_token_accuracy": 0.15278058052062987,
      "num_tokens": 11061274.0,
      "step": 5455
    },
    {
      "entropy": 6.060980224609375,
      "epoch": 0.47714760115354365,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004982860593430509,
      "loss": 5.7692,
      "mean_token_accuracy": 0.16059326827526094,
      "num_tokens": 11070644.0,
      "step": 5460
    },
    {
      "entropy": 6.0266416549682615,
      "epoch": 0.47758454950624835,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004982822183110431,
      "loss": 5.8256,
      "mean_token_accuracy": 0.16112038344144822,
      "num_tokens": 11081154.0,
      "step": 5465
    },
    {
      "entropy": 6.0233681201934814,
      "epoch": 0.4780214978589531,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004982783729963654,
      "loss": 5.9141,
      "mean_token_accuracy": 0.1576973170042038,
      "num_tokens": 11091929.0,
      "step": 5470
    },
    {
      "entropy": 5.962234926223755,
      "epoch": 0.4784584462116578,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004982745233990914,
      "loss": 5.7457,
      "mean_token_accuracy": 0.16974477767944335,
      "num_tokens": 11102075.0,
      "step": 5475
    },
    {
      "entropy": 6.016850423812866,
      "epoch": 0.4788953945643625,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004982706695192949,
      "loss": 5.8322,
      "mean_token_accuracy": 0.16025734543800355,
      "num_tokens": 11111886.0,
      "step": 5480
    },
    {
      "entropy": 6.073599863052368,
      "epoch": 0.4793323429170672,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004982668113570503,
      "loss": 5.8412,
      "mean_token_accuracy": 0.15446365475654603,
      "num_tokens": 11122329.0,
      "step": 5485
    },
    {
      "entropy": 6.024536085128784,
      "epoch": 0.4797692912697719,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000498262948912431,
      "loss": 5.7635,
      "mean_token_accuracy": 0.1623012751340866,
      "num_tokens": 11132348.0,
      "step": 5490
    },
    {
      "entropy": 6.040790319442749,
      "epoch": 0.4802062396224766,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004982590821855115,
      "loss": 5.8298,
      "mean_token_accuracy": 0.16335973143577576,
      "num_tokens": 11141975.0,
      "step": 5495
    },
    {
      "entropy": 6.059398031234741,
      "epoch": 0.48064318797518135,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004982552111763657,
      "loss": 5.8426,
      "mean_token_accuracy": 0.16131259351968766,
      "num_tokens": 11152174.0,
      "step": 5500
    },
    {
      "entropy": 6.019055700302124,
      "epoch": 0.48108013632788604,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004982513358850681,
      "loss": 5.8351,
      "mean_token_accuracy": 0.15886461287736892,
      "num_tokens": 11162563.0,
      "step": 5505
    },
    {
      "entropy": 6.026846933364868,
      "epoch": 0.48151708468059073,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004982474563116928,
      "loss": 5.7304,
      "mean_token_accuracy": 0.16977405399084092,
      "num_tokens": 11172572.0,
      "step": 5510
    },
    {
      "entropy": 6.012419843673706,
      "epoch": 0.4819540330332955,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004982435724563143,
      "loss": 5.8284,
      "mean_token_accuracy": 0.16790070682764052,
      "num_tokens": 11183404.0,
      "step": 5515
    },
    {
      "entropy": 6.001372241973877,
      "epoch": 0.4823909813860002,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004982396843190072,
      "loss": 5.836,
      "mean_token_accuracy": 0.16337916404008865,
      "num_tokens": 11193905.0,
      "step": 5520
    },
    {
      "entropy": 6.035560655593872,
      "epoch": 0.48282792973870486,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004982357918998459,
      "loss": 5.7365,
      "mean_token_accuracy": 0.17206798493862152,
      "num_tokens": 11203127.0,
      "step": 5525
    },
    {
      "entropy": 5.979992294311524,
      "epoch": 0.4832648780914096,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004982318951989053,
      "loss": 5.7384,
      "mean_token_accuracy": 0.1704808935523033,
      "num_tokens": 11213468.0,
      "step": 5530
    },
    {
      "entropy": 5.996744871139526,
      "epoch": 0.4837018264441143,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004982279942162597,
      "loss": 5.8394,
      "mean_token_accuracy": 0.16231720745563508,
      "num_tokens": 11224722.0,
      "step": 5535
    },
    {
      "entropy": 6.065119409561158,
      "epoch": 0.484138774796819,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004982240889519844,
      "loss": 5.8138,
      "mean_token_accuracy": 0.16752776354551316,
      "num_tokens": 11234565.0,
      "step": 5540
    },
    {
      "entropy": 6.033210182189942,
      "epoch": 0.48457572314952374,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000498220179406154,
      "loss": 5.813,
      "mean_token_accuracy": 0.1631343275308609,
      "num_tokens": 11244496.0,
      "step": 5545
    },
    {
      "entropy": 6.082521963119507,
      "epoch": 0.48501267150222843,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004982162655788436,
      "loss": 5.791,
      "mean_token_accuracy": 0.16825300455093384,
      "num_tokens": 11254187.0,
      "step": 5550
    },
    {
      "entropy": 6.020895385742188,
      "epoch": 0.4854496198549331,
      "grad_norm": 1.0,
      "learning_rate": 0.0004982123474701283,
      "loss": 5.7618,
      "mean_token_accuracy": 0.1722523882985115,
      "num_tokens": 11264580.0,
      "step": 5555
    },
    {
      "entropy": 6.036573171615601,
      "epoch": 0.48588656820763787,
      "grad_norm": 1.0,
      "learning_rate": 0.0004982084250800832,
      "loss": 5.9193,
      "mean_token_accuracy": 0.15699466690421104,
      "num_tokens": 11275104.0,
      "step": 5560
    },
    {
      "entropy": 6.105719757080078,
      "epoch": 0.48632351656034256,
      "grad_norm": 1.0,
      "learning_rate": 0.0004982044984087836,
      "loss": 5.9343,
      "mean_token_accuracy": 0.15429940968751907,
      "num_tokens": 11285983.0,
      "step": 5565
    },
    {
      "entropy": 6.0477532863616945,
      "epoch": 0.48676046491304725,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004982005674563048,
      "loss": 5.8102,
      "mean_token_accuracy": 0.1640915408730507,
      "num_tokens": 11295378.0,
      "step": 5570
    },
    {
      "entropy": 6.027726554870606,
      "epoch": 0.487197413265752,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000498196632222722,
      "loss": 5.916,
      "mean_token_accuracy": 0.15790842175483705,
      "num_tokens": 11305500.0,
      "step": 5575
    },
    {
      "entropy": 6.0338860034942625,
      "epoch": 0.4876343616184567,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004981926927081109,
      "loss": 5.8265,
      "mean_token_accuracy": 0.16164934933185576,
      "num_tokens": 11316455.0,
      "step": 5580
    },
    {
      "entropy": 6.046772527694702,
      "epoch": 0.4880713099711614,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004981887489125471,
      "loss": 5.7915,
      "mean_token_accuracy": 0.1576798029243946,
      "num_tokens": 11325865.0,
      "step": 5585
    },
    {
      "entropy": 5.925597429275513,
      "epoch": 0.48850825832386613,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000498184800836106,
      "loss": 5.7527,
      "mean_token_accuracy": 0.17598473727703096,
      "num_tokens": 11337368.0,
      "step": 5590
    },
    {
      "entropy": 5.969305610656738,
      "epoch": 0.4889452066765708,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004981808484788635,
      "loss": 5.7063,
      "mean_token_accuracy": 0.1729299783706665,
      "num_tokens": 11347129.0,
      "step": 5595
    },
    {
      "entropy": 6.066465854644775,
      "epoch": 0.48938215502927557,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004981768918408955,
      "loss": 5.795,
      "mean_token_accuracy": 0.16336847692728043,
      "num_tokens": 11356641.0,
      "step": 5600
    },
    {
      "entropy": 5.977285671234131,
      "epoch": 0.48981910338198026,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004981729309222777,
      "loss": 5.682,
      "mean_token_accuracy": 0.16987183690071106,
      "num_tokens": 11366485.0,
      "step": 5605
    },
    {
      "entropy": 5.995461177825928,
      "epoch": 0.49025605173468495,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004981689657230861,
      "loss": 5.8322,
      "mean_token_accuracy": 0.1667294904589653,
      "num_tokens": 11376305.0,
      "step": 5610
    },
    {
      "entropy": 6.0156971454620365,
      "epoch": 0.4906930000873897,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000498164996243397,
      "loss": 5.793,
      "mean_token_accuracy": 0.16228654682636262,
      "num_tokens": 11385837.0,
      "step": 5615
    },
    {
      "entropy": 6.026220655441284,
      "epoch": 0.4911299484400944,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004981610224832861,
      "loss": 5.8491,
      "mean_token_accuracy": 0.16134712249040603,
      "num_tokens": 11396719.0,
      "step": 5620
    },
    {
      "entropy": 6.069628572463989,
      "epoch": 0.4915668967927991,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004981570444428299,
      "loss": 5.8547,
      "mean_token_accuracy": 0.1600213348865509,
      "num_tokens": 11408107.0,
      "step": 5625
    },
    {
      "entropy": 6.007379341125488,
      "epoch": 0.4920038451455038,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004981530621221048,
      "loss": 5.7696,
      "mean_token_accuracy": 0.17298493683338165,
      "num_tokens": 11417547.0,
      "step": 5630
    },
    {
      "entropy": 6.029811763763428,
      "epoch": 0.4924407934982085,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004981490755211869,
      "loss": 5.8441,
      "mean_token_accuracy": 0.16345840990543364,
      "num_tokens": 11427660.0,
      "step": 5635
    },
    {
      "entropy": 5.960938930511475,
      "epoch": 0.4928777418509132,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004981450846401528,
      "loss": 5.7179,
      "mean_token_accuracy": 0.16941920965909957,
      "num_tokens": 11438032.0,
      "step": 5640
    },
    {
      "entropy": 6.041319370269775,
      "epoch": 0.49331469020361796,
      "grad_norm": 0.96875,
      "learning_rate": 0.000498141089479079,
      "loss": 5.8244,
      "mean_token_accuracy": 0.16358433961868285,
      "num_tokens": 11448987.0,
      "step": 5645
    },
    {
      "entropy": 6.022552537918091,
      "epoch": 0.49375163855632265,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004981370900380423,
      "loss": 5.8003,
      "mean_token_accuracy": 0.16702569872140885,
      "num_tokens": 11458659.0,
      "step": 5650
    },
    {
      "entropy": 5.984081888198853,
      "epoch": 0.49418858690902734,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004981330863171194,
      "loss": 5.7798,
      "mean_token_accuracy": 0.16197975873947143,
      "num_tokens": 11468104.0,
      "step": 5655
    },
    {
      "entropy": 6.013318347930908,
      "epoch": 0.4946255352617321,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004981290783163867,
      "loss": 5.7438,
      "mean_token_accuracy": 0.17086322605609894,
      "num_tokens": 11479579.0,
      "step": 5660
    },
    {
      "entropy": 6.02326135635376,
      "epoch": 0.4950624836144368,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004981250660359216,
      "loss": 5.8257,
      "mean_token_accuracy": 0.16383726745843888,
      "num_tokens": 11489202.0,
      "step": 5665
    },
    {
      "entropy": 5.9639606952667235,
      "epoch": 0.49549943196714147,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004981210494758007,
      "loss": 5.8543,
      "mean_token_accuracy": 0.15504157543182373,
      "num_tokens": 11499425.0,
      "step": 5670
    },
    {
      "entropy": 6.047420835494995,
      "epoch": 0.4959363803198462,
      "grad_norm": 1.0,
      "learning_rate": 0.000498117028636101,
      "loss": 5.8268,
      "mean_token_accuracy": 0.16233543455600738,
      "num_tokens": 11508983.0,
      "step": 5675
    },
    {
      "entropy": 6.032330703735352,
      "epoch": 0.4963733286725509,
      "grad_norm": 0.875,
      "learning_rate": 0.0004981130035169,
      "loss": 5.8713,
      "mean_token_accuracy": 0.16642006635665893,
      "num_tokens": 11519791.0,
      "step": 5680
    },
    {
      "entropy": 6.007494974136352,
      "epoch": 0.4968102770252556,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004981089741182747,
      "loss": 5.7978,
      "mean_token_accuracy": 0.16401655375957488,
      "num_tokens": 11530537.0,
      "step": 5685
    },
    {
      "entropy": 6.02313346862793,
      "epoch": 0.49724722537796034,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004981049404403023,
      "loss": 5.8067,
      "mean_token_accuracy": 0.16553872525691987,
      "num_tokens": 11540591.0,
      "step": 5690
    },
    {
      "entropy": 5.976443481445313,
      "epoch": 0.49768417373066504,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004981009024830602,
      "loss": 5.7499,
      "mean_token_accuracy": 0.1701885774731636,
      "num_tokens": 11550719.0,
      "step": 5695
    },
    {
      "entropy": 5.960111045837403,
      "epoch": 0.4981211220833697,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000498096860246626,
      "loss": 5.8119,
      "mean_token_accuracy": 0.16723384857177734,
      "num_tokens": 11561805.0,
      "step": 5700
    },
    {
      "entropy": 6.1308740139007565,
      "epoch": 0.4985580704360745,
      "grad_norm": 0.9375,
      "learning_rate": 0.000498092813731077,
      "loss": 5.8255,
      "mean_token_accuracy": 0.1577245056629181,
      "num_tokens": 11571380.0,
      "step": 5705
    },
    {
      "entropy": 5.985809421539306,
      "epoch": 0.49899501878877917,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000498088762936491,
      "loss": 5.816,
      "mean_token_accuracy": 0.1598203718662262,
      "num_tokens": 11580853.0,
      "step": 5710
    },
    {
      "entropy": 5.8938737392425535,
      "epoch": 0.49943196714148386,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004980847078629457,
      "loss": 5.7191,
      "mean_token_accuracy": 0.17613861560821534,
      "num_tokens": 11591966.0,
      "step": 5715
    },
    {
      "entropy": 6.013378667831421,
      "epoch": 0.4998689154941886,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004980806485105186,
      "loss": 5.7772,
      "mean_token_accuracy": 0.16970082744956017,
      "num_tokens": 11601239.0,
      "step": 5720
    },
    {
      "entropy": 6.021574783325195,
      "epoch": 0.5003058638468934,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004980765848792879,
      "loss": 5.705,
      "mean_token_accuracy": 0.17401468604803086,
      "num_tokens": 11609857.0,
      "step": 5725
    },
    {
      "entropy": 6.02330961227417,
      "epoch": 0.500742812199598,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004980725169693314,
      "loss": 5.7692,
      "mean_token_accuracy": 0.16660242825746535,
      "num_tokens": 11618952.0,
      "step": 5730
    },
    {
      "entropy": 6.010266590118408,
      "epoch": 0.5011797605523027,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004980684447807272,
      "loss": 5.6935,
      "mean_token_accuracy": 0.17109654247760772,
      "num_tokens": 11627638.0,
      "step": 5735
    },
    {
      "entropy": 5.9998329162597654,
      "epoch": 0.5016167089050074,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004980643683135533,
      "loss": 5.813,
      "mean_token_accuracy": 0.16527167707681656,
      "num_tokens": 11637842.0,
      "step": 5740
    },
    {
      "entropy": 5.97807559967041,
      "epoch": 0.5020536572577121,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004980602875678878,
      "loss": 5.7308,
      "mean_token_accuracy": 0.1756144255399704,
      "num_tokens": 11647217.0,
      "step": 5745
    },
    {
      "entropy": 5.922547435760498,
      "epoch": 0.5024906056104168,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004980562025438093,
      "loss": 5.7189,
      "mean_token_accuracy": 0.16963032633066177,
      "num_tokens": 11657558.0,
      "step": 5750
    },
    {
      "entropy": 5.993762111663818,
      "epoch": 0.5029275539631216,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004980521132413958,
      "loss": 5.7102,
      "mean_token_accuracy": 0.16456425487995147,
      "num_tokens": 11668054.0,
      "step": 5755
    },
    {
      "entropy": 6.022767782211304,
      "epoch": 0.5033645023158263,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000498048019660726,
      "loss": 5.7478,
      "mean_token_accuracy": 0.16621067076921464,
      "num_tokens": 11677779.0,
      "step": 5760
    },
    {
      "entropy": 5.984483480453491,
      "epoch": 0.503801450668531,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004980439218018783,
      "loss": 5.7922,
      "mean_token_accuracy": 0.16176709830760955,
      "num_tokens": 11687577.0,
      "step": 5765
    },
    {
      "entropy": 6.115398693084717,
      "epoch": 0.5042383990212357,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004980398196649312,
      "loss": 5.8941,
      "mean_token_accuracy": 0.15421511828899384,
      "num_tokens": 11699630.0,
      "step": 5770
    },
    {
      "entropy": 6.0269455909729,
      "epoch": 0.5046753473739404,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004980357132499636,
      "loss": 5.827,
      "mean_token_accuracy": 0.16120329201221467,
      "num_tokens": 11710139.0,
      "step": 5775
    },
    {
      "entropy": 6.037773275375367,
      "epoch": 0.5051122957266451,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004980316025570542,
      "loss": 5.7921,
      "mean_token_accuracy": 0.16295965909957885,
      "num_tokens": 11719070.0,
      "step": 5780
    },
    {
      "entropy": 5.978439092636108,
      "epoch": 0.5055492440793499,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004980274875862816,
      "loss": 5.8152,
      "mean_token_accuracy": 0.165883868932724,
      "num_tokens": 11728770.0,
      "step": 5785
    },
    {
      "entropy": 6.0388671398162845,
      "epoch": 0.5059861924320546,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000498023368337725,
      "loss": 5.7555,
      "mean_token_accuracy": 0.16983137428760528,
      "num_tokens": 11737753.0,
      "step": 5790
    },
    {
      "entropy": 5.9694465637207035,
      "epoch": 0.5064231407847593,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004980192448114635,
      "loss": 5.74,
      "mean_token_accuracy": 0.16601070612668992,
      "num_tokens": 11747837.0,
      "step": 5795
    },
    {
      "entropy": 5.977008104324341,
      "epoch": 0.5068600891374639,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004980151170075758,
      "loss": 5.8607,
      "mean_token_accuracy": 0.16577140539884566,
      "num_tokens": 11757911.0,
      "step": 5800
    },
    {
      "entropy": 6.070715713500976,
      "epoch": 0.5072970374901686,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004980109849261414,
      "loss": 5.7336,
      "mean_token_accuracy": 0.1649346336722374,
      "num_tokens": 11766213.0,
      "step": 5805
    },
    {
      "entropy": 5.971834135055542,
      "epoch": 0.5077339858428733,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004980068485672394,
      "loss": 5.6836,
      "mean_token_accuracy": 0.17372626066207886,
      "num_tokens": 11775640.0,
      "step": 5810
    },
    {
      "entropy": 6.012606239318847,
      "epoch": 0.5081709341955781,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004980027079309493,
      "loss": 5.7271,
      "mean_token_accuracy": 0.17473618388175965,
      "num_tokens": 11785664.0,
      "step": 5815
    },
    {
      "entropy": 6.0354461669921875,
      "epoch": 0.5086078825482828,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004979985630173503,
      "loss": 5.8481,
      "mean_token_accuracy": 0.1585448458790779,
      "num_tokens": 11795402.0,
      "step": 5820
    },
    {
      "entropy": 5.950788307189941,
      "epoch": 0.5090448309009875,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004979944138265221,
      "loss": 5.7048,
      "mean_token_accuracy": 0.16548294425010682,
      "num_tokens": 11805923.0,
      "step": 5825
    },
    {
      "entropy": 5.9886155128479,
      "epoch": 0.5094817792536922,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004979902603585442,
      "loss": 5.8142,
      "mean_token_accuracy": 0.1598925217986107,
      "num_tokens": 11816097.0,
      "step": 5830
    },
    {
      "entropy": 6.005649852752685,
      "epoch": 0.5099187276063969,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004979861026134961,
      "loss": 5.7296,
      "mean_token_accuracy": 0.16915328949689865,
      "num_tokens": 11826889.0,
      "step": 5835
    },
    {
      "entropy": 6.01967487335205,
      "epoch": 0.5103556759591016,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004979819405914579,
      "loss": 5.7935,
      "mean_token_accuracy": 0.16395938098430635,
      "num_tokens": 11837935.0,
      "step": 5840
    },
    {
      "entropy": 5.996092700958252,
      "epoch": 0.5107926243118064,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004979777742925092,
      "loss": 5.7841,
      "mean_token_accuracy": 0.16417268365621568,
      "num_tokens": 11847715.0,
      "step": 5845
    },
    {
      "entropy": 6.059479188919068,
      "epoch": 0.5112295726645111,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004979736037167298,
      "loss": 5.8055,
      "mean_token_accuracy": 0.16161997616291046,
      "num_tokens": 11857841.0,
      "step": 5850
    },
    {
      "entropy": 6.0096320629119875,
      "epoch": 0.5116665210172158,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004979694288642,
      "loss": 5.8056,
      "mean_token_accuracy": 0.169195756316185,
      "num_tokens": 11868135.0,
      "step": 5855
    },
    {
      "entropy": 5.994083452224731,
      "epoch": 0.5121034693699205,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004979652497349997,
      "loss": 5.725,
      "mean_token_accuracy": 0.16251475363969803,
      "num_tokens": 11877637.0,
      "step": 5860
    },
    {
      "entropy": 5.909412384033203,
      "epoch": 0.5125404177226252,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000497961066329209,
      "loss": 5.7969,
      "mean_token_accuracy": 0.16380739212036133,
      "num_tokens": 11887128.0,
      "step": 5865
    },
    {
      "entropy": 6.075037622451783,
      "epoch": 0.5129773660753298,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004979568786469083,
      "loss": 5.8141,
      "mean_token_accuracy": 0.1570396229624748,
      "num_tokens": 11897885.0,
      "step": 5870
    },
    {
      "entropy": 5.9767515659332275,
      "epoch": 0.5134143144280346,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004979526866881777,
      "loss": 5.7509,
      "mean_token_accuracy": 0.1694442570209503,
      "num_tokens": 11907919.0,
      "step": 5875
    },
    {
      "entropy": 5.898437595367431,
      "epoch": 0.5138512627807393,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000497948490453098,
      "loss": 5.7241,
      "mean_token_accuracy": 0.171146397292614,
      "num_tokens": 11918612.0,
      "step": 5880
    },
    {
      "entropy": 5.944434452056885,
      "epoch": 0.514288211133444,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004979442899417493,
      "loss": 5.7895,
      "mean_token_accuracy": 0.1660420849919319,
      "num_tokens": 11927759.0,
      "step": 5885
    },
    {
      "entropy": 6.031778621673584,
      "epoch": 0.5147251594861487,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004979400851542122,
      "loss": 5.7445,
      "mean_token_accuracy": 0.1709442839026451,
      "num_tokens": 11937522.0,
      "step": 5890
    },
    {
      "entropy": 5.930755472183227,
      "epoch": 0.5151621078388534,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004979358760905675,
      "loss": 5.7375,
      "mean_token_accuracy": 0.16642869561910628,
      "num_tokens": 11947787.0,
      "step": 5895
    },
    {
      "entropy": 5.957692050933838,
      "epoch": 0.5155990561915581,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000497931662750896,
      "loss": 5.7815,
      "mean_token_accuracy": 0.16530563682317734,
      "num_tokens": 11958929.0,
      "step": 5900
    },
    {
      "entropy": 5.989711761474609,
      "epoch": 0.5160360045442629,
      "grad_norm": 1.0,
      "learning_rate": 0.0004979274451352783,
      "loss": 5.7245,
      "mean_token_accuracy": 0.1655576378107071,
      "num_tokens": 11969242.0,
      "step": 5905
    },
    {
      "entropy": 6.0093732357025145,
      "epoch": 0.5164729528969676,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004979232232437954,
      "loss": 5.7893,
      "mean_token_accuracy": 0.16588158458471297,
      "num_tokens": 11980149.0,
      "step": 5910
    },
    {
      "entropy": 6.018730211257934,
      "epoch": 0.5169099012496723,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004979189970765282,
      "loss": 5.7867,
      "mean_token_accuracy": 0.15827691107988356,
      "num_tokens": 11990636.0,
      "step": 5915
    },
    {
      "entropy": 6.0136371612548825,
      "epoch": 0.517346849602377,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004979147666335579,
      "loss": 5.7873,
      "mean_token_accuracy": 0.16637275069952012,
      "num_tokens": 12000615.0,
      "step": 5920
    },
    {
      "entropy": 5.960154342651367,
      "epoch": 0.5177837979550817,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004979105319149656,
      "loss": 5.8209,
      "mean_token_accuracy": 0.16430925726890563,
      "num_tokens": 12011072.0,
      "step": 5925
    },
    {
      "entropy": 5.926241779327393,
      "epoch": 0.5182207463077865,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004979062929208324,
      "loss": 5.7217,
      "mean_token_accuracy": 0.16912839114665984,
      "num_tokens": 12021726.0,
      "step": 5930
    },
    {
      "entropy": 6.048215723037719,
      "epoch": 0.5186576946604912,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004979020496512398,
      "loss": 5.8524,
      "mean_token_accuracy": 0.15961608439683914,
      "num_tokens": 12031353.0,
      "step": 5935
    },
    {
      "entropy": 6.028223180770874,
      "epoch": 0.5190946430131959,
      "grad_norm": 0.9375,
      "learning_rate": 0.000497897802106269,
      "loss": 5.8001,
      "mean_token_accuracy": 0.15659957528114318,
      "num_tokens": 12041234.0,
      "step": 5940
    },
    {
      "entropy": 5.897720575332642,
      "epoch": 0.5195315913659005,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004978935502860017,
      "loss": 5.7419,
      "mean_token_accuracy": 0.16626310497522354,
      "num_tokens": 12051100.0,
      "step": 5945
    },
    {
      "entropy": 6.006535768508911,
      "epoch": 0.5199685397186052,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004978892941905191,
      "loss": 5.7788,
      "mean_token_accuracy": 0.167525415122509,
      "num_tokens": 12062012.0,
      "step": 5950
    },
    {
      "entropy": 6.064031553268433,
      "epoch": 0.5204054880713099,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004978850338199033,
      "loss": 5.817,
      "mean_token_accuracy": 0.1626892387866974,
      "num_tokens": 12072763.0,
      "step": 5955
    },
    {
      "entropy": 5.957067680358887,
      "epoch": 0.5208424364240147,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004978807691742356,
      "loss": 5.6711,
      "mean_token_accuracy": 0.1723813995718956,
      "num_tokens": 12081892.0,
      "step": 5960
    },
    {
      "entropy": 5.986256551742554,
      "epoch": 0.5212793847767194,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000497876500253598,
      "loss": 5.7524,
      "mean_token_accuracy": 0.16899172365665435,
      "num_tokens": 12091428.0,
      "step": 5965
    },
    {
      "entropy": 5.939373922348023,
      "epoch": 0.5217163331294241,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004978722270580722,
      "loss": 5.7485,
      "mean_token_accuracy": 0.16559201925992967,
      "num_tokens": 12101687.0,
      "step": 5970
    },
    {
      "entropy": 5.890044689178467,
      "epoch": 0.5221532814821288,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004978679495877405,
      "loss": 5.7066,
      "mean_token_accuracy": 0.16988663226366044,
      "num_tokens": 12111726.0,
      "step": 5975
    },
    {
      "entropy": 6.032839250564575,
      "epoch": 0.5225902298348335,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004978636678426847,
      "loss": 5.7241,
      "mean_token_accuracy": 0.1752467632293701,
      "num_tokens": 12122171.0,
      "step": 5980
    },
    {
      "entropy": 6.0335760593414305,
      "epoch": 0.5230271781875382,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004978593818229869,
      "loss": 5.7715,
      "mean_token_accuracy": 0.16260040998458863,
      "num_tokens": 12131414.0,
      "step": 5985
    },
    {
      "entropy": 5.922642612457276,
      "epoch": 0.523464126540243,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004978550915287294,
      "loss": 5.741,
      "mean_token_accuracy": 0.16121231764554977,
      "num_tokens": 12141665.0,
      "step": 5990
    },
    {
      "entropy": 6.030305671691894,
      "epoch": 0.5239010748929477,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004978507969599945,
      "loss": 5.806,
      "mean_token_accuracy": 0.16103749722242355,
      "num_tokens": 12151263.0,
      "step": 5995
    },
    {
      "entropy": 6.065587520599365,
      "epoch": 0.5243380232456524,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004978464981168645,
      "loss": 5.7856,
      "mean_token_accuracy": 0.163425475358963,
      "num_tokens": 12161070.0,
      "step": 6000
    },
    {
      "epoch": 0.5243380232456524,
      "eval_entropy": 5.752074069484202,
      "eval_loss": 5.778484344482422,
      "eval_mean_token_accuracy": 0.17260141236261128,
      "eval_num_tokens": 12161070.0,
      "eval_runtime": 21.5329,
      "eval_samples_per_second": 1718.951,
      "eval_steps_per_second": 214.881,
      "step": 6000
    },
    {
      "entropy": 5.974909782409668,
      "epoch": 0.5247749715983571,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000497842194999422,
      "loss": 5.7616,
      "mean_token_accuracy": 0.16672465205192566,
      "num_tokens": 12171197.0,
      "step": 6005
    },
    {
      "entropy": 5.973691844940186,
      "epoch": 0.5252119199510618,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004978378876077493,
      "loss": 5.7756,
      "mean_token_accuracy": 0.16448745578527452,
      "num_tokens": 12181462.0,
      "step": 6010
    },
    {
      "entropy": 5.974299335479737,
      "epoch": 0.5256488683037664,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004978335759419293,
      "loss": 5.7632,
      "mean_token_accuracy": 0.16959105134010316,
      "num_tokens": 12191473.0,
      "step": 6015
    },
    {
      "entropy": 6.015347814559936,
      "epoch": 0.5260858166564713,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004978292600020444,
      "loss": 5.7075,
      "mean_token_accuracy": 0.1727922573685646,
      "num_tokens": 12200920.0,
      "step": 6020
    },
    {
      "entropy": 6.005795431137085,
      "epoch": 0.5265227650091759,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004978249397881776,
      "loss": 5.7814,
      "mean_token_accuracy": 0.16636140793561935,
      "num_tokens": 12210823.0,
      "step": 6025
    },
    {
      "entropy": 5.995762586593628,
      "epoch": 0.5269597133618806,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004978206153004117,
      "loss": 5.7932,
      "mean_token_accuracy": 0.16732889264822007,
      "num_tokens": 12221967.0,
      "step": 6030
    },
    {
      "entropy": 6.000564813613892,
      "epoch": 0.5273966617145853,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004978162865388297,
      "loss": 5.7554,
      "mean_token_accuracy": 0.16809408664703368,
      "num_tokens": 12232519.0,
      "step": 6035
    },
    {
      "entropy": 6.010684633255005,
      "epoch": 0.52783361006729,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004978119535035144,
      "loss": 5.7316,
      "mean_token_accuracy": 0.16341073662042618,
      "num_tokens": 12243193.0,
      "step": 6040
    },
    {
      "entropy": 6.026944923400879,
      "epoch": 0.5282705584199947,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004978076161945492,
      "loss": 5.7403,
      "mean_token_accuracy": 0.16596628874540328,
      "num_tokens": 12253640.0,
      "step": 6045
    },
    {
      "entropy": 6.085190963745117,
      "epoch": 0.5287075067726995,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004978032746120173,
      "loss": 5.8587,
      "mean_token_accuracy": 0.15808591544628142,
      "num_tokens": 12265072.0,
      "step": 6050
    },
    {
      "entropy": 6.01777834892273,
      "epoch": 0.5291444551254042,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004977989287560016,
      "loss": 5.7826,
      "mean_token_accuracy": 0.16498706340789795,
      "num_tokens": 12274449.0,
      "step": 6055
    },
    {
      "entropy": 6.044222211837768,
      "epoch": 0.5295814034781089,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004977945786265857,
      "loss": 5.8021,
      "mean_token_accuracy": 0.15623485445976257,
      "num_tokens": 12284910.0,
      "step": 6060
    },
    {
      "entropy": 6.054895067214966,
      "epoch": 0.5300183518308136,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004977902242238533,
      "loss": 5.8099,
      "mean_token_accuracy": 0.16117410659790038,
      "num_tokens": 12294046.0,
      "step": 6065
    },
    {
      "entropy": 5.953406906127929,
      "epoch": 0.5304553001835183,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004977858655478874,
      "loss": 5.7964,
      "mean_token_accuracy": 0.16127994209527968,
      "num_tokens": 12303900.0,
      "step": 6070
    },
    {
      "entropy": 6.004986047744751,
      "epoch": 0.530892248536223,
      "grad_norm": 1.046875,
      "learning_rate": 0.000497781502598772,
      "loss": 5.7323,
      "mean_token_accuracy": 0.16589521020650863,
      "num_tokens": 12313326.0,
      "step": 6075
    },
    {
      "entropy": 5.967529296875,
      "epoch": 0.5313291968889278,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004977771353765906,
      "loss": 5.7308,
      "mean_token_accuracy": 0.16263570487499238,
      "num_tokens": 12322792.0,
      "step": 6080
    },
    {
      "entropy": 6.049457740783692,
      "epoch": 0.5317661452416325,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000497772763881427,
      "loss": 5.836,
      "mean_token_accuracy": 0.15844803303480148,
      "num_tokens": 12333417.0,
      "step": 6085
    },
    {
      "entropy": 6.0368751049041744,
      "epoch": 0.5322030935943372,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004977683881133651,
      "loss": 5.7356,
      "mean_token_accuracy": 0.16882071048021316,
      "num_tokens": 12343944.0,
      "step": 6090
    },
    {
      "entropy": 5.969963932037354,
      "epoch": 0.5326400419470418,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004977640080724888,
      "loss": 5.7479,
      "mean_token_accuracy": 0.1624076709151268,
      "num_tokens": 12353838.0,
      "step": 6095
    },
    {
      "entropy": 6.0397275447845455,
      "epoch": 0.5330769902997465,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004977596237588821,
      "loss": 5.8113,
      "mean_token_accuracy": 0.16100122779607773,
      "num_tokens": 12364996.0,
      "step": 6100
    },
    {
      "entropy": 5.983057260513306,
      "epoch": 0.5335139386524512,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000497755235172629,
      "loss": 5.8069,
      "mean_token_accuracy": 0.15979383140802383,
      "num_tokens": 12375111.0,
      "step": 6105
    },
    {
      "entropy": 5.940854549407959,
      "epoch": 0.533950887005156,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004977508423138139,
      "loss": 5.7669,
      "mean_token_accuracy": 0.1622542768716812,
      "num_tokens": 12384376.0,
      "step": 6110
    },
    {
      "entropy": 6.021222448348999,
      "epoch": 0.5343878353578607,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004977464451825208,
      "loss": 5.8169,
      "mean_token_accuracy": 0.15347349345684053,
      "num_tokens": 12395248.0,
      "step": 6115
    },
    {
      "entropy": 6.064636135101319,
      "epoch": 0.5348247837105654,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004977420437788341,
      "loss": 5.8062,
      "mean_token_accuracy": 0.15986451208591462,
      "num_tokens": 12405083.0,
      "step": 6120
    },
    {
      "entropy": 5.898163461685181,
      "epoch": 0.5352617320632701,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004977376381028385,
      "loss": 5.7053,
      "mean_token_accuracy": 0.17480639070272447,
      "num_tokens": 12415787.0,
      "step": 6125
    },
    {
      "entropy": 5.9801506996154785,
      "epoch": 0.5356986804159748,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004977332281546181,
      "loss": 5.7596,
      "mean_token_accuracy": 0.1672113135457039,
      "num_tokens": 12425911.0,
      "step": 6130
    },
    {
      "entropy": 6.015285921096802,
      "epoch": 0.5361356287686796,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004977288139342578,
      "loss": 5.7852,
      "mean_token_accuracy": 0.16462033092975617,
      "num_tokens": 12437270.0,
      "step": 6135
    },
    {
      "entropy": 6.003399229049682,
      "epoch": 0.5365725771213843,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004977243954418422,
      "loss": 5.742,
      "mean_token_accuracy": 0.16710813641548156,
      "num_tokens": 12446778.0,
      "step": 6140
    },
    {
      "entropy": 5.902034568786621,
      "epoch": 0.537009525474089,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004977199726774558,
      "loss": 5.7052,
      "mean_token_accuracy": 0.16986558884382247,
      "num_tokens": 12456550.0,
      "step": 6145
    },
    {
      "entropy": 5.995639228820801,
      "epoch": 0.5374464738267937,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004977155456411837,
      "loss": 5.7,
      "mean_token_accuracy": 0.16762072443962098,
      "num_tokens": 12465562.0,
      "step": 6150
    },
    {
      "entropy": 6.040780735015869,
      "epoch": 0.5378834221794984,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004977111143331108,
      "loss": 5.7724,
      "mean_token_accuracy": 0.161351802945137,
      "num_tokens": 12475277.0,
      "step": 6155
    },
    {
      "entropy": 5.968061065673828,
      "epoch": 0.538320370532203,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004977066787533219,
      "loss": 5.7144,
      "mean_token_accuracy": 0.16704001426696777,
      "num_tokens": 12485478.0,
      "step": 6160
    },
    {
      "entropy": 6.013670539855957,
      "epoch": 0.5387573188849079,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004977022389019023,
      "loss": 5.8285,
      "mean_token_accuracy": 0.15940337404608726,
      "num_tokens": 12495714.0,
      "step": 6165
    },
    {
      "entropy": 6.083322763442993,
      "epoch": 0.5391942672376125,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004976977947789371,
      "loss": 5.7392,
      "mean_token_accuracy": 0.1704345867037773,
      "num_tokens": 12505125.0,
      "step": 6170
    },
    {
      "entropy": 5.9623583316802975,
      "epoch": 0.5396312155903172,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004976933463845115,
      "loss": 5.7463,
      "mean_token_accuracy": 0.16843317002058028,
      "num_tokens": 12515051.0,
      "step": 6175
    },
    {
      "entropy": 5.961470985412598,
      "epoch": 0.5400681639430219,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004976888937187107,
      "loss": 5.6926,
      "mean_token_accuracy": 0.1747165709733963,
      "num_tokens": 12525389.0,
      "step": 6180
    },
    {
      "entropy": 5.999250841140747,
      "epoch": 0.5405051122957266,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004976844367816203,
      "loss": 5.8098,
      "mean_token_accuracy": 0.1527354672551155,
      "num_tokens": 12535454.0,
      "step": 6185
    },
    {
      "entropy": 5.957386016845703,
      "epoch": 0.5409420606484313,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004976799755733257,
      "loss": 5.6413,
      "mean_token_accuracy": 0.1719467520713806,
      "num_tokens": 12546092.0,
      "step": 6190
    },
    {
      "entropy": 5.964174890518189,
      "epoch": 0.5413790090011361,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004976755100939126,
      "loss": 5.7402,
      "mean_token_accuracy": 0.16424873620271682,
      "num_tokens": 12555462.0,
      "step": 6195
    },
    {
      "entropy": 6.0099259376525875,
      "epoch": 0.5418159573538408,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004976710403434663,
      "loss": 5.784,
      "mean_token_accuracy": 0.16797115057706832,
      "num_tokens": 12564506.0,
      "step": 6200
    },
    {
      "entropy": 5.9581639766693115,
      "epoch": 0.5422529057065455,
      "grad_norm": 0.921875,
      "learning_rate": 0.000497666566322073,
      "loss": 5.7564,
      "mean_token_accuracy": 0.16673678308725357,
      "num_tokens": 12574887.0,
      "step": 6205
    },
    {
      "entropy": 6.057765007019043,
      "epoch": 0.5426898540592502,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004976620880298182,
      "loss": 5.7721,
      "mean_token_accuracy": 0.16573724150657654,
      "num_tokens": 12584499.0,
      "step": 6210
    },
    {
      "entropy": 5.962661933898926,
      "epoch": 0.5431268024119549,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004976576054667878,
      "loss": 5.7177,
      "mean_token_accuracy": 0.1664577066898346,
      "num_tokens": 12593796.0,
      "step": 6215
    },
    {
      "entropy": 5.938970899581909,
      "epoch": 0.5435637507646596,
      "grad_norm": 1.0,
      "learning_rate": 0.000497653118633068,
      "loss": 5.7752,
      "mean_token_accuracy": 0.16016299426555633,
      "num_tokens": 12603719.0,
      "step": 6220
    },
    {
      "entropy": 5.944729328155518,
      "epoch": 0.5440006991173644,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004976486275287446,
      "loss": 5.7656,
      "mean_token_accuracy": 0.16738038659095764,
      "num_tokens": 12614407.0,
      "step": 6225
    },
    {
      "entropy": 6.016261053085327,
      "epoch": 0.5444376474700691,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004976441321539038,
      "loss": 5.7455,
      "mean_token_accuracy": 0.1729540765285492,
      "num_tokens": 12624414.0,
      "step": 6230
    },
    {
      "entropy": 5.969081115722656,
      "epoch": 0.5448745958227738,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004976396325086321,
      "loss": 5.7647,
      "mean_token_accuracy": 0.16788340210914612,
      "num_tokens": 12635679.0,
      "step": 6235
    },
    {
      "entropy": 5.993377923965454,
      "epoch": 0.5453115441754784,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004976351285930154,
      "loss": 5.7084,
      "mean_token_accuracy": 0.1628316432237625,
      "num_tokens": 12645253.0,
      "step": 6240
    },
    {
      "entropy": 5.883375453948974,
      "epoch": 0.5457484925281831,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004976306204071404,
      "loss": 5.666,
      "mean_token_accuracy": 0.18156086206436156,
      "num_tokens": 12655815.0,
      "step": 6245
    },
    {
      "entropy": 5.948963737487793,
      "epoch": 0.5461854408808878,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004976261079510934,
      "loss": 5.7589,
      "mean_token_accuracy": 0.1667147010564804,
      "num_tokens": 12666173.0,
      "step": 6250
    },
    {
      "entropy": 5.975498628616333,
      "epoch": 0.5466223892335926,
      "grad_norm": 0.9375,
      "learning_rate": 0.000497621591224961,
      "loss": 5.7036,
      "mean_token_accuracy": 0.17155962735414504,
      "num_tokens": 12676008.0,
      "step": 6255
    },
    {
      "entropy": 5.995073509216309,
      "epoch": 0.5470593375862973,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004976170702288298,
      "loss": 5.7527,
      "mean_token_accuracy": 0.16572992503643036,
      "num_tokens": 12686217.0,
      "step": 6260
    },
    {
      "entropy": 5.976833963394165,
      "epoch": 0.547496285939002,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004976125449627866,
      "loss": 5.7685,
      "mean_token_accuracy": 0.16199978739023208,
      "num_tokens": 12696809.0,
      "step": 6265
    },
    {
      "entropy": 6.048425579071045,
      "epoch": 0.5479332342917067,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004976080154269181,
      "loss": 5.8493,
      "mean_token_accuracy": 0.156369149684906,
      "num_tokens": 12708328.0,
      "step": 6270
    },
    {
      "entropy": 5.983789443969727,
      "epoch": 0.5483701826444114,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004976034816213112,
      "loss": 5.656,
      "mean_token_accuracy": 0.16893780529499053,
      "num_tokens": 12718478.0,
      "step": 6275
    },
    {
      "entropy": 5.875085735321045,
      "epoch": 0.5488071309971161,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004975989435460528,
      "loss": 5.7521,
      "mean_token_accuracy": 0.16918388605117798,
      "num_tokens": 12728853.0,
      "step": 6280
    },
    {
      "entropy": 5.966558456420898,
      "epoch": 0.5492440793498209,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004975944012012302,
      "loss": 5.8114,
      "mean_token_accuracy": 0.16706595420837403,
      "num_tokens": 12739036.0,
      "step": 6285
    },
    {
      "entropy": 5.9536333084106445,
      "epoch": 0.5496810277025256,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004975898545869303,
      "loss": 5.7053,
      "mean_token_accuracy": 0.1725954070687294,
      "num_tokens": 12748908.0,
      "step": 6290
    },
    {
      "entropy": 5.926786184310913,
      "epoch": 0.5501179760552303,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004975853037032403,
      "loss": 5.7758,
      "mean_token_accuracy": 0.16151309609413148,
      "num_tokens": 12758671.0,
      "step": 6295
    },
    {
      "entropy": 5.9707073211669925,
      "epoch": 0.550554924407935,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004975807485502476,
      "loss": 5.7446,
      "mean_token_accuracy": 0.1654016062617302,
      "num_tokens": 12769552.0,
      "step": 6300
    },
    {
      "entropy": 6.055487442016601,
      "epoch": 0.5509918727606397,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004975761891280394,
      "loss": 5.8852,
      "mean_token_accuracy": 0.15503082424402237,
      "num_tokens": 12781032.0,
      "step": 6305
    },
    {
      "entropy": 6.068560409545898,
      "epoch": 0.5514288211133443,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004975716254367033,
      "loss": 5.8075,
      "mean_token_accuracy": 0.1635914146900177,
      "num_tokens": 12790818.0,
      "step": 6310
    },
    {
      "entropy": 5.88771014213562,
      "epoch": 0.5518657694660491,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000497567057476327,
      "loss": 5.6933,
      "mean_token_accuracy": 0.17007147073745726,
      "num_tokens": 12799701.0,
      "step": 6315
    },
    {
      "entropy": 5.973051309585571,
      "epoch": 0.5523027178187538,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004975624852469976,
      "loss": 5.8373,
      "mean_token_accuracy": 0.16561338901519776,
      "num_tokens": 12809575.0,
      "step": 6320
    },
    {
      "entropy": 6.0136439323425295,
      "epoch": 0.5527396661714585,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004975579087488033,
      "loss": 5.7239,
      "mean_token_accuracy": 0.17029444575309755,
      "num_tokens": 12820278.0,
      "step": 6325
    },
    {
      "entropy": 6.039294481277466,
      "epoch": 0.5531766145241632,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004975533279818317,
      "loss": 5.7597,
      "mean_token_accuracy": 0.16447666883468628,
      "num_tokens": 12830774.0,
      "step": 6330
    },
    {
      "entropy": 5.9324737071990965,
      "epoch": 0.5536135628768679,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004975487429461706,
      "loss": 5.7226,
      "mean_token_accuracy": 0.17386601865291595,
      "num_tokens": 12839487.0,
      "step": 6335
    },
    {
      "entropy": 5.91487684249878,
      "epoch": 0.5540505112295727,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000497544153641908,
      "loss": 5.6581,
      "mean_token_accuracy": 0.16970890909433364,
      "num_tokens": 12849614.0,
      "step": 6340
    },
    {
      "entropy": 5.985746192932129,
      "epoch": 0.5544874595822774,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004975395600691319,
      "loss": 5.7742,
      "mean_token_accuracy": 0.16493808329105378,
      "num_tokens": 12859428.0,
      "step": 6345
    },
    {
      "entropy": 5.964206218719482,
      "epoch": 0.5549244079349821,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004975349622279305,
      "loss": 5.7864,
      "mean_token_accuracy": 0.16574744284152984,
      "num_tokens": 12869167.0,
      "step": 6350
    },
    {
      "entropy": 5.96634464263916,
      "epoch": 0.5553613562876868,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004975303601183919,
      "loss": 5.7137,
      "mean_token_accuracy": 0.16944707930088043,
      "num_tokens": 12878410.0,
      "step": 6355
    },
    {
      "entropy": 5.988892889022827,
      "epoch": 0.5557983046403915,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004975257537406043,
      "loss": 5.7468,
      "mean_token_accuracy": 0.16139499247074127,
      "num_tokens": 12888372.0,
      "step": 6360
    },
    {
      "entropy": 6.000051879882813,
      "epoch": 0.5562352529930962,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004975211430946563,
      "loss": 5.7836,
      "mean_token_accuracy": 0.16662488430738448,
      "num_tokens": 12897958.0,
      "step": 6365
    },
    {
      "entropy": 5.981447029113769,
      "epoch": 0.556672201345801,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000497516528180636,
      "loss": 5.7764,
      "mean_token_accuracy": 0.1681187391281128,
      "num_tokens": 12907890.0,
      "step": 6370
    },
    {
      "entropy": 5.99637541770935,
      "epoch": 0.5571091496985057,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004975119089986322,
      "loss": 5.7189,
      "mean_token_accuracy": 0.164870785176754,
      "num_tokens": 12917914.0,
      "step": 6375
    },
    {
      "entropy": 5.999019908905029,
      "epoch": 0.5575460980512104,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004975072855487333,
      "loss": 5.7569,
      "mean_token_accuracy": 0.16910712867975236,
      "num_tokens": 12927090.0,
      "step": 6380
    },
    {
      "entropy": 5.971538877487182,
      "epoch": 0.557983046403915,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004975026578310282,
      "loss": 5.7453,
      "mean_token_accuracy": 0.16372237652540206,
      "num_tokens": 12937236.0,
      "step": 6385
    },
    {
      "entropy": 5.945543479919434,
      "epoch": 0.5584199947566197,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004974980258456055,
      "loss": 5.6996,
      "mean_token_accuracy": 0.16848786175251007,
      "num_tokens": 12947635.0,
      "step": 6390
    },
    {
      "entropy": 6.026625108718872,
      "epoch": 0.5588569431093244,
      "grad_norm": 0.96875,
      "learning_rate": 0.000497493389592554,
      "loss": 5.6518,
      "mean_token_accuracy": 0.1690240368247032,
      "num_tokens": 12957236.0,
      "step": 6395
    },
    {
      "entropy": 6.0162303924560545,
      "epoch": 0.5592938914620292,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004974887490719627,
      "loss": 5.7756,
      "mean_token_accuracy": 0.1611407682299614,
      "num_tokens": 12967497.0,
      "step": 6400
    },
    {
      "entropy": 5.898978519439697,
      "epoch": 0.5597308398147339,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004974841042839207,
      "loss": 5.7006,
      "mean_token_accuracy": 0.17128993272781373,
      "num_tokens": 12977467.0,
      "step": 6405
    },
    {
      "entropy": 5.993454933166504,
      "epoch": 0.5601677881674386,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000497479455228517,
      "loss": 5.765,
      "mean_token_accuracy": 0.1637665346264839,
      "num_tokens": 12987867.0,
      "step": 6410
    },
    {
      "entropy": 5.991722774505615,
      "epoch": 0.5606047365201433,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004974748019058407,
      "loss": 5.7627,
      "mean_token_accuracy": 0.16298171430826186,
      "num_tokens": 12998081.0,
      "step": 6415
    },
    {
      "entropy": 5.932345581054688,
      "epoch": 0.561041684872848,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004974701443159812,
      "loss": 5.7488,
      "mean_token_accuracy": 0.1667596787214279,
      "num_tokens": 13008047.0,
      "step": 6420
    },
    {
      "entropy": 5.987129354476929,
      "epoch": 0.5614786332255527,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004974654824590277,
      "loss": 5.7166,
      "mean_token_accuracy": 0.16370745301246642,
      "num_tokens": 13018048.0,
      "step": 6425
    },
    {
      "entropy": 5.912373161315918,
      "epoch": 0.5619155815782575,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004974608163350697,
      "loss": 5.7268,
      "mean_token_accuracy": 0.16715857833623887,
      "num_tokens": 13028156.0,
      "step": 6430
    },
    {
      "entropy": 6.01599760055542,
      "epoch": 0.5623525299309622,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004974561459441967,
      "loss": 5.7524,
      "mean_token_accuracy": 0.16166191399097443,
      "num_tokens": 13038191.0,
      "step": 6435
    },
    {
      "entropy": 5.953306198120117,
      "epoch": 0.5627894782836669,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004974514712864981,
      "loss": 5.7371,
      "mean_token_accuracy": 0.1662627562880516,
      "num_tokens": 13049085.0,
      "step": 6440
    },
    {
      "entropy": 5.944247102737426,
      "epoch": 0.5632264266363716,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004974467923620639,
      "loss": 5.7598,
      "mean_token_accuracy": 0.1655406966805458,
      "num_tokens": 13060010.0,
      "step": 6445
    },
    {
      "entropy": 6.071695613861084,
      "epoch": 0.5636633749890763,
      "grad_norm": 0.875,
      "learning_rate": 0.0004974421091709836,
      "loss": 5.7472,
      "mean_token_accuracy": 0.16250929832458497,
      "num_tokens": 13069768.0,
      "step": 6450
    },
    {
      "entropy": 5.96634840965271,
      "epoch": 0.564100323341781,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004974374217133471,
      "loss": 5.7416,
      "mean_token_accuracy": 0.16375129520893097,
      "num_tokens": 13079842.0,
      "step": 6455
    },
    {
      "entropy": 5.958716106414795,
      "epoch": 0.5645372716944858,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004974327299892443,
      "loss": 5.6762,
      "mean_token_accuracy": 0.16915546208620072,
      "num_tokens": 13089235.0,
      "step": 6460
    },
    {
      "entropy": 5.955935382843018,
      "epoch": 0.5649742200471904,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004974280339987651,
      "loss": 5.7412,
      "mean_token_accuracy": 0.16048220098018645,
      "num_tokens": 13098364.0,
      "step": 6465
    },
    {
      "entropy": 6.020211553573608,
      "epoch": 0.5654111683998951,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004974233337419997,
      "loss": 5.879,
      "mean_token_accuracy": 0.15505149215459824,
      "num_tokens": 13108876.0,
      "step": 6470
    },
    {
      "entropy": 6.071588563919067,
      "epoch": 0.5658481167525998,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004974186292190383,
      "loss": 5.8513,
      "mean_token_accuracy": 0.1634252041578293,
      "num_tokens": 13119591.0,
      "step": 6475
    },
    {
      "entropy": 5.996852397918701,
      "epoch": 0.5662850651053045,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000497413920429971,
      "loss": 5.7317,
      "mean_token_accuracy": 0.17271516621112823,
      "num_tokens": 13129796.0,
      "step": 6480
    },
    {
      "entropy": 5.949346733093262,
      "epoch": 0.5667220134580092,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004974092073748882,
      "loss": 5.6962,
      "mean_token_accuracy": 0.17277736514806746,
      "num_tokens": 13139302.0,
      "step": 6485
    },
    {
      "entropy": 5.942293405532837,
      "epoch": 0.567158961810714,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004974044900538802,
      "loss": 5.7582,
      "mean_token_accuracy": 0.1677715688943863,
      "num_tokens": 13148983.0,
      "step": 6490
    },
    {
      "entropy": 5.999657773971558,
      "epoch": 0.5675959101634187,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004973997684670375,
      "loss": 5.7311,
      "mean_token_accuracy": 0.16669172197580337,
      "num_tokens": 13159016.0,
      "step": 6495
    },
    {
      "entropy": 5.9895348072052,
      "epoch": 0.5680328585161234,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004973950426144508,
      "loss": 5.7804,
      "mean_token_accuracy": 0.161250601708889,
      "num_tokens": 13169705.0,
      "step": 6500
    },
    {
      "entropy": 6.001507377624511,
      "epoch": 0.5684698068688281,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004973903124962106,
      "loss": 5.7699,
      "mean_token_accuracy": 0.1560562565922737,
      "num_tokens": 13181372.0,
      "step": 6505
    },
    {
      "entropy": 5.9984156608581545,
      "epoch": 0.5689067552215328,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004973855781124077,
      "loss": 5.7958,
      "mean_token_accuracy": 0.1600258618593216,
      "num_tokens": 13192581.0,
      "step": 6510
    },
    {
      "entropy": 6.005476808547973,
      "epoch": 0.5693437035742375,
      "grad_norm": 1.0,
      "learning_rate": 0.0004973808394631329,
      "loss": 5.83,
      "mean_token_accuracy": 0.15928068459033967,
      "num_tokens": 13203269.0,
      "step": 6515
    },
    {
      "entropy": 5.967685270309448,
      "epoch": 0.5697806519269423,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000497376096548477,
      "loss": 5.7016,
      "mean_token_accuracy": 0.1716892898082733,
      "num_tokens": 13213183.0,
      "step": 6520
    },
    {
      "entropy": 5.963852739334106,
      "epoch": 0.570217600279647,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004973713493685311,
      "loss": 5.7807,
      "mean_token_accuracy": 0.16549151092767717,
      "num_tokens": 13223196.0,
      "step": 6525
    },
    {
      "entropy": 5.970004081726074,
      "epoch": 0.5706545486323517,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004973665979233861,
      "loss": 5.7067,
      "mean_token_accuracy": 0.16781740188598632,
      "num_tokens": 13233915.0,
      "step": 6530
    },
    {
      "entropy": 5.953088569641113,
      "epoch": 0.5710914969850563,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004973618422131334,
      "loss": 5.7406,
      "mean_token_accuracy": 0.16723136752843856,
      "num_tokens": 13245108.0,
      "step": 6535
    },
    {
      "entropy": 5.925341558456421,
      "epoch": 0.571528445337761,
      "grad_norm": 0.953125,
      "learning_rate": 0.000497357082237864,
      "loss": 5.7605,
      "mean_token_accuracy": 0.1683202549815178,
      "num_tokens": 13255508.0,
      "step": 6540
    },
    {
      "entropy": 6.031052112579346,
      "epoch": 0.5719653936904658,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004973523179976692,
      "loss": 5.77,
      "mean_token_accuracy": 0.1679235577583313,
      "num_tokens": 13265118.0,
      "step": 6545
    },
    {
      "entropy": 6.012777090072632,
      "epoch": 0.5724023420431705,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004973475494926404,
      "loss": 5.6697,
      "mean_token_accuracy": 0.16822943091392517,
      "num_tokens": 13275947.0,
      "step": 6550
    },
    {
      "entropy": 5.925443029403686,
      "epoch": 0.5728392903958752,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004973427767228692,
      "loss": 5.7079,
      "mean_token_accuracy": 0.1724183291196823,
      "num_tokens": 13284662.0,
      "step": 6555
    },
    {
      "entropy": 5.989375829696655,
      "epoch": 0.5732762387485799,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000497337999688447,
      "loss": 5.7653,
      "mean_token_accuracy": 0.16682929098606109,
      "num_tokens": 13295627.0,
      "step": 6560
    },
    {
      "entropy": 6.001201534271241,
      "epoch": 0.5737131871012846,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004973332183894653,
      "loss": 5.6792,
      "mean_token_accuracy": 0.16851622611284256,
      "num_tokens": 13305205.0,
      "step": 6565
    },
    {
      "entropy": 5.889416980743408,
      "epoch": 0.5741501354539893,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004973284328260162,
      "loss": 5.7034,
      "mean_token_accuracy": 0.16735193133354187,
      "num_tokens": 13315464.0,
      "step": 6570
    },
    {
      "entropy": 5.875208044052124,
      "epoch": 0.5745870838066941,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004973236429981912,
      "loss": 5.637,
      "mean_token_accuracy": 0.17321056127548218,
      "num_tokens": 13325468.0,
      "step": 6575
    },
    {
      "entropy": 6.004652500152588,
      "epoch": 0.5750240321593988,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004973188489060821,
      "loss": 5.747,
      "mean_token_accuracy": 0.16478069573640824,
      "num_tokens": 13334719.0,
      "step": 6580
    },
    {
      "entropy": 5.934004306793213,
      "epoch": 0.5754609805121035,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004973140505497812,
      "loss": 5.6972,
      "mean_token_accuracy": 0.16642517298460008,
      "num_tokens": 13344545.0,
      "step": 6585
    },
    {
      "entropy": 5.982285881042481,
      "epoch": 0.5758979288648082,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004973092479293802,
      "loss": 5.7825,
      "mean_token_accuracy": 0.16220623403787612,
      "num_tokens": 13355212.0,
      "step": 6590
    },
    {
      "entropy": 5.955848264694214,
      "epoch": 0.5763348772175129,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004973044410449713,
      "loss": 5.684,
      "mean_token_accuracy": 0.16877468079328536,
      "num_tokens": 13364869.0,
      "step": 6595
    },
    {
      "entropy": 5.954962682723999,
      "epoch": 0.5767718255702176,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004972996298966468,
      "loss": 5.7296,
      "mean_token_accuracy": 0.16358875185251237,
      "num_tokens": 13376511.0,
      "step": 6600
    },
    {
      "entropy": 5.953457450866699,
      "epoch": 0.5772087739229224,
      "grad_norm": 0.953125,
      "learning_rate": 0.000497294814484499,
      "loss": 5.7512,
      "mean_token_accuracy": 0.1643550157546997,
      "num_tokens": 13385470.0,
      "step": 6605
    },
    {
      "entropy": 5.992721509933472,
      "epoch": 0.577645722275627,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00049728999480862,
      "loss": 5.8122,
      "mean_token_accuracy": 0.1704306498169899,
      "num_tokens": 13395328.0,
      "step": 6610
    },
    {
      "entropy": 5.996850776672363,
      "epoch": 0.5780826706283317,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004972851708691025,
      "loss": 5.72,
      "mean_token_accuracy": 0.17624374926090242,
      "num_tokens": 13405319.0,
      "step": 6615
    },
    {
      "entropy": 5.8612841129302975,
      "epoch": 0.5785196189810364,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004972803426660389,
      "loss": 5.7266,
      "mean_token_accuracy": 0.16552847474813462,
      "num_tokens": 13416119.0,
      "step": 6620
    },
    {
      "entropy": 5.950865125656128,
      "epoch": 0.5789565673337411,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004972755101995219,
      "loss": 5.7194,
      "mean_token_accuracy": 0.16360987722873688,
      "num_tokens": 13425824.0,
      "step": 6625
    },
    {
      "entropy": 5.915494632720947,
      "epoch": 0.5793935156864458,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004972706734696442,
      "loss": 5.6357,
      "mean_token_accuracy": 0.17530417740345,
      "num_tokens": 13435506.0,
      "step": 6630
    },
    {
      "entropy": 5.894372844696045,
      "epoch": 0.5798304640391506,
      "grad_norm": 1.0,
      "learning_rate": 0.0004972658324764984,
      "loss": 5.676,
      "mean_token_accuracy": 0.17498529255390166,
      "num_tokens": 13445313.0,
      "step": 6635
    },
    {
      "entropy": 5.907119607925415,
      "epoch": 0.5802674123918553,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004972609872201775,
      "loss": 5.6311,
      "mean_token_accuracy": 0.17491835206747056,
      "num_tokens": 13455312.0,
      "step": 6640
    },
    {
      "entropy": 5.974127006530762,
      "epoch": 0.58070436074456,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004972561377007744,
      "loss": 5.7488,
      "mean_token_accuracy": 0.16536202132701874,
      "num_tokens": 13465721.0,
      "step": 6645
    },
    {
      "entropy": 5.9745584487915036,
      "epoch": 0.5811413090972647,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004972512839183822,
      "loss": 5.7711,
      "mean_token_accuracy": 0.16656267195940017,
      "num_tokens": 13477185.0,
      "step": 6650
    },
    {
      "entropy": 5.889624452590942,
      "epoch": 0.5815782574499694,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004972464258730939,
      "loss": 5.6856,
      "mean_token_accuracy": 0.18008620291948318,
      "num_tokens": 13487846.0,
      "step": 6655
    },
    {
      "entropy": 5.882382011413574,
      "epoch": 0.5820152058026741,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004972415635650027,
      "loss": 5.7704,
      "mean_token_accuracy": 0.17557527720928193,
      "num_tokens": 13498175.0,
      "step": 6660
    },
    {
      "entropy": 6.014629983901978,
      "epoch": 0.5824521541553789,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004972366969942017,
      "loss": 5.6957,
      "mean_token_accuracy": 0.1605272948741913,
      "num_tokens": 13506864.0,
      "step": 6665
    },
    {
      "entropy": 5.937314748764038,
      "epoch": 0.5828891025080836,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004972318261607845,
      "loss": 5.7079,
      "mean_token_accuracy": 0.16786991506814958,
      "num_tokens": 13516208.0,
      "step": 6670
    },
    {
      "entropy": 5.977537775039673,
      "epoch": 0.5833260508607883,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004972269510648444,
      "loss": 5.7794,
      "mean_token_accuracy": 0.16236940771341324,
      "num_tokens": 13526902.0,
      "step": 6675
    },
    {
      "entropy": 5.965627193450928,
      "epoch": 0.583762999213493,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000497222071706475,
      "loss": 5.6901,
      "mean_token_accuracy": 0.17912275791168214,
      "num_tokens": 13537001.0,
      "step": 6680
    },
    {
      "entropy": 5.88482894897461,
      "epoch": 0.5841999475661976,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004972171880857698,
      "loss": 5.6753,
      "mean_token_accuracy": 0.1726643517613411,
      "num_tokens": 13546971.0,
      "step": 6685
    },
    {
      "entropy": 5.886464691162109,
      "epoch": 0.5846368959189023,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004972123002028226,
      "loss": 5.6578,
      "mean_token_accuracy": 0.17230957597494126,
      "num_tokens": 13557202.0,
      "step": 6690
    },
    {
      "entropy": 5.877233695983887,
      "epoch": 0.5850738442716071,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004972074080577269,
      "loss": 5.6558,
      "mean_token_accuracy": 0.17665928453207017,
      "num_tokens": 13567291.0,
      "step": 6695
    },
    {
      "entropy": 5.982480716705322,
      "epoch": 0.5855107926243118,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004972025116505767,
      "loss": 5.7338,
      "mean_token_accuracy": 0.17252862304449082,
      "num_tokens": 13576864.0,
      "step": 6700
    },
    {
      "entropy": 5.950371217727661,
      "epoch": 0.5859477409770165,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004971976109814658,
      "loss": 5.6519,
      "mean_token_accuracy": 0.17174218147993087,
      "num_tokens": 13586728.0,
      "step": 6705
    },
    {
      "entropy": 5.909093046188355,
      "epoch": 0.5863846893297212,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004971927060504885,
      "loss": 5.7179,
      "mean_token_accuracy": 0.1715800240635872,
      "num_tokens": 13596875.0,
      "step": 6710
    },
    {
      "entropy": 6.0003053665161135,
      "epoch": 0.5868216376824259,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004971877968577386,
      "loss": 5.7377,
      "mean_token_accuracy": 0.1553231045603752,
      "num_tokens": 13606483.0,
      "step": 6715
    },
    {
      "entropy": 5.97728123664856,
      "epoch": 0.5872585860351307,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004971828834033102,
      "loss": 5.7735,
      "mean_token_accuracy": 0.16625204980373381,
      "num_tokens": 13616780.0,
      "step": 6720
    },
    {
      "entropy": 5.951204586029053,
      "epoch": 0.5876955343878354,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004971779656872978,
      "loss": 5.7274,
      "mean_token_accuracy": 0.16728439480066298,
      "num_tokens": 13626418.0,
      "step": 6725
    },
    {
      "entropy": 6.013874578475952,
      "epoch": 0.5881324827405401,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004971730437097955,
      "loss": 5.796,
      "mean_token_accuracy": 0.1662833124399185,
      "num_tokens": 13636956.0,
      "step": 6730
    },
    {
      "entropy": 5.980191850662232,
      "epoch": 0.5885694310932448,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000497168117470898,
      "loss": 5.7148,
      "mean_token_accuracy": 0.170338873565197,
      "num_tokens": 13647458.0,
      "step": 6735
    },
    {
      "entropy": 5.886775112152099,
      "epoch": 0.5890063794459495,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004971631869706995,
      "loss": 5.6742,
      "mean_token_accuracy": 0.17262455075979233,
      "num_tokens": 13657655.0,
      "step": 6740
    },
    {
      "entropy": 6.011074113845825,
      "epoch": 0.5894433277986542,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004971582522092946,
      "loss": 5.7575,
      "mean_token_accuracy": 0.162050524353981,
      "num_tokens": 13668472.0,
      "step": 6745
    },
    {
      "entropy": 5.934489011764526,
      "epoch": 0.589880276151359,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000497153313186778,
      "loss": 5.6395,
      "mean_token_accuracy": 0.17543691247701645,
      "num_tokens": 13677319.0,
      "step": 6750
    },
    {
      "entropy": 5.881658554077148,
      "epoch": 0.5903172245040637,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004971483699032445,
      "loss": 5.6806,
      "mean_token_accuracy": 0.17045584470033645,
      "num_tokens": 13688037.0,
      "step": 6755
    },
    {
      "entropy": 6.036482334136963,
      "epoch": 0.5907541728567683,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004971434223587888,
      "loss": 5.7839,
      "mean_token_accuracy": 0.16187653988599776,
      "num_tokens": 13698441.0,
      "step": 6760
    },
    {
      "entropy": 5.9418800354003904,
      "epoch": 0.591191121209473,
      "grad_norm": 1.015625,
      "learning_rate": 0.000497138470553506,
      "loss": 5.7108,
      "mean_token_accuracy": 0.1708940878510475,
      "num_tokens": 13708140.0,
      "step": 6765
    },
    {
      "entropy": 5.977269840240479,
      "epoch": 0.5916280695621777,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004971335144874909,
      "loss": 5.705,
      "mean_token_accuracy": 0.17302952110767364,
      "num_tokens": 13718069.0,
      "step": 6770
    },
    {
      "entropy": 5.918741083145141,
      "epoch": 0.5920650179148824,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004971285541608385,
      "loss": 5.6742,
      "mean_token_accuracy": 0.18713206499814988,
      "num_tokens": 13727932.0,
      "step": 6775
    },
    {
      "entropy": 6.035796403884888,
      "epoch": 0.5925019662675872,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004971235895736441,
      "loss": 5.7665,
      "mean_token_accuracy": 0.16273656785488128,
      "num_tokens": 13738239.0,
      "step": 6780
    },
    {
      "entropy": 5.953591442108154,
      "epoch": 0.5929389146202919,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004971186207260029,
      "loss": 5.7171,
      "mean_token_accuracy": 0.1649756669998169,
      "num_tokens": 13748622.0,
      "step": 6785
    },
    {
      "entropy": 5.842089557647705,
      "epoch": 0.5933758629729966,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00049711364761801,
      "loss": 5.6552,
      "mean_token_accuracy": 0.17211836874485015,
      "num_tokens": 13759221.0,
      "step": 6790
    },
    {
      "entropy": 5.9833746433258055,
      "epoch": 0.5938128113257013,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000497108670249761,
      "loss": 5.6507,
      "mean_token_accuracy": 0.1691410258412361,
      "num_tokens": 13768343.0,
      "step": 6795
    },
    {
      "entropy": 5.943966627120972,
      "epoch": 0.594249759678406,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004971036886213515,
      "loss": 5.6215,
      "mean_token_accuracy": 0.17078949362039567,
      "num_tokens": 13778225.0,
      "step": 6800
    },
    {
      "entropy": 5.920519733428955,
      "epoch": 0.5946867080311107,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004970987027328767,
      "loss": 5.7289,
      "mean_token_accuracy": 0.1676565796136856,
      "num_tokens": 13788633.0,
      "step": 6805
    },
    {
      "entropy": 5.954409980773926,
      "epoch": 0.5951236563838155,
      "grad_norm": 1.0,
      "learning_rate": 0.0004970937125844325,
      "loss": 5.6403,
      "mean_token_accuracy": 0.17139566987752913,
      "num_tokens": 13798172.0,
      "step": 6810
    },
    {
      "entropy": 5.86420373916626,
      "epoch": 0.5955606047365202,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004970887181761144,
      "loss": 5.632,
      "mean_token_accuracy": 0.17911376953125,
      "num_tokens": 13807284.0,
      "step": 6815
    },
    {
      "entropy": 5.979779958724976,
      "epoch": 0.5959975530892249,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004970837195080184,
      "loss": 5.6955,
      "mean_token_accuracy": 0.17271262109279634,
      "num_tokens": 13816807.0,
      "step": 6820
    },
    {
      "entropy": 5.959517240524292,
      "epoch": 0.5964345014419296,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004970787165802403,
      "loss": 5.6634,
      "mean_token_accuracy": 0.16695165783166885,
      "num_tokens": 13827635.0,
      "step": 6825
    },
    {
      "entropy": 5.903842496871948,
      "epoch": 0.5968714497946342,
      "grad_norm": 0.890625,
      "learning_rate": 0.000497073709392876,
      "loss": 5.7453,
      "mean_token_accuracy": 0.16901420205831527,
      "num_tokens": 13838060.0,
      "step": 6830
    },
    {
      "entropy": 5.983510446548462,
      "epoch": 0.5973083981473389,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004970686979460216,
      "loss": 5.7221,
      "mean_token_accuracy": 0.16646428257226945,
      "num_tokens": 13848669.0,
      "step": 6835
    },
    {
      "entropy": 5.916943836212158,
      "epoch": 0.5977453465000437,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004970636822397733,
      "loss": 5.6914,
      "mean_token_accuracy": 0.16970677375793458,
      "num_tokens": 13857949.0,
      "step": 6840
    },
    {
      "entropy": 5.927335548400879,
      "epoch": 0.5981822948527484,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004970586622742272,
      "loss": 5.6682,
      "mean_token_accuracy": 0.1799116387963295,
      "num_tokens": 13867726.0,
      "step": 6845
    },
    {
      "entropy": 6.023752498626709,
      "epoch": 0.5986192432054531,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004970536380494795,
      "loss": 5.7569,
      "mean_token_accuracy": 0.16346921622753144,
      "num_tokens": 13878072.0,
      "step": 6850
    },
    {
      "entropy": 5.845704841613769,
      "epoch": 0.5990561915581578,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004970486095656268,
      "loss": 5.6177,
      "mean_token_accuracy": 0.18290360569953917,
      "num_tokens": 13887771.0,
      "step": 6855
    },
    {
      "entropy": 5.943838453292846,
      "epoch": 0.5994931399108625,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004970435768227652,
      "loss": 5.6713,
      "mean_token_accuracy": 0.1734023317694664,
      "num_tokens": 13897867.0,
      "step": 6860
    },
    {
      "entropy": 5.9663163185119625,
      "epoch": 0.5999300882635672,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004970385398209916,
      "loss": 5.7308,
      "mean_token_accuracy": 0.1692700207233429,
      "num_tokens": 13909171.0,
      "step": 6865
    },
    {
      "entropy": 5.96572790145874,
      "epoch": 0.600367036616272,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004970334985604024,
      "loss": 5.7266,
      "mean_token_accuracy": 0.17131581157445908,
      "num_tokens": 13920374.0,
      "step": 6870
    },
    {
      "entropy": 5.974809646606445,
      "epoch": 0.6008039849689767,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004970284530410945,
      "loss": 5.7242,
      "mean_token_accuracy": 0.16636893451213836,
      "num_tokens": 13930150.0,
      "step": 6875
    },
    {
      "entropy": 5.941253757476806,
      "epoch": 0.6012409333216814,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004970234032631644,
      "loss": 5.679,
      "mean_token_accuracy": 0.17380213290452956,
      "num_tokens": 13940553.0,
      "step": 6880
    },
    {
      "entropy": 5.908690881729126,
      "epoch": 0.6016778816743861,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004970183492267091,
      "loss": 5.7915,
      "mean_token_accuracy": 0.1736527144908905,
      "num_tokens": 13951086.0,
      "step": 6885
    },
    {
      "entropy": 5.973720264434815,
      "epoch": 0.6021148300270908,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004970132909318254,
      "loss": 5.7127,
      "mean_token_accuracy": 0.1649186909198761,
      "num_tokens": 13960684.0,
      "step": 6890
    },
    {
      "entropy": 5.815363359451294,
      "epoch": 0.6025517783797955,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004970082283786107,
      "loss": 5.5476,
      "mean_token_accuracy": 0.19229229539632797,
      "num_tokens": 13971409.0,
      "step": 6895
    },
    {
      "entropy": 5.947085905075073,
      "epoch": 0.6029887267325003,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004970031615671617,
      "loss": 5.7267,
      "mean_token_accuracy": 0.17262187600135803,
      "num_tokens": 13981185.0,
      "step": 6900
    },
    {
      "entropy": 5.928203487396241,
      "epoch": 0.603425675085205,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004969980904975758,
      "loss": 5.7352,
      "mean_token_accuracy": 0.1727176308631897,
      "num_tokens": 13991670.0,
      "step": 6905
    },
    {
      "entropy": 5.878294849395752,
      "epoch": 0.6038626234379096,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004969930151699501,
      "loss": 5.6758,
      "mean_token_accuracy": 0.17196464240550996,
      "num_tokens": 14002343.0,
      "step": 6910
    },
    {
      "entropy": 5.94720573425293,
      "epoch": 0.6042995717906143,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000496987935584382,
      "loss": 5.7377,
      "mean_token_accuracy": 0.16700234413146972,
      "num_tokens": 14013373.0,
      "step": 6915
    },
    {
      "entropy": 5.98115291595459,
      "epoch": 0.604736520143319,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004969828517409693,
      "loss": 5.666,
      "mean_token_accuracy": 0.17324753999710082,
      "num_tokens": 14023233.0,
      "step": 6920
    },
    {
      "entropy": 5.888190889358521,
      "epoch": 0.6051734684960238,
      "grad_norm": 0.953125,
      "learning_rate": 0.000496977763639809,
      "loss": 5.7291,
      "mean_token_accuracy": 0.1673596829175949,
      "num_tokens": 14033054.0,
      "step": 6925
    },
    {
      "entropy": 5.933578014373779,
      "epoch": 0.6056104168487285,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004969726712809988,
      "loss": 5.6405,
      "mean_token_accuracy": 0.16903098672628403,
      "num_tokens": 14042039.0,
      "step": 6930
    },
    {
      "entropy": 6.000370264053345,
      "epoch": 0.6060473652014332,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004969675746646364,
      "loss": 5.7968,
      "mean_token_accuracy": 0.16199924647808076,
      "num_tokens": 14051957.0,
      "step": 6935
    },
    {
      "entropy": 5.968285322189331,
      "epoch": 0.6064843135541379,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004969624737908197,
      "loss": 5.6751,
      "mean_token_accuracy": 0.17337231189012528,
      "num_tokens": 14062438.0,
      "step": 6940
    },
    {
      "entropy": 5.9567700862884525,
      "epoch": 0.6069212619068426,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004969573686596464,
      "loss": 5.7069,
      "mean_token_accuracy": 0.1631592631340027,
      "num_tokens": 14073196.0,
      "step": 6945
    },
    {
      "entropy": 5.923959589004516,
      "epoch": 0.6073582102595473,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004969522592712146,
      "loss": 5.6906,
      "mean_token_accuracy": 0.17131973654031754,
      "num_tokens": 14083462.0,
      "step": 6950
    },
    {
      "entropy": 5.925915050506592,
      "epoch": 0.6077951586122521,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004969471456256221,
      "loss": 5.7555,
      "mean_token_accuracy": 0.16844635009765624,
      "num_tokens": 14094194.0,
      "step": 6955
    },
    {
      "entropy": 5.979562854766845,
      "epoch": 0.6082321069649568,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000496942027722967,
      "loss": 5.7067,
      "mean_token_accuracy": 0.16755084395408631,
      "num_tokens": 14104553.0,
      "step": 6960
    },
    {
      "entropy": 6.000029373168945,
      "epoch": 0.6086690553176615,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004969369055633476,
      "loss": 5.6398,
      "mean_token_accuracy": 0.17094933986663818,
      "num_tokens": 14114140.0,
      "step": 6965
    },
    {
      "entropy": 5.9206178188323975,
      "epoch": 0.6091060036703662,
      "grad_norm": 0.890625,
      "learning_rate": 0.000496931779146862,
      "loss": 5.772,
      "mean_token_accuracy": 0.16876591742038727,
      "num_tokens": 14124723.0,
      "step": 6970
    },
    {
      "entropy": 5.939309024810791,
      "epoch": 0.6095429520230708,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004969266484736085,
      "loss": 5.6243,
      "mean_token_accuracy": 0.1735442742705345,
      "num_tokens": 14134030.0,
      "step": 6975
    },
    {
      "entropy": 5.905480194091797,
      "epoch": 0.6099799003757755,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004969215135436858,
      "loss": 5.6065,
      "mean_token_accuracy": 0.17367956936359405,
      "num_tokens": 14144691.0,
      "step": 6980
    },
    {
      "entropy": 5.90670599937439,
      "epoch": 0.6104168487284803,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004969163743571921,
      "loss": 5.7713,
      "mean_token_accuracy": 0.16909185349941253,
      "num_tokens": 14155114.0,
      "step": 6985
    },
    {
      "entropy": 6.021292018890381,
      "epoch": 0.610853797081185,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000496911230914226,
      "loss": 5.7786,
      "mean_token_accuracy": 0.1627641126513481,
      "num_tokens": 14166097.0,
      "step": 6990
    },
    {
      "entropy": 5.937092208862305,
      "epoch": 0.6112907454338897,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004969060832148862,
      "loss": 5.6421,
      "mean_token_accuracy": 0.1662045642733574,
      "num_tokens": 14176538.0,
      "step": 6995
    },
    {
      "entropy": 5.890448474884034,
      "epoch": 0.6117276937865944,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004969009312592715,
      "loss": 5.6679,
      "mean_token_accuracy": 0.16790677458047867,
      "num_tokens": 14187912.0,
      "step": 7000
    },
    {
      "entropy": 5.8025237083435055,
      "epoch": 0.6121646421392991,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004968957750474806,
      "loss": 5.5523,
      "mean_token_accuracy": 0.18280513137578963,
      "num_tokens": 14198446.0,
      "step": 7005
    },
    {
      "entropy": 6.018140268325806,
      "epoch": 0.6126015904920038,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004968906145796126,
      "loss": 5.7943,
      "mean_token_accuracy": 0.16550207287073135,
      "num_tokens": 14207723.0,
      "step": 7010
    },
    {
      "entropy": 6.006080532073975,
      "epoch": 0.6130385388447086,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004968854498557662,
      "loss": 5.6628,
      "mean_token_accuracy": 0.17237055897712708,
      "num_tokens": 14217590.0,
      "step": 7015
    },
    {
      "entropy": 5.95548152923584,
      "epoch": 0.6134754871974133,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004968802808760407,
      "loss": 5.7355,
      "mean_token_accuracy": 0.1631538063287735,
      "num_tokens": 14226629.0,
      "step": 7020
    },
    {
      "entropy": 6.013607406616211,
      "epoch": 0.613912435550118,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004968751076405351,
      "loss": 5.8046,
      "mean_token_accuracy": 0.1598005250096321,
      "num_tokens": 14235782.0,
      "step": 7025
    },
    {
      "entropy": 5.933961296081543,
      "epoch": 0.6143493839028227,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004968699301493488,
      "loss": 5.6821,
      "mean_token_accuracy": 0.17899044305086137,
      "num_tokens": 14246687.0,
      "step": 7030
    },
    {
      "entropy": 5.987144088745117,
      "epoch": 0.6147863322555274,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004968647484025808,
      "loss": 5.6997,
      "mean_token_accuracy": 0.17080247551202773,
      "num_tokens": 14255961.0,
      "step": 7035
    },
    {
      "entropy": 5.953332567214966,
      "epoch": 0.6152232806082321,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004968595624003307,
      "loss": 5.6962,
      "mean_token_accuracy": 0.17000192999839783,
      "num_tokens": 14266024.0,
      "step": 7040
    },
    {
      "entropy": 5.918338632583618,
      "epoch": 0.6156602289609369,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000496854372142698,
      "loss": 5.7182,
      "mean_token_accuracy": 0.16677912771701814,
      "num_tokens": 14276590.0,
      "step": 7045
    },
    {
      "entropy": 5.953415250778198,
      "epoch": 0.6160971773136416,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004968491776297821,
      "loss": 5.7559,
      "mean_token_accuracy": 0.16753247678279876,
      "num_tokens": 14285992.0,
      "step": 7050
    },
    {
      "entropy": 5.925625371932983,
      "epoch": 0.6165341256663462,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004968439788616829,
      "loss": 5.6572,
      "mean_token_accuracy": 0.16945477277040483,
      "num_tokens": 14295843.0,
      "step": 7055
    },
    {
      "entropy": 5.913970708847046,
      "epoch": 0.6169710740190509,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004968387758384999,
      "loss": 5.6796,
      "mean_token_accuracy": 0.17228251099586486,
      "num_tokens": 14305643.0,
      "step": 7060
    },
    {
      "entropy": 5.96819543838501,
      "epoch": 0.6174080223717556,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004968335685603328,
      "loss": 5.6658,
      "mean_token_accuracy": 0.17211828082799913,
      "num_tokens": 14315907.0,
      "step": 7065
    },
    {
      "entropy": 5.921641874313354,
      "epoch": 0.6178449707244603,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004968283570272817,
      "loss": 5.7331,
      "mean_token_accuracy": 0.16253844350576402,
      "num_tokens": 14325893.0,
      "step": 7070
    },
    {
      "entropy": 5.9236249923706055,
      "epoch": 0.6182819190771651,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004968231412394464,
      "loss": 5.7236,
      "mean_token_accuracy": 0.16289602518081664,
      "num_tokens": 14337510.0,
      "step": 7075
    },
    {
      "entropy": 5.979340839385986,
      "epoch": 0.6187188674298698,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004968179211969272,
      "loss": 5.6639,
      "mean_token_accuracy": 0.17410714775323868,
      "num_tokens": 14346373.0,
      "step": 7080
    },
    {
      "entropy": 5.843090963363648,
      "epoch": 0.6191558157825745,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004968126968998239,
      "loss": 5.5522,
      "mean_token_accuracy": 0.1894397884607315,
      "num_tokens": 14356303.0,
      "step": 7085
    },
    {
      "entropy": 5.921578359603882,
      "epoch": 0.6195927641352792,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004968074683482369,
      "loss": 5.693,
      "mean_token_accuracy": 0.17338753044605254,
      "num_tokens": 14366171.0,
      "step": 7090
    },
    {
      "entropy": 6.00875334739685,
      "epoch": 0.6200297124879839,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004968022355422663,
      "loss": 5.7933,
      "mean_token_accuracy": 0.16383754909038545,
      "num_tokens": 14376133.0,
      "step": 7095
    },
    {
      "entropy": 5.976847076416016,
      "epoch": 0.6204666608406886,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004967969984820128,
      "loss": 5.6689,
      "mean_token_accuracy": 0.16863363832235337,
      "num_tokens": 14385670.0,
      "step": 7100
    },
    {
      "entropy": 5.931465339660645,
      "epoch": 0.6209036091933934,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004967917571675765,
      "loss": 5.6699,
      "mean_token_accuracy": 0.17102366983890532,
      "num_tokens": 14396002.0,
      "step": 7105
    },
    {
      "entropy": 5.969066715240478,
      "epoch": 0.6213405575460981,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004967865115990582,
      "loss": 5.792,
      "mean_token_accuracy": 0.16192105785012245,
      "num_tokens": 14406718.0,
      "step": 7110
    },
    {
      "entropy": 6.016126394271851,
      "epoch": 0.6217775058988028,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004967812617765583,
      "loss": 5.6994,
      "mean_token_accuracy": 0.16700837016105652,
      "num_tokens": 14417027.0,
      "step": 7115
    },
    {
      "entropy": 5.91008358001709,
      "epoch": 0.6222144542515075,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004967760077001775,
      "loss": 5.6674,
      "mean_token_accuracy": 0.17592547535896302,
      "num_tokens": 14426885.0,
      "step": 7120
    },
    {
      "entropy": 5.879356527328492,
      "epoch": 0.6226514026042121,
      "grad_norm": 0.875,
      "learning_rate": 0.0004967707493700169,
      "loss": 5.6575,
      "mean_token_accuracy": 0.17039713710546495,
      "num_tokens": 14437743.0,
      "step": 7125
    },
    {
      "entropy": 5.898950290679932,
      "epoch": 0.623088350956917,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004967654867861769,
      "loss": 5.6756,
      "mean_token_accuracy": 0.1753813624382019,
      "num_tokens": 14448808.0,
      "step": 7130
    },
    {
      "entropy": 5.9142289638519285,
      "epoch": 0.6235252993096216,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004967602199487588,
      "loss": 5.7142,
      "mean_token_accuracy": 0.17009922862052917,
      "num_tokens": 14458560.0,
      "step": 7135
    },
    {
      "entropy": 6.001022863388061,
      "epoch": 0.6239622476623263,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004967549488578635,
      "loss": 5.772,
      "mean_token_accuracy": 0.16817139834165573,
      "num_tokens": 14469375.0,
      "step": 7140
    },
    {
      "entropy": 5.981230783462524,
      "epoch": 0.624399196015031,
      "grad_norm": 0.90625,
      "learning_rate": 0.000496749673513592,
      "loss": 5.6925,
      "mean_token_accuracy": 0.16427991092205046,
      "num_tokens": 14478444.0,
      "step": 7145
    },
    {
      "entropy": 5.881944084167481,
      "epoch": 0.6248361443677357,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004967443939160456,
      "loss": 5.5761,
      "mean_token_accuracy": 0.18010010570287704,
      "num_tokens": 14487776.0,
      "step": 7150
    },
    {
      "entropy": 5.936268758773804,
      "epoch": 0.6252730927204404,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004967391100653255,
      "loss": 5.7059,
      "mean_token_accuracy": 0.17313706129789352,
      "num_tokens": 14497879.0,
      "step": 7155
    },
    {
      "entropy": 5.929531240463257,
      "epoch": 0.6257100410731452,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004967338219615331,
      "loss": 5.6308,
      "mean_token_accuracy": 0.17721510976552962,
      "num_tokens": 14507753.0,
      "step": 7160
    },
    {
      "entropy": 5.90009446144104,
      "epoch": 0.6261469894258499,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004967285296047699,
      "loss": 5.5962,
      "mean_token_accuracy": 0.18243916928768159,
      "num_tokens": 14517086.0,
      "step": 7165
    },
    {
      "entropy": 5.909822225570679,
      "epoch": 0.6265839377785546,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004967232329951372,
      "loss": 5.6781,
      "mean_token_accuracy": 0.16627199947834015,
      "num_tokens": 14527097.0,
      "step": 7170
    },
    {
      "entropy": 5.903944730758667,
      "epoch": 0.6270208861312593,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004967179321327367,
      "loss": 5.7157,
      "mean_token_accuracy": 0.1690282329916954,
      "num_tokens": 14537924.0,
      "step": 7175
    },
    {
      "entropy": 5.972730398178101,
      "epoch": 0.627457834483964,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004967126270176702,
      "loss": 5.7249,
      "mean_token_accuracy": 0.16390668004751205,
      "num_tokens": 14547542.0,
      "step": 7180
    },
    {
      "entropy": 5.916732931137085,
      "epoch": 0.6278947828366687,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004967073176500392,
      "loss": 5.6343,
      "mean_token_accuracy": 0.17581770718097686,
      "num_tokens": 14557150.0,
      "step": 7185
    },
    {
      "entropy": 5.93960337638855,
      "epoch": 0.6283317311893735,
      "grad_norm": 1.0,
      "learning_rate": 0.0004967020040299458,
      "loss": 5.671,
      "mean_token_accuracy": 0.17066039741039277,
      "num_tokens": 14567242.0,
      "step": 7190
    },
    {
      "entropy": 5.81256251335144,
      "epoch": 0.6287686795420782,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004966966861574917,
      "loss": 5.6281,
      "mean_token_accuracy": 0.17198438495397567,
      "num_tokens": 14576631.0,
      "step": 7195
    },
    {
      "entropy": 5.969675731658936,
      "epoch": 0.6292056278947828,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000496691364032779,
      "loss": 5.7364,
      "mean_token_accuracy": 0.16725654006004334,
      "num_tokens": 14587731.0,
      "step": 7200
    },
    {
      "entropy": 6.035846805572509,
      "epoch": 0.6296425762474875,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004966860376559097,
      "loss": 5.7326,
      "mean_token_accuracy": 0.17020094990730286,
      "num_tokens": 14598136.0,
      "step": 7205
    },
    {
      "entropy": 5.877604818344116,
      "epoch": 0.6300795246001922,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004966807070269862,
      "loss": 5.6892,
      "mean_token_accuracy": 0.1691464900970459,
      "num_tokens": 14607243.0,
      "step": 7210
    },
    {
      "entropy": 5.906348419189453,
      "epoch": 0.6305164729528969,
      "grad_norm": 1.125,
      "learning_rate": 0.0004966753721461105,
      "loss": 5.6941,
      "mean_token_accuracy": 0.16796606481075288,
      "num_tokens": 14617094.0,
      "step": 7215
    },
    {
      "entropy": 6.001829957962036,
      "epoch": 0.6309534213056017,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004966700330133849,
      "loss": 5.7269,
      "mean_token_accuracy": 0.16659176349639893,
      "num_tokens": 14626486.0,
      "step": 7220
    },
    {
      "entropy": 5.800681781768799,
      "epoch": 0.6313903696583064,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004966646896289121,
      "loss": 5.5863,
      "mean_token_accuracy": 0.1889587014913559,
      "num_tokens": 14637170.0,
      "step": 7225
    },
    {
      "entropy": 5.89100751876831,
      "epoch": 0.6318273180110111,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004966593419927944,
      "loss": 5.666,
      "mean_token_accuracy": 0.17304012030363083,
      "num_tokens": 14646396.0,
      "step": 7230
    },
    {
      "entropy": 5.977441883087158,
      "epoch": 0.6322642663637158,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004966539901051343,
      "loss": 5.723,
      "mean_token_accuracy": 0.16518759429454805,
      "num_tokens": 14656097.0,
      "step": 7235
    },
    {
      "entropy": 5.907673168182373,
      "epoch": 0.6327012147164205,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004966486339660345,
      "loss": 5.677,
      "mean_token_accuracy": 0.16812118738889695,
      "num_tokens": 14667124.0,
      "step": 7240
    },
    {
      "entropy": 5.925240612030029,
      "epoch": 0.6331381630691252,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004966432735755978,
      "loss": 5.7759,
      "mean_token_accuracy": 0.1663908302783966,
      "num_tokens": 14677427.0,
      "step": 7245
    },
    {
      "entropy": 5.905351161956787,
      "epoch": 0.63357511142183,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000496637908933927,
      "loss": 5.5878,
      "mean_token_accuracy": 0.17652758359909057,
      "num_tokens": 14686429.0,
      "step": 7250
    },
    {
      "entropy": 5.9338939666748045,
      "epoch": 0.6340120597745347,
      "grad_norm": 0.953125,
      "learning_rate": 0.000496632540041125,
      "loss": 5.6757,
      "mean_token_accuracy": 0.17885968536138536,
      "num_tokens": 14696264.0,
      "step": 7255
    },
    {
      "entropy": 5.968767929077148,
      "epoch": 0.6344490081272394,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004966271668972946,
      "loss": 5.6876,
      "mean_token_accuracy": 0.16377898454666137,
      "num_tokens": 14705935.0,
      "step": 7260
    },
    {
      "entropy": 5.923043203353882,
      "epoch": 0.6348859564799441,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004966217895025391,
      "loss": 5.6615,
      "mean_token_accuracy": 0.16858061403036118,
      "num_tokens": 14715978.0,
      "step": 7265
    },
    {
      "entropy": 5.960965919494629,
      "epoch": 0.6353229048326487,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004966164078569616,
      "loss": 5.6511,
      "mean_token_accuracy": 0.1691838711500168,
      "num_tokens": 14726170.0,
      "step": 7270
    },
    {
      "entropy": 5.911863613128662,
      "epoch": 0.6357598531853534,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004966110219606652,
      "loss": 5.6725,
      "mean_token_accuracy": 0.17294366657733917,
      "num_tokens": 14735775.0,
      "step": 7275
    },
    {
      "entropy": 5.906346416473388,
      "epoch": 0.6361968015380582,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004966056318137532,
      "loss": 5.6357,
      "mean_token_accuracy": 0.17020394802093505,
      "num_tokens": 14746143.0,
      "step": 7280
    },
    {
      "entropy": 5.936255025863647,
      "epoch": 0.6366337498907629,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004966002374163293,
      "loss": 5.6901,
      "mean_token_accuracy": 0.17279675751924514,
      "num_tokens": 14756707.0,
      "step": 7285
    },
    {
      "entropy": 5.9292398452758786,
      "epoch": 0.6370706982434676,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004965948387684964,
      "loss": 5.6894,
      "mean_token_accuracy": 0.17305904775857925,
      "num_tokens": 14765778.0,
      "step": 7290
    },
    {
      "entropy": 5.91487340927124,
      "epoch": 0.6375076465961723,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004965894358703587,
      "loss": 5.7178,
      "mean_token_accuracy": 0.16445838660001755,
      "num_tokens": 14775264.0,
      "step": 7295
    },
    {
      "entropy": 5.955113172531128,
      "epoch": 0.637944594948877,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004965840287220193,
      "loss": 5.6673,
      "mean_token_accuracy": 0.17495448142290115,
      "num_tokens": 14785754.0,
      "step": 7300
    },
    {
      "entropy": 5.96008825302124,
      "epoch": 0.6383815433015817,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004965786173235821,
      "loss": 5.7181,
      "mean_token_accuracy": 0.16657958626747132,
      "num_tokens": 14795936.0,
      "step": 7305
    },
    {
      "entropy": 5.981540250778198,
      "epoch": 0.6388184916542865,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004965732016751511,
      "loss": 5.6998,
      "mean_token_accuracy": 0.1693900302052498,
      "num_tokens": 14805721.0,
      "step": 7310
    },
    {
      "entropy": 5.91663236618042,
      "epoch": 0.6392554400069912,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004965677817768299,
      "loss": 5.7085,
      "mean_token_accuracy": 0.1714463621377945,
      "num_tokens": 14816610.0,
      "step": 7315
    },
    {
      "entropy": 5.971685791015625,
      "epoch": 0.6396923883596959,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004965623576287225,
      "loss": 5.6991,
      "mean_token_accuracy": 0.1705488532781601,
      "num_tokens": 14826126.0,
      "step": 7320
    },
    {
      "entropy": 5.900341033935547,
      "epoch": 0.6401293367124006,
      "grad_norm": 1.0,
      "learning_rate": 0.0004965569292309331,
      "loss": 5.5956,
      "mean_token_accuracy": 0.17734986841678618,
      "num_tokens": 14835657.0,
      "step": 7325
    },
    {
      "entropy": 5.885954093933106,
      "epoch": 0.6405662850651053,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004965514965835655,
      "loss": 5.6858,
      "mean_token_accuracy": 0.173714479804039,
      "num_tokens": 14845430.0,
      "step": 7330
    },
    {
      "entropy": 5.929603099822998,
      "epoch": 0.6410032334178101,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004965460596867243,
      "loss": 5.646,
      "mean_token_accuracy": 0.17476627677679063,
      "num_tokens": 14855459.0,
      "step": 7335
    },
    {
      "entropy": 5.976365089416504,
      "epoch": 0.6414401817705148,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004965406185405135,
      "loss": 5.6766,
      "mean_token_accuracy": 0.1709641382098198,
      "num_tokens": 14865375.0,
      "step": 7340
    },
    {
      "entropy": 5.93745436668396,
      "epoch": 0.6418771301232195,
      "grad_norm": 1.0,
      "learning_rate": 0.0004965351731450376,
      "loss": 5.6497,
      "mean_token_accuracy": 0.1737244874238968,
      "num_tokens": 14875211.0,
      "step": 7345
    },
    {
      "entropy": 5.8836335182189945,
      "epoch": 0.6423140784759241,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000496529723500401,
      "loss": 5.7199,
      "mean_token_accuracy": 0.1699363425374031,
      "num_tokens": 14886221.0,
      "step": 7350
    },
    {
      "entropy": 5.97067699432373,
      "epoch": 0.6427510268286288,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004965242696067081,
      "loss": 5.676,
      "mean_token_accuracy": 0.16997317373752593,
      "num_tokens": 14896002.0,
      "step": 7355
    },
    {
      "entropy": 5.964481687545776,
      "epoch": 0.6431879751813335,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004965188114640637,
      "loss": 5.733,
      "mean_token_accuracy": 0.1675407201051712,
      "num_tokens": 14906311.0,
      "step": 7360
    },
    {
      "entropy": 5.999446058273316,
      "epoch": 0.6436249235340383,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004965133490725725,
      "loss": 5.684,
      "mean_token_accuracy": 0.167180660367012,
      "num_tokens": 14915890.0,
      "step": 7365
    },
    {
      "entropy": 5.948170948028564,
      "epoch": 0.644061871886743,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004965078824323391,
      "loss": 5.6683,
      "mean_token_accuracy": 0.17221832275390625,
      "num_tokens": 14925891.0,
      "step": 7370
    },
    {
      "entropy": 5.909536457061767,
      "epoch": 0.6444988202394477,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004965024115434685,
      "loss": 5.6966,
      "mean_token_accuracy": 0.16815910637378692,
      "num_tokens": 14937272.0,
      "step": 7375
    },
    {
      "entropy": 5.941659307479858,
      "epoch": 0.6449357685921524,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004964969364060656,
      "loss": 5.7109,
      "mean_token_accuracy": 0.16980713456869126,
      "num_tokens": 14948135.0,
      "step": 7380
    },
    {
      "entropy": 6.027208423614502,
      "epoch": 0.6453727169448571,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004964914570202353,
      "loss": 5.7779,
      "mean_token_accuracy": 0.1631114214658737,
      "num_tokens": 14959254.0,
      "step": 7385
    },
    {
      "entropy": 5.942892360687256,
      "epoch": 0.6458096652975618,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004964859733860829,
      "loss": 5.6773,
      "mean_token_accuracy": 0.17487737387418748,
      "num_tokens": 14969313.0,
      "step": 7390
    },
    {
      "entropy": 5.906868171691895,
      "epoch": 0.6462466136502666,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004964804855037134,
      "loss": 5.6842,
      "mean_token_accuracy": 0.16762391477823257,
      "num_tokens": 14978963.0,
      "step": 7395
    },
    {
      "entropy": 6.004027223587036,
      "epoch": 0.6466835620029713,
      "grad_norm": 1.0,
      "learning_rate": 0.0004964749933732322,
      "loss": 5.7089,
      "mean_token_accuracy": 0.17225397676229476,
      "num_tokens": 14989355.0,
      "step": 7400
    },
    {
      "entropy": 5.934569358825684,
      "epoch": 0.647120510355676,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004964694969947445,
      "loss": 5.6543,
      "mean_token_accuracy": 0.17083992511034013,
      "num_tokens": 14998995.0,
      "step": 7405
    },
    {
      "entropy": 5.952825212478638,
      "epoch": 0.6475574587083807,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004964639963683558,
      "loss": 5.6994,
      "mean_token_accuracy": 0.17286603301763534,
      "num_tokens": 15008762.0,
      "step": 7410
    },
    {
      "entropy": 5.997217607498169,
      "epoch": 0.6479944070610854,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004964584914941716,
      "loss": 5.6642,
      "mean_token_accuracy": 0.17172364592552186,
      "num_tokens": 15020062.0,
      "step": 7415
    },
    {
      "entropy": 5.999050331115723,
      "epoch": 0.64843135541379,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004964529823722975,
      "loss": 5.8258,
      "mean_token_accuracy": 0.15894084870815278,
      "num_tokens": 15030440.0,
      "step": 7420
    },
    {
      "entropy": 5.935611581802368,
      "epoch": 0.6488683037664948,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004964474690028393,
      "loss": 5.6589,
      "mean_token_accuracy": 0.17266058325767517,
      "num_tokens": 15041709.0,
      "step": 7425
    },
    {
      "entropy": 5.943481826782227,
      "epoch": 0.6493052521191995,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004964419513859025,
      "loss": 5.7803,
      "mean_token_accuracy": 0.16661767214536666,
      "num_tokens": 15052277.0,
      "step": 7430
    },
    {
      "entropy": 5.9189001560211185,
      "epoch": 0.6497422004719042,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000496436429521593,
      "loss": 5.6859,
      "mean_token_accuracy": 0.166048763692379,
      "num_tokens": 15061870.0,
      "step": 7435
    },
    {
      "entropy": 5.850366067886353,
      "epoch": 0.6501791488246089,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004964309034100168,
      "loss": 5.5621,
      "mean_token_accuracy": 0.17796765118837357,
      "num_tokens": 15071403.0,
      "step": 7440
    },
    {
      "entropy": 5.903576469421386,
      "epoch": 0.6506160971773136,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004964253730512799,
      "loss": 5.6703,
      "mean_token_accuracy": 0.1732289806008339,
      "num_tokens": 15081651.0,
      "step": 7445
    },
    {
      "entropy": 5.978784513473511,
      "epoch": 0.6510530455300183,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004964198384454882,
      "loss": 5.7002,
      "mean_token_accuracy": 0.16599079668521882,
      "num_tokens": 15093403.0,
      "step": 7450
    },
    {
      "entropy": 5.939753103256225,
      "epoch": 0.6514899938827231,
      "grad_norm": 1.0,
      "learning_rate": 0.000496414299592748,
      "loss": 5.7754,
      "mean_token_accuracy": 0.15990479588508605,
      "num_tokens": 15103618.0,
      "step": 7455
    },
    {
      "entropy": 5.937116527557373,
      "epoch": 0.6519269422354278,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004964087564931656,
      "loss": 5.7058,
      "mean_token_accuracy": 0.17586928755044937,
      "num_tokens": 15113926.0,
      "step": 7460
    },
    {
      "entropy": 6.0514894962310795,
      "epoch": 0.6523638905881325,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004964032091468472,
      "loss": 5.7446,
      "mean_token_accuracy": 0.16631711274385452,
      "num_tokens": 15125302.0,
      "step": 7465
    },
    {
      "entropy": 5.933960723876953,
      "epoch": 0.6528008389408372,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004963976575538991,
      "loss": 5.6227,
      "mean_token_accuracy": 0.1753958597779274,
      "num_tokens": 15135943.0,
      "step": 7470
    },
    {
      "entropy": 5.900322437286377,
      "epoch": 0.6532377872935419,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000496392101714428,
      "loss": 5.6994,
      "mean_token_accuracy": 0.17226468324661254,
      "num_tokens": 15145001.0,
      "step": 7475
    },
    {
      "entropy": 5.93157172203064,
      "epoch": 0.6536747356462466,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004963865416285405,
      "loss": 5.6252,
      "mean_token_accuracy": 0.16903005838394164,
      "num_tokens": 15154567.0,
      "step": 7480
    },
    {
      "entropy": 5.998332357406616,
      "epoch": 0.6541116839989514,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004963809772963431,
      "loss": 5.7643,
      "mean_token_accuracy": 0.16783703714609147,
      "num_tokens": 15164311.0,
      "step": 7485
    },
    {
      "entropy": 5.967646741867066,
      "epoch": 0.6545486323516561,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004963754087179425,
      "loss": 5.6608,
      "mean_token_accuracy": 0.17073347121477128,
      "num_tokens": 15175217.0,
      "step": 7490
    },
    {
      "entropy": 5.957986211776733,
      "epoch": 0.6549855807043607,
      "grad_norm": 1.0,
      "learning_rate": 0.0004963698358934455,
      "loss": 5.6683,
      "mean_token_accuracy": 0.17260702401399614,
      "num_tokens": 15184895.0,
      "step": 7495
    },
    {
      "entropy": 5.86179404258728,
      "epoch": 0.6554225290570654,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004963642588229591,
      "loss": 5.6567,
      "mean_token_accuracy": 0.16766242384910585,
      "num_tokens": 15194380.0,
      "step": 7500
    },
    {
      "entropy": 5.947136259078979,
      "epoch": 0.6558594774097701,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004963586775065902,
      "loss": 5.6963,
      "mean_token_accuracy": 0.16782759875059128,
      "num_tokens": 15204364.0,
      "step": 7505
    },
    {
      "entropy": 5.937696647644043,
      "epoch": 0.6562964257624748,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004963530919444459,
      "loss": 5.6485,
      "mean_token_accuracy": 0.17060399204492568,
      "num_tokens": 15214485.0,
      "step": 7510
    },
    {
      "entropy": 5.981035280227661,
      "epoch": 0.6567333741151796,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004963475021366333,
      "loss": 5.7784,
      "mean_token_accuracy": 0.16234237998723983,
      "num_tokens": 15224764.0,
      "step": 7515
    },
    {
      "entropy": 5.983535242080689,
      "epoch": 0.6571703224678843,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004963419080832597,
      "loss": 5.7762,
      "mean_token_accuracy": 0.16502948254346847,
      "num_tokens": 15235481.0,
      "step": 7520
    },
    {
      "entropy": 6.034067535400391,
      "epoch": 0.657607270820589,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004963363097844323,
      "loss": 5.7452,
      "mean_token_accuracy": 0.16899672597646714,
      "num_tokens": 15245332.0,
      "step": 7525
    },
    {
      "entropy": 6.0035747528076175,
      "epoch": 0.6580442191732937,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004963307072402584,
      "loss": 5.7417,
      "mean_token_accuracy": 0.16613221317529678,
      "num_tokens": 15255562.0,
      "step": 7530
    },
    {
      "entropy": 5.949329566955567,
      "epoch": 0.6584811675259984,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004963251004508456,
      "loss": 5.6784,
      "mean_token_accuracy": 0.1677081897854805,
      "num_tokens": 15265655.0,
      "step": 7535
    },
    {
      "entropy": 5.862000465393066,
      "epoch": 0.6589181158787032,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004963194894163015,
      "loss": 5.5961,
      "mean_token_accuracy": 0.17913943380117417,
      "num_tokens": 15276311.0,
      "step": 7540
    },
    {
      "entropy": 5.941941213607788,
      "epoch": 0.6593550642314079,
      "grad_norm": 1.0,
      "learning_rate": 0.0004963138741367334,
      "loss": 5.757,
      "mean_token_accuracy": 0.1678944393992424,
      "num_tokens": 15287142.0,
      "step": 7545
    },
    {
      "entropy": 6.031400728225708,
      "epoch": 0.6597920125841126,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004963082546122495,
      "loss": 5.7333,
      "mean_token_accuracy": 0.16926798075437546,
      "num_tokens": 15297484.0,
      "step": 7550
    },
    {
      "entropy": 5.958199119567871,
      "epoch": 0.6602289609368173,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004963026308429571,
      "loss": 5.6856,
      "mean_token_accuracy": 0.171526762843132,
      "num_tokens": 15307468.0,
      "step": 7555
    },
    {
      "entropy": 5.860775756835937,
      "epoch": 0.660665909289522,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004962970028289644,
      "loss": 5.6564,
      "mean_token_accuracy": 0.17619750499725342,
      "num_tokens": 15318487.0,
      "step": 7560
    },
    {
      "entropy": 5.932335996627808,
      "epoch": 0.6611028576422266,
      "grad_norm": 1.0,
      "learning_rate": 0.0004962913705703791,
      "loss": 5.7041,
      "mean_token_accuracy": 0.1643641322851181,
      "num_tokens": 15329193.0,
      "step": 7565
    },
    {
      "entropy": 5.949157190322876,
      "epoch": 0.6615398059949315,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004962857340673095,
      "loss": 5.6538,
      "mean_token_accuracy": 0.17204876393079757,
      "num_tokens": 15338886.0,
      "step": 7570
    },
    {
      "entropy": 5.874380826950073,
      "epoch": 0.6619767543476361,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004962800933198635,
      "loss": 5.5759,
      "mean_token_accuracy": 0.18257878422737123,
      "num_tokens": 15348629.0,
      "step": 7575
    },
    {
      "entropy": 5.89663143157959,
      "epoch": 0.6624137027003408,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004962744483281495,
      "loss": 5.6955,
      "mean_token_accuracy": 0.17697014957666396,
      "num_tokens": 15358004.0,
      "step": 7580
    },
    {
      "entropy": 5.944311094284058,
      "epoch": 0.6628506510530455,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004962687990922754,
      "loss": 5.6849,
      "mean_token_accuracy": 0.16907844245433806,
      "num_tokens": 15368018.0,
      "step": 7585
    },
    {
      "entropy": 6.0428955078125,
      "epoch": 0.6632875994057502,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00049626314561235,
      "loss": 5.6755,
      "mean_token_accuracy": 0.1675059363245964,
      "num_tokens": 15377443.0,
      "step": 7590
    },
    {
      "entropy": 5.944485425949097,
      "epoch": 0.6637245477584549,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004962574878884814,
      "loss": 5.6928,
      "mean_token_accuracy": 0.16727231740951537,
      "num_tokens": 15388278.0,
      "step": 7595
    },
    {
      "entropy": 5.911031484603882,
      "epoch": 0.6641614961111597,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004962518259207782,
      "loss": 5.6994,
      "mean_token_accuracy": 0.16796059906482697,
      "num_tokens": 15398619.0,
      "step": 7600
    },
    {
      "entropy": 5.881869792938232,
      "epoch": 0.6645984444638644,
      "grad_norm": 1.0,
      "learning_rate": 0.0004962461597093491,
      "loss": 5.5908,
      "mean_token_accuracy": 0.17623851746320723,
      "num_tokens": 15408635.0,
      "step": 7605
    },
    {
      "entropy": 5.983315372467041,
      "epoch": 0.6650353928165691,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004962404892543028,
      "loss": 5.7113,
      "mean_token_accuracy": 0.1645165652036667,
      "num_tokens": 15419113.0,
      "step": 7610
    },
    {
      "entropy": 6.027986001968384,
      "epoch": 0.6654723411692738,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000496234814555748,
      "loss": 5.6775,
      "mean_token_accuracy": 0.16941139101982117,
      "num_tokens": 15428959.0,
      "step": 7615
    },
    {
      "entropy": 5.916666793823242,
      "epoch": 0.6659092895219785,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004962291356137934,
      "loss": 5.7532,
      "mean_token_accuracy": 0.16224540323019027,
      "num_tokens": 15439116.0,
      "step": 7620
    },
    {
      "entropy": 5.922262525558471,
      "epoch": 0.6663462378746832,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004962234524285481,
      "loss": 5.7504,
      "mean_token_accuracy": 0.17196505665779113,
      "num_tokens": 15449715.0,
      "step": 7625
    },
    {
      "entropy": 5.915909910202027,
      "epoch": 0.666783186227388,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004962177650001212,
      "loss": 5.5154,
      "mean_token_accuracy": 0.18639762252569197,
      "num_tokens": 15459170.0,
      "step": 7630
    },
    {
      "entropy": 5.846242237091064,
      "epoch": 0.6672201345800927,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004962120733286215,
      "loss": 5.6308,
      "mean_token_accuracy": 0.17514614760875702,
      "num_tokens": 15469497.0,
      "step": 7635
    },
    {
      "entropy": 5.891697072982788,
      "epoch": 0.6676570829327974,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004962063774141583,
      "loss": 5.769,
      "mean_token_accuracy": 0.16937299966812133,
      "num_tokens": 15481298.0,
      "step": 7640
    },
    {
      "entropy": 6.009493207931518,
      "epoch": 0.668094031285502,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004962006772568409,
      "loss": 5.7027,
      "mean_token_accuracy": 0.17172098457813262,
      "num_tokens": 15491696.0,
      "step": 7645
    },
    {
      "entropy": 5.985794448852539,
      "epoch": 0.6685309796382067,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004961949728567787,
      "loss": 5.7148,
      "mean_token_accuracy": 0.16959842443466186,
      "num_tokens": 15502412.0,
      "step": 7650
    },
    {
      "entropy": 5.902047252655029,
      "epoch": 0.6689679279909114,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004961892642140809,
      "loss": 5.7049,
      "mean_token_accuracy": 0.16745323985815047,
      "num_tokens": 15513436.0,
      "step": 7655
    },
    {
      "entropy": 5.9506182193756105,
      "epoch": 0.6694048763436162,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004961835513288572,
      "loss": 5.7029,
      "mean_token_accuracy": 0.17352998554706572,
      "num_tokens": 15524564.0,
      "step": 7660
    },
    {
      "entropy": 5.939547157287597,
      "epoch": 0.6698418246963209,
      "grad_norm": 0.953125,
      "learning_rate": 0.000496177834201217,
      "loss": 5.6825,
      "mean_token_accuracy": 0.16373797804117202,
      "num_tokens": 15534539.0,
      "step": 7665
    },
    {
      "entropy": 5.948321199417114,
      "epoch": 0.6702787730490256,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004961721128312702,
      "loss": 5.7454,
      "mean_token_accuracy": 0.15733132809400557,
      "num_tokens": 15544571.0,
      "step": 7670
    },
    {
      "entropy": 5.991916990280151,
      "epoch": 0.6707157214017303,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004961663872191263,
      "loss": 5.6703,
      "mean_token_accuracy": 0.1792220413684845,
      "num_tokens": 15555066.0,
      "step": 7675
    },
    {
      "entropy": 5.939787483215332,
      "epoch": 0.671152669754435,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004961606573648951,
      "loss": 5.6977,
      "mean_token_accuracy": 0.17095106095075607,
      "num_tokens": 15565128.0,
      "step": 7680
    },
    {
      "entropy": 5.916919469833374,
      "epoch": 0.6715896181071397,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004961549232686867,
      "loss": 5.7534,
      "mean_token_accuracy": 0.16654478758573532,
      "num_tokens": 15576394.0,
      "step": 7685
    },
    {
      "entropy": 6.020664215087891,
      "epoch": 0.6720265664598445,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000496149184930611,
      "loss": 5.686,
      "mean_token_accuracy": 0.16511461958289148,
      "num_tokens": 15585806.0,
      "step": 7690
    },
    {
      "entropy": 5.953474760055542,
      "epoch": 0.6724635148125492,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004961434423507779,
      "loss": 5.6729,
      "mean_token_accuracy": 0.17099466174840927,
      "num_tokens": 15596471.0,
      "step": 7695
    },
    {
      "entropy": 5.897314214706421,
      "epoch": 0.6729004631652539,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004961376955292979,
      "loss": 5.5755,
      "mean_token_accuracy": 0.1859306275844574,
      "num_tokens": 15605720.0,
      "step": 7700
    },
    {
      "entropy": 5.905326318740845,
      "epoch": 0.6733374115179586,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004961319444662809,
      "loss": 5.6635,
      "mean_token_accuracy": 0.1735016033053398,
      "num_tokens": 15616167.0,
      "step": 7705
    },
    {
      "entropy": 6.010176801681519,
      "epoch": 0.6737743598706633,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004961261891618374,
      "loss": 5.7514,
      "mean_token_accuracy": 0.16001667827367783,
      "num_tokens": 15627033.0,
      "step": 7710
    },
    {
      "entropy": 5.947105360031128,
      "epoch": 0.6742113082233679,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004961204296160777,
      "loss": 5.6974,
      "mean_token_accuracy": 0.175602974742651,
      "num_tokens": 15637460.0,
      "step": 7715
    },
    {
      "entropy": 5.931848621368408,
      "epoch": 0.6746482565760727,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004961146658291123,
      "loss": 5.627,
      "mean_token_accuracy": 0.17534350454807282,
      "num_tokens": 15646027.0,
      "step": 7720
    },
    {
      "entropy": 5.9268574714660645,
      "epoch": 0.6750852049287774,
      "grad_norm": 1.0,
      "learning_rate": 0.0004961088978010518,
      "loss": 5.6565,
      "mean_token_accuracy": 0.1713066041469574,
      "num_tokens": 15655241.0,
      "step": 7725
    },
    {
      "entropy": 5.906084823608398,
      "epoch": 0.6755221532814821,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004961031255320067,
      "loss": 5.6063,
      "mean_token_accuracy": 0.17116497904062272,
      "num_tokens": 15666213.0,
      "step": 7730
    },
    {
      "entropy": 6.020251226425171,
      "epoch": 0.6759591016341868,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004960973490220878,
      "loss": 5.6969,
      "mean_token_accuracy": 0.1677973374724388,
      "num_tokens": 15676985.0,
      "step": 7735
    },
    {
      "entropy": 5.963055992126465,
      "epoch": 0.6763960499868915,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004960915682714059,
      "loss": 5.7262,
      "mean_token_accuracy": 0.1648624762892723,
      "num_tokens": 15687519.0,
      "step": 7740
    },
    {
      "entropy": 5.9893025875091555,
      "epoch": 0.6768329983395963,
      "grad_norm": 1.0,
      "learning_rate": 0.0004960857832800719,
      "loss": 5.6977,
      "mean_token_accuracy": 0.16722078174352645,
      "num_tokens": 15697401.0,
      "step": 7745
    },
    {
      "entropy": 5.940856695175171,
      "epoch": 0.677269946692301,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004960799940481968,
      "loss": 5.7193,
      "mean_token_accuracy": 0.16715561598539352,
      "num_tokens": 15708159.0,
      "step": 7750
    },
    {
      "entropy": 5.927948236465454,
      "epoch": 0.6777068950450057,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004960742005758914,
      "loss": 5.6677,
      "mean_token_accuracy": 0.16917172819375992,
      "num_tokens": 15717749.0,
      "step": 7755
    },
    {
      "entropy": 5.910120582580566,
      "epoch": 0.6781438433977104,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004960684028632671,
      "loss": 5.6168,
      "mean_token_accuracy": 0.173377887904644,
      "num_tokens": 15726868.0,
      "step": 7760
    },
    {
      "entropy": 5.937097024917603,
      "epoch": 0.6785807917504151,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004960626009104349,
      "loss": 5.6403,
      "mean_token_accuracy": 0.17782494723796843,
      "num_tokens": 15736038.0,
      "step": 7765
    },
    {
      "entropy": 5.91299467086792,
      "epoch": 0.6790177401031198,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004960567947175064,
      "loss": 5.6715,
      "mean_token_accuracy": 0.16713650077581405,
      "num_tokens": 15746676.0,
      "step": 7770
    },
    {
      "entropy": 5.99061861038208,
      "epoch": 0.6794546884558246,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004960509842845926,
      "loss": 5.6517,
      "mean_token_accuracy": 0.17163523137569428,
      "num_tokens": 15756064.0,
      "step": 7775
    },
    {
      "entropy": 5.933780717849731,
      "epoch": 0.6798916368085293,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004960451696118049,
      "loss": 5.6606,
      "mean_token_accuracy": 0.17397007197141648,
      "num_tokens": 15767414.0,
      "step": 7780
    },
    {
      "entropy": 5.949579191207886,
      "epoch": 0.680328585161234,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004960393506992553,
      "loss": 5.6382,
      "mean_token_accuracy": 0.1722015157341957,
      "num_tokens": 15778276.0,
      "step": 7785
    },
    {
      "entropy": 5.876014947891235,
      "epoch": 0.6807655335139386,
      "grad_norm": 0.875,
      "learning_rate": 0.0004960335275470551,
      "loss": 5.5784,
      "mean_token_accuracy": 0.16979634165763854,
      "num_tokens": 15788365.0,
      "step": 7790
    },
    {
      "entropy": 5.920602130889892,
      "epoch": 0.6812024818666433,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004960277001553159,
      "loss": 5.5839,
      "mean_token_accuracy": 0.17887528389692306,
      "num_tokens": 15799021.0,
      "step": 7795
    },
    {
      "entropy": 5.845925426483154,
      "epoch": 0.681639430219348,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004960218685241498,
      "loss": 5.6656,
      "mean_token_accuracy": 0.171703739464283,
      "num_tokens": 15808779.0,
      "step": 7800
    },
    {
      "entropy": 5.965919542312622,
      "epoch": 0.6820763785720528,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004960160326536681,
      "loss": 5.6766,
      "mean_token_accuracy": 0.1737473264336586,
      "num_tokens": 15818515.0,
      "step": 7805
    },
    {
      "entropy": 5.957288694381714,
      "epoch": 0.6825133269247575,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004960101925439834,
      "loss": 5.6363,
      "mean_token_accuracy": 0.1736165091395378,
      "num_tokens": 15828843.0,
      "step": 7810
    },
    {
      "entropy": 5.9391459941864015,
      "epoch": 0.6829502752774622,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004960043481952072,
      "loss": 5.6274,
      "mean_token_accuracy": 0.17138408571481706,
      "num_tokens": 15839323.0,
      "step": 7815
    },
    {
      "entropy": 5.899854946136474,
      "epoch": 0.6833872236301669,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000495998499607452,
      "loss": 5.6801,
      "mean_token_accuracy": 0.16953393816947937,
      "num_tokens": 15848399.0,
      "step": 7820
    },
    {
      "entropy": 5.90320987701416,
      "epoch": 0.6838241719828716,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004959926467808296,
      "loss": 5.554,
      "mean_token_accuracy": 0.1736243635416031,
      "num_tokens": 15858137.0,
      "step": 7825
    },
    {
      "entropy": 5.9101379871368405,
      "epoch": 0.6842611203355763,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004959867897154526,
      "loss": 5.649,
      "mean_token_accuracy": 0.17085463255643846,
      "num_tokens": 15867643.0,
      "step": 7830
    },
    {
      "entropy": 5.9303202629089355,
      "epoch": 0.6846980686882811,
      "grad_norm": 0.9375,
      "learning_rate": 0.000495980928411433,
      "loss": 5.5467,
      "mean_token_accuracy": 0.17508207708597184,
      "num_tokens": 15877523.0,
      "step": 7835
    },
    {
      "entropy": 5.937829542160034,
      "epoch": 0.6851350170409858,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004959750628688834,
      "loss": 5.7398,
      "mean_token_accuracy": 0.16561159044504165,
      "num_tokens": 15886946.0,
      "step": 7840
    },
    {
      "entropy": 5.90155553817749,
      "epoch": 0.6855719653936905,
      "grad_norm": 1.0,
      "learning_rate": 0.0004959691930879164,
      "loss": 5.5818,
      "mean_token_accuracy": 0.17879643887281418,
      "num_tokens": 15896646.0,
      "step": 7845
    },
    {
      "entropy": 5.894581508636475,
      "epoch": 0.6860089137463952,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004959633190686445,
      "loss": 5.6131,
      "mean_token_accuracy": 0.17592983990907668,
      "num_tokens": 15907204.0,
      "step": 7850
    },
    {
      "entropy": 5.909151601791382,
      "epoch": 0.6864458620990999,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004959574408111804,
      "loss": 5.6697,
      "mean_token_accuracy": 0.17775077223777772,
      "num_tokens": 15917634.0,
      "step": 7855
    },
    {
      "entropy": 5.90779185295105,
      "epoch": 0.6868828104518045,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004959515583156367,
      "loss": 5.6255,
      "mean_token_accuracy": 0.1735110491514206,
      "num_tokens": 15926584.0,
      "step": 7860
    },
    {
      "entropy": 5.917984056472778,
      "epoch": 0.6873197588045094,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004959456715821264,
      "loss": 5.6056,
      "mean_token_accuracy": 0.17545076012611388,
      "num_tokens": 15936636.0,
      "step": 7865
    },
    {
      "entropy": 5.912478017807007,
      "epoch": 0.687756707157214,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004959397806107624,
      "loss": 5.6205,
      "mean_token_accuracy": 0.18108900636434555,
      "num_tokens": 15946113.0,
      "step": 7870
    },
    {
      "entropy": 5.973181533813476,
      "epoch": 0.6881936555099187,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004959338854016574,
      "loss": 5.6748,
      "mean_token_accuracy": 0.16875917911529542,
      "num_tokens": 15956630.0,
      "step": 7875
    },
    {
      "entropy": 5.886916780471802,
      "epoch": 0.6886306038626234,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004959279859549249,
      "loss": 5.5946,
      "mean_token_accuracy": 0.1790658324956894,
      "num_tokens": 15967129.0,
      "step": 7880
    },
    {
      "entropy": 5.902432584762574,
      "epoch": 0.6890675522153281,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004959220822706779,
      "loss": 5.5954,
      "mean_token_accuracy": 0.17571737766265869,
      "num_tokens": 15976663.0,
      "step": 7885
    },
    {
      "entropy": 5.849058389663696,
      "epoch": 0.6895045005680328,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004959161743490295,
      "loss": 5.6099,
      "mean_token_accuracy": 0.17869622856378556,
      "num_tokens": 15986666.0,
      "step": 7890
    },
    {
      "entropy": 5.850893115997314,
      "epoch": 0.6899414489207376,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004959102621900931,
      "loss": 5.5497,
      "mean_token_accuracy": 0.18442929536104202,
      "num_tokens": 15996233.0,
      "step": 7895
    },
    {
      "entropy": 5.972593116760254,
      "epoch": 0.6903783972734423,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004959043457939822,
      "loss": 5.7351,
      "mean_token_accuracy": 0.16863043904304503,
      "num_tokens": 16008188.0,
      "step": 7900
    },
    {
      "entropy": 5.903003835678101,
      "epoch": 0.690815345626147,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004958984251608101,
      "loss": 5.6334,
      "mean_token_accuracy": 0.17173693776130677,
      "num_tokens": 16018494.0,
      "step": 7905
    },
    {
      "entropy": 5.8806493282318115,
      "epoch": 0.6912522939788517,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004958925002906905,
      "loss": 5.6573,
      "mean_token_accuracy": 0.17597612589597703,
      "num_tokens": 16028531.0,
      "step": 7910
    },
    {
      "entropy": 5.965801620483399,
      "epoch": 0.6916892423315564,
      "grad_norm": 0.9375,
      "learning_rate": 0.000495886571183737,
      "loss": 5.6328,
      "mean_token_accuracy": 0.17182032614946366,
      "num_tokens": 16039050.0,
      "step": 7915
    },
    {
      "entropy": 5.935761642456055,
      "epoch": 0.6921261906842611,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004958806378400634,
      "loss": 5.7285,
      "mean_token_accuracy": 0.16290564388036727,
      "num_tokens": 16050751.0,
      "step": 7920
    },
    {
      "entropy": 5.942456531524658,
      "epoch": 0.6925631390369659,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004958747002597834,
      "loss": 5.6521,
      "mean_token_accuracy": 0.1673956736922264,
      "num_tokens": 16060758.0,
      "step": 7925
    },
    {
      "entropy": 5.9601507663726805,
      "epoch": 0.6930000873896706,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004958687584430109,
      "loss": 5.6568,
      "mean_token_accuracy": 0.1696046158671379,
      "num_tokens": 16070852.0,
      "step": 7930
    },
    {
      "entropy": 5.8889611721038815,
      "epoch": 0.6934370357423753,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004958628123898597,
      "loss": 5.5813,
      "mean_token_accuracy": 0.1747801423072815,
      "num_tokens": 16080081.0,
      "step": 7935
    },
    {
      "entropy": 5.875192928314209,
      "epoch": 0.6938739840950799,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004958568621004442,
      "loss": 5.615,
      "mean_token_accuracy": 0.17415339052677153,
      "num_tokens": 16090483.0,
      "step": 7940
    },
    {
      "entropy": 6.004095411300659,
      "epoch": 0.6943109324477846,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004958509075748785,
      "loss": 5.7047,
      "mean_token_accuracy": 0.16883806586265565,
      "num_tokens": 16101136.0,
      "step": 7945
    },
    {
      "entropy": 5.914346313476562,
      "epoch": 0.6947478808004894,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004958449488132764,
      "loss": 5.6001,
      "mean_token_accuracy": 0.17818938493728637,
      "num_tokens": 16110124.0,
      "step": 7950
    },
    {
      "entropy": 5.850813484191894,
      "epoch": 0.6951848291531941,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004958389858157526,
      "loss": 5.5518,
      "mean_token_accuracy": 0.17802151739597322,
      "num_tokens": 16119581.0,
      "step": 7955
    },
    {
      "entropy": 5.880281066894531,
      "epoch": 0.6956217775058988,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004958330185824213,
      "loss": 5.6644,
      "mean_token_accuracy": 0.17115467935800552,
      "num_tokens": 16130119.0,
      "step": 7960
    },
    {
      "entropy": 5.9511512279510494,
      "epoch": 0.6960587258586035,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000495827047113397,
      "loss": 5.6963,
      "mean_token_accuracy": 0.16921136230230333,
      "num_tokens": 16140702.0,
      "step": 7965
    },
    {
      "entropy": 5.862891721725464,
      "epoch": 0.6964956742113082,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004958210714087942,
      "loss": 5.6765,
      "mean_token_accuracy": 0.1721287861466408,
      "num_tokens": 16151107.0,
      "step": 7970
    },
    {
      "entropy": 5.8997307300567625,
      "epoch": 0.6969326225640129,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004958150914687276,
      "loss": 5.5932,
      "mean_token_accuracy": 0.1755025014281273,
      "num_tokens": 16160779.0,
      "step": 7975
    },
    {
      "entropy": 5.851053857803345,
      "epoch": 0.6973695709167177,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004958091072933118,
      "loss": 5.5346,
      "mean_token_accuracy": 0.1791098326444626,
      "num_tokens": 16169624.0,
      "step": 7980
    },
    {
      "entropy": 5.820342493057251,
      "epoch": 0.6978065192694224,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004958031188826616,
      "loss": 5.5488,
      "mean_token_accuracy": 0.1850172221660614,
      "num_tokens": 16179192.0,
      "step": 7985
    },
    {
      "entropy": 5.849523830413818,
      "epoch": 0.6982434676221271,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004957971262368919,
      "loss": 5.6311,
      "mean_token_accuracy": 0.17664679437875747,
      "num_tokens": 16188847.0,
      "step": 7990
    },
    {
      "entropy": 5.988716077804566,
      "epoch": 0.6986804159748318,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004957911293561176,
      "loss": 5.7399,
      "mean_token_accuracy": 0.1690984234213829,
      "num_tokens": 16199480.0,
      "step": 7995
    },
    {
      "entropy": 5.936985635757447,
      "epoch": 0.6991173643275365,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004957851282404538,
      "loss": 5.6111,
      "mean_token_accuracy": 0.18212860822677612,
      "num_tokens": 16208818.0,
      "step": 8000
    },
    {
      "entropy": 5.927171945571899,
      "epoch": 0.6995543126802412,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004957791228900154,
      "loss": 5.6005,
      "mean_token_accuracy": 0.17574614137411118,
      "num_tokens": 16219298.0,
      "step": 8005
    },
    {
      "entropy": 5.884048509597778,
      "epoch": 0.699991261032946,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004957731133049179,
      "loss": 5.5979,
      "mean_token_accuracy": 0.18096419274806977,
      "num_tokens": 16228170.0,
      "step": 8010
    },
    {
      "entropy": 5.942883157730103,
      "epoch": 0.7004282093856506,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004957670994852764,
      "loss": 5.6787,
      "mean_token_accuracy": 0.16906016170978547,
      "num_tokens": 16237552.0,
      "step": 8015
    },
    {
      "entropy": 5.971673059463501,
      "epoch": 0.7008651577383553,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004957610814312062,
      "loss": 5.7029,
      "mean_token_accuracy": 0.16702363938093184,
      "num_tokens": 16248157.0,
      "step": 8020
    },
    {
      "entropy": 5.931836748123169,
      "epoch": 0.70130210609106,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004957550591428229,
      "loss": 5.5927,
      "mean_token_accuracy": 0.177716201543808,
      "num_tokens": 16258266.0,
      "step": 8025
    },
    {
      "entropy": 5.864434099197387,
      "epoch": 0.7017390544437647,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004957490326202418,
      "loss": 5.6819,
      "mean_token_accuracy": 0.1753667786717415,
      "num_tokens": 16268769.0,
      "step": 8030
    },
    {
      "entropy": 5.961971092224121,
      "epoch": 0.7021760027964694,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004957430018635785,
      "loss": 5.6512,
      "mean_token_accuracy": 0.17281595766544341,
      "num_tokens": 16279688.0,
      "step": 8035
    },
    {
      "entropy": 5.919993829727173,
      "epoch": 0.7026129511491742,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004957369668729488,
      "loss": 5.6286,
      "mean_token_accuracy": 0.17452212870121003,
      "num_tokens": 16289441.0,
      "step": 8040
    },
    {
      "entropy": 5.9076621532440186,
      "epoch": 0.7030498995018789,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004957309276484685,
      "loss": 5.6097,
      "mean_token_accuracy": 0.17459991723299026,
      "num_tokens": 16299802.0,
      "step": 8045
    },
    {
      "entropy": 5.868761157989502,
      "epoch": 0.7034868478545836,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004957248841902533,
      "loss": 5.5833,
      "mean_token_accuracy": 0.18094270527362824,
      "num_tokens": 16309577.0,
      "step": 8050
    },
    {
      "entropy": 5.880494499206543,
      "epoch": 0.7039237962072883,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004957188364984191,
      "loss": 5.5403,
      "mean_token_accuracy": 0.1820886641740799,
      "num_tokens": 16319205.0,
      "step": 8055
    },
    {
      "entropy": 5.81786961555481,
      "epoch": 0.704360744559993,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000495712784573082,
      "loss": 5.6241,
      "mean_token_accuracy": 0.17667629718780517,
      "num_tokens": 16329650.0,
      "step": 8060
    },
    {
      "entropy": 5.878563117980957,
      "epoch": 0.7047976929126977,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004957067284143581,
      "loss": 5.6447,
      "mean_token_accuracy": 0.17422914654016494,
      "num_tokens": 16339242.0,
      "step": 8065
    },
    {
      "entropy": 5.950465154647827,
      "epoch": 0.7052346412654025,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004957006680223634,
      "loss": 5.5997,
      "mean_token_accuracy": 0.18178627043962478,
      "num_tokens": 16348441.0,
      "step": 8070
    },
    {
      "entropy": 5.880155563354492,
      "epoch": 0.7056715896181072,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004956946033972143,
      "loss": 5.6061,
      "mean_token_accuracy": 0.17598044872283936,
      "num_tokens": 16357425.0,
      "step": 8075
    },
    {
      "entropy": 5.852259731292724,
      "epoch": 0.7061085379708119,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000495688534539027,
      "loss": 5.5781,
      "mean_token_accuracy": 0.1723771497607231,
      "num_tokens": 16366180.0,
      "step": 8080
    },
    {
      "entropy": 5.893771171569824,
      "epoch": 0.7065454863235165,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004956824614479181,
      "loss": 5.6433,
      "mean_token_accuracy": 0.172347816824913,
      "num_tokens": 16375714.0,
      "step": 8085
    },
    {
      "entropy": 5.985397624969482,
      "epoch": 0.7069824346762212,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004956763841240038,
      "loss": 5.6232,
      "mean_token_accuracy": 0.17065414637327195,
      "num_tokens": 16386051.0,
      "step": 8090
    },
    {
      "entropy": 5.92710599899292,
      "epoch": 0.7074193830289259,
      "grad_norm": 0.96875,
      "learning_rate": 0.000495670302567401,
      "loss": 5.6816,
      "mean_token_accuracy": 0.17041144222021104,
      "num_tokens": 16396616.0,
      "step": 8095
    },
    {
      "entropy": 5.879296970367432,
      "epoch": 0.7078563313816307,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000495664216778226,
      "loss": 5.6091,
      "mean_token_accuracy": 0.17852067649364473,
      "num_tokens": 16406301.0,
      "step": 8100
    },
    {
      "entropy": 5.920216417312622,
      "epoch": 0.7082932797343354,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004956581267565957,
      "loss": 5.6714,
      "mean_token_accuracy": 0.16531501710414886,
      "num_tokens": 16417107.0,
      "step": 8105
    },
    {
      "entropy": 5.9186375617980955,
      "epoch": 0.7087302280870401,
      "grad_norm": 0.953125,
      "learning_rate": 0.000495652032502627,
      "loss": 5.5944,
      "mean_token_accuracy": 0.17682256698608398,
      "num_tokens": 16427292.0,
      "step": 8110
    },
    {
      "entropy": 5.885567665100098,
      "epoch": 0.7091671764397448,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004956459340164366,
      "loss": 5.6097,
      "mean_token_accuracy": 0.1808635488152504,
      "num_tokens": 16438191.0,
      "step": 8115
    },
    {
      "entropy": 5.8984901905059814,
      "epoch": 0.7096041247924495,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004956398312981417,
      "loss": 5.5892,
      "mean_token_accuracy": 0.17588261216878892,
      "num_tokens": 16448237.0,
      "step": 8120
    },
    {
      "entropy": 5.882715034484863,
      "epoch": 0.7100410731451542,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004956337243478592,
      "loss": 5.6659,
      "mean_token_accuracy": 0.1711867243051529,
      "num_tokens": 16458947.0,
      "step": 8125
    },
    {
      "entropy": 5.985532426834107,
      "epoch": 0.710478021497859,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004956276131657061,
      "loss": 5.6091,
      "mean_token_accuracy": 0.17293916493654252,
      "num_tokens": 16467464.0,
      "step": 8130
    },
    {
      "entropy": 5.91393346786499,
      "epoch": 0.7109149698505637,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004956214977518,
      "loss": 5.6168,
      "mean_token_accuracy": 0.1807863384485245,
      "num_tokens": 16477747.0,
      "step": 8135
    },
    {
      "entropy": 5.926320266723633,
      "epoch": 0.7113519182032684,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004956153781062578,
      "loss": 5.731,
      "mean_token_accuracy": 0.17039488404989242,
      "num_tokens": 16488111.0,
      "step": 8140
    },
    {
      "entropy": 5.937343692779541,
      "epoch": 0.7117888665559731,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004956092542291971,
      "loss": 5.607,
      "mean_token_accuracy": 0.17745495587587357,
      "num_tokens": 16497689.0,
      "step": 8145
    },
    {
      "entropy": 5.913807106018067,
      "epoch": 0.7122258149086778,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004956031261207353,
      "loss": 5.7265,
      "mean_token_accuracy": 0.17039292305707932,
      "num_tokens": 16508515.0,
      "step": 8150
    },
    {
      "entropy": 5.805982160568237,
      "epoch": 0.7126627632613826,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00049559699378099,
      "loss": 5.6153,
      "mean_token_accuracy": 0.18247250467538834,
      "num_tokens": 16519184.0,
      "step": 8155
    },
    {
      "entropy": 5.9372594356536865,
      "epoch": 0.7130997116140873,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004955908572100787,
      "loss": 5.7075,
      "mean_token_accuracy": 0.16851256489753724,
      "num_tokens": 16529917.0,
      "step": 8160
    },
    {
      "entropy": 5.95330228805542,
      "epoch": 0.7135366599667919,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004955847164081192,
      "loss": 5.5945,
      "mean_token_accuracy": 0.17610006034374237,
      "num_tokens": 16539687.0,
      "step": 8165
    },
    {
      "entropy": 5.91069450378418,
      "epoch": 0.7139736083194966,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004955785713752292,
      "loss": 5.6865,
      "mean_token_accuracy": 0.17063601911067963,
      "num_tokens": 16550273.0,
      "step": 8170
    },
    {
      "entropy": 5.97544231414795,
      "epoch": 0.7144105566722013,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004955724221115267,
      "loss": 5.7525,
      "mean_token_accuracy": 0.165391306579113,
      "num_tokens": 16562215.0,
      "step": 8175
    },
    {
      "entropy": 5.991822910308838,
      "epoch": 0.714847505024906,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004955662686171295,
      "loss": 5.6988,
      "mean_token_accuracy": 0.17201162576675416,
      "num_tokens": 16573218.0,
      "step": 8180
    },
    {
      "entropy": 5.9940324306488035,
      "epoch": 0.7152844533776108,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004955601108921557,
      "loss": 5.6744,
      "mean_token_accuracy": 0.1693201407790184,
      "num_tokens": 16582740.0,
      "step": 8185
    },
    {
      "entropy": 5.869105625152588,
      "epoch": 0.7157214017303155,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004955539489367235,
      "loss": 5.626,
      "mean_token_accuracy": 0.1832931935787201,
      "num_tokens": 16592169.0,
      "step": 8190
    },
    {
      "entropy": 5.937628555297851,
      "epoch": 0.7161583500830202,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004955477827509509,
      "loss": 5.615,
      "mean_token_accuracy": 0.17084167748689652,
      "num_tokens": 16602706.0,
      "step": 8195
    },
    {
      "entropy": 5.940625619888306,
      "epoch": 0.7165952984357249,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004955416123349562,
      "loss": 5.5815,
      "mean_token_accuracy": 0.17855869084596634,
      "num_tokens": 16612012.0,
      "step": 8200
    },
    {
      "entropy": 5.900930786132813,
      "epoch": 0.7170322467884296,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000495535437688858,
      "loss": 5.7194,
      "mean_token_accuracy": 0.16814900636672975,
      "num_tokens": 16621798.0,
      "step": 8205
    },
    {
      "entropy": 5.927228307723999,
      "epoch": 0.7174691951411343,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004955292588127744,
      "loss": 5.6273,
      "mean_token_accuracy": 0.17556482404470444,
      "num_tokens": 16631792.0,
      "step": 8210
    },
    {
      "entropy": 5.928174352645874,
      "epoch": 0.7179061434938391,
      "grad_norm": 0.984375,
      "learning_rate": 0.000495523075706824,
      "loss": 5.5777,
      "mean_token_accuracy": 0.1742174431681633,
      "num_tokens": 16641655.0,
      "step": 8215
    },
    {
      "entropy": 5.926209831237793,
      "epoch": 0.7183430918465438,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004955168883711257,
      "loss": 5.6724,
      "mean_token_accuracy": 0.16749219596385956,
      "num_tokens": 16651210.0,
      "step": 8220
    },
    {
      "entropy": 5.984452772140503,
      "epoch": 0.7187800401992485,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004955106968057977,
      "loss": 5.6389,
      "mean_token_accuracy": 0.17034388184547425,
      "num_tokens": 16660078.0,
      "step": 8225
    },
    {
      "entropy": 5.932273578643799,
      "epoch": 0.7192169885519532,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004955045010109592,
      "loss": 5.6468,
      "mean_token_accuracy": 0.17081798613071442,
      "num_tokens": 16671211.0,
      "step": 8230
    },
    {
      "entropy": 5.908595132827759,
      "epoch": 0.7196539369046578,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004954983009867286,
      "loss": 5.5834,
      "mean_token_accuracy": 0.1770865201950073,
      "num_tokens": 16680850.0,
      "step": 8235
    },
    {
      "entropy": 5.892777967453003,
      "epoch": 0.7200908852573625,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004954920967332252,
      "loss": 5.5639,
      "mean_token_accuracy": 0.17654247283935548,
      "num_tokens": 16690447.0,
      "step": 8240
    },
    {
      "entropy": 5.918341732025146,
      "epoch": 0.7205278336100673,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004954858882505678,
      "loss": 5.6772,
      "mean_token_accuracy": 0.17430853694677353,
      "num_tokens": 16701061.0,
      "step": 8245
    },
    {
      "entropy": 5.936386251449585,
      "epoch": 0.720964781962772,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004954796755388755,
      "loss": 5.5607,
      "mean_token_accuracy": 0.18070277124643325,
      "num_tokens": 16711139.0,
      "step": 8250
    },
    {
      "entropy": 5.923426389694214,
      "epoch": 0.7214017303154767,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004954734585982675,
      "loss": 5.7038,
      "mean_token_accuracy": 0.16543979793787003,
      "num_tokens": 16722343.0,
      "step": 8255
    },
    {
      "entropy": 5.938275575637817,
      "epoch": 0.7218386786681814,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004954672374288631,
      "loss": 5.6865,
      "mean_token_accuracy": 0.16848636120557786,
      "num_tokens": 16733475.0,
      "step": 8260
    },
    {
      "entropy": 5.938981246948242,
      "epoch": 0.7222756270208861,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004954610120307815,
      "loss": 5.6716,
      "mean_token_accuracy": 0.17030773907899857,
      "num_tokens": 16744356.0,
      "step": 8265
    },
    {
      "entropy": 5.900533771514892,
      "epoch": 0.7227125753735908,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004954547824041422,
      "loss": 5.6855,
      "mean_token_accuracy": 0.17325286865234374,
      "num_tokens": 16755275.0,
      "step": 8270
    },
    {
      "entropy": 5.8641987323760985,
      "epoch": 0.7231495237262956,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004954485485490647,
      "loss": 5.6012,
      "mean_token_accuracy": 0.17692860215902328,
      "num_tokens": 16766114.0,
      "step": 8275
    },
    {
      "entropy": 5.913595867156983,
      "epoch": 0.7235864720790003,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004954423104656685,
      "loss": 5.596,
      "mean_token_accuracy": 0.18173280209302903,
      "num_tokens": 16776391.0,
      "step": 8280
    },
    {
      "entropy": 5.8187273979187015,
      "epoch": 0.724023420431705,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004954360681540732,
      "loss": 5.558,
      "mean_token_accuracy": 0.1858651041984558,
      "num_tokens": 16784845.0,
      "step": 8285
    },
    {
      "entropy": 5.8222010135650635,
      "epoch": 0.7244603687844097,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004954298216143986,
      "loss": 5.588,
      "mean_token_accuracy": 0.18075579702854155,
      "num_tokens": 16794735.0,
      "step": 8290
    },
    {
      "entropy": 5.997547817230225,
      "epoch": 0.7248973171371144,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004954235708467647,
      "loss": 5.6773,
      "mean_token_accuracy": 0.173848494887352,
      "num_tokens": 16803643.0,
      "step": 8295
    },
    {
      "entropy": 5.892679882049561,
      "epoch": 0.725334265489819,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000495417315851291,
      "loss": 5.6596,
      "mean_token_accuracy": 0.1739158570766449,
      "num_tokens": 16813064.0,
      "step": 8300
    },
    {
      "entropy": 5.899724340438842,
      "epoch": 0.7257712138425239,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004954110566280978,
      "loss": 5.6597,
      "mean_token_accuracy": 0.17038166373968125,
      "num_tokens": 16823166.0,
      "step": 8305
    },
    {
      "entropy": 5.937746953964234,
      "epoch": 0.7262081621952285,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000495404793177305,
      "loss": 5.6792,
      "mean_token_accuracy": 0.1684171438217163,
      "num_tokens": 16833546.0,
      "step": 8310
    },
    {
      "entropy": 5.938686990737915,
      "epoch": 0.7266451105479332,
      "grad_norm": 1.0,
      "learning_rate": 0.0004953985254990328,
      "loss": 5.5526,
      "mean_token_accuracy": 0.17863503098487854,
      "num_tokens": 16843794.0,
      "step": 8315
    },
    {
      "entropy": 5.953123712539673,
      "epoch": 0.7270820589006379,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004953922535934014,
      "loss": 5.7018,
      "mean_token_accuracy": 0.17126530706882476,
      "num_tokens": 16855677.0,
      "step": 8320
    },
    {
      "entropy": 5.931733560562134,
      "epoch": 0.7275190072533426,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000495385977460531,
      "loss": 5.6396,
      "mean_token_accuracy": 0.1685136526823044,
      "num_tokens": 16866137.0,
      "step": 8325
    },
    {
      "entropy": 5.9244402885437015,
      "epoch": 0.7279559556060473,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004953796971005422,
      "loss": 5.5697,
      "mean_token_accuracy": 0.17562247812747955,
      "num_tokens": 16875997.0,
      "step": 8330
    },
    {
      "entropy": 5.818835020065308,
      "epoch": 0.7283929039587521,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004953734125135552,
      "loss": 5.6119,
      "mean_token_accuracy": 0.18268579691648484,
      "num_tokens": 16885838.0,
      "step": 8335
    },
    {
      "entropy": 5.853946733474731,
      "epoch": 0.7288298523114568,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004953671236996909,
      "loss": 5.53,
      "mean_token_accuracy": 0.1850700780749321,
      "num_tokens": 16895533.0,
      "step": 8340
    },
    {
      "entropy": 5.968580961227417,
      "epoch": 0.7292668006641615,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004953608306590696,
      "loss": 5.669,
      "mean_token_accuracy": 0.17497962713241577,
      "num_tokens": 16906258.0,
      "step": 8345
    },
    {
      "entropy": 5.9226020812988285,
      "epoch": 0.7297037490168662,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000495354533391812,
      "loss": 5.6391,
      "mean_token_accuracy": 0.1779405176639557,
      "num_tokens": 16917123.0,
      "step": 8350
    },
    {
      "entropy": 5.8896783828735355,
      "epoch": 0.7301406973695709,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004953482318980389,
      "loss": 5.6421,
      "mean_token_accuracy": 0.17930960655212402,
      "num_tokens": 16927533.0,
      "step": 8355
    },
    {
      "entropy": 5.97347321510315,
      "epoch": 0.7305776457222757,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004953419261778714,
      "loss": 5.6167,
      "mean_token_accuracy": 0.17626841366291046,
      "num_tokens": 16937763.0,
      "step": 8360
    },
    {
      "entropy": 5.893387365341186,
      "epoch": 0.7310145940749804,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004953356162314304,
      "loss": 5.5505,
      "mean_token_accuracy": 0.18100274205207825,
      "num_tokens": 16947173.0,
      "step": 8365
    },
    {
      "entropy": 5.906252050399781,
      "epoch": 0.7314515424276851,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004953293020588368,
      "loss": 5.6366,
      "mean_token_accuracy": 0.17950859367847444,
      "num_tokens": 16956856.0,
      "step": 8370
    },
    {
      "entropy": 5.907759475708008,
      "epoch": 0.7318884907803898,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004953229836602115,
      "loss": 5.6397,
      "mean_token_accuracy": 0.17214396744966506,
      "num_tokens": 16966705.0,
      "step": 8375
    },
    {
      "entropy": 5.915832138061523,
      "epoch": 0.7323254391330944,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000495316661035676,
      "loss": 5.6374,
      "mean_token_accuracy": 0.17281220704317093,
      "num_tokens": 16977026.0,
      "step": 8380
    },
    {
      "entropy": 5.896400260925293,
      "epoch": 0.7327623874857991,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004953103341853517,
      "loss": 5.5599,
      "mean_token_accuracy": 0.18390518873929979,
      "num_tokens": 16986800.0,
      "step": 8385
    },
    {
      "entropy": 5.870542430877686,
      "epoch": 0.7331993358385039,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004953040031093596,
      "loss": 5.5376,
      "mean_token_accuracy": 0.17737241983413696,
      "num_tokens": 16996392.0,
      "step": 8390
    },
    {
      "entropy": 5.910156488418579,
      "epoch": 0.7336362841912086,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004952976678078213,
      "loss": 5.5783,
      "mean_token_accuracy": 0.17474534958600998,
      "num_tokens": 17006430.0,
      "step": 8395
    },
    {
      "entropy": 5.935700273513794,
      "epoch": 0.7340732325439133,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004952913282808584,
      "loss": 5.625,
      "mean_token_accuracy": 0.1702142834663391,
      "num_tokens": 17017168.0,
      "step": 8400
    },
    {
      "entropy": 5.807715225219726,
      "epoch": 0.734510180896618,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004952849845285922,
      "loss": 5.5525,
      "mean_token_accuracy": 0.179844668507576,
      "num_tokens": 17027337.0,
      "step": 8405
    },
    {
      "entropy": 5.903041696548462,
      "epoch": 0.7349471292493227,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004952786365511447,
      "loss": 5.5343,
      "mean_token_accuracy": 0.18069631606340408,
      "num_tokens": 17036839.0,
      "step": 8410
    },
    {
      "entropy": 5.903202915191651,
      "epoch": 0.7353840776020274,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004952722843486375,
      "loss": 5.7372,
      "mean_token_accuracy": 0.1686851128935814,
      "num_tokens": 17048221.0,
      "step": 8415
    },
    {
      "entropy": 5.907151365280152,
      "epoch": 0.7358210259547322,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004952659279211924,
      "loss": 5.568,
      "mean_token_accuracy": 0.1762968882918358,
      "num_tokens": 17057694.0,
      "step": 8420
    },
    {
      "entropy": 5.932665252685547,
      "epoch": 0.7362579743074369,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004952595672689315,
      "loss": 5.6553,
      "mean_token_accuracy": 0.1603608325123787,
      "num_tokens": 17066492.0,
      "step": 8425
    },
    {
      "entropy": 5.94142780303955,
      "epoch": 0.7366949226601416,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004952532023919766,
      "loss": 5.6418,
      "mean_token_accuracy": 0.17349047660827638,
      "num_tokens": 17077211.0,
      "step": 8430
    },
    {
      "entropy": 5.9196868419647215,
      "epoch": 0.7371318710128463,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004952468332904498,
      "loss": 5.6233,
      "mean_token_accuracy": 0.1780526503920555,
      "num_tokens": 17088665.0,
      "step": 8435
    },
    {
      "entropy": 5.873553323745727,
      "epoch": 0.737568819365551,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004952404599644735,
      "loss": 5.6053,
      "mean_token_accuracy": 0.17678896337747574,
      "num_tokens": 17099767.0,
      "step": 8440
    },
    {
      "entropy": 5.881426095962524,
      "epoch": 0.7380057677182557,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004952340824141698,
      "loss": 5.5362,
      "mean_token_accuracy": 0.18210793137550355,
      "num_tokens": 17109414.0,
      "step": 8445
    },
    {
      "entropy": 5.86700177192688,
      "epoch": 0.7384427160709605,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004952277006396609,
      "loss": 5.6252,
      "mean_token_accuracy": 0.17041776925325394,
      "num_tokens": 17119107.0,
      "step": 8450
    },
    {
      "entropy": 5.882149410247803,
      "epoch": 0.7388796644236652,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004952213146410693,
      "loss": 5.6495,
      "mean_token_accuracy": 0.17009395062923433,
      "num_tokens": 17129010.0,
      "step": 8455
    },
    {
      "entropy": 5.987139081954956,
      "epoch": 0.7393166127763698,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004952149244185174,
      "loss": 5.6758,
      "mean_token_accuracy": 0.16742247641086577,
      "num_tokens": 17138951.0,
      "step": 8460
    },
    {
      "entropy": 5.9611622333526615,
      "epoch": 0.7397535611290745,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004952085299721279,
      "loss": 5.5969,
      "mean_token_accuracy": 0.17259318381547928,
      "num_tokens": 17149416.0,
      "step": 8465
    },
    {
      "entropy": 5.861917543411255,
      "epoch": 0.7401905094817792,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004952021313020235,
      "loss": 5.5467,
      "mean_token_accuracy": 0.1854512169957161,
      "num_tokens": 17160068.0,
      "step": 8470
    },
    {
      "entropy": 5.862675523757934,
      "epoch": 0.7406274578344839,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004951957284083269,
      "loss": 5.6006,
      "mean_token_accuracy": 0.1744087442755699,
      "num_tokens": 17168972.0,
      "step": 8475
    },
    {
      "entropy": 5.9068115711212155,
      "epoch": 0.7410644061871887,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004951893212911607,
      "loss": 5.5809,
      "mean_token_accuracy": 0.18033021241426467,
      "num_tokens": 17178239.0,
      "step": 8480
    },
    {
      "entropy": 5.91066837310791,
      "epoch": 0.7415013545398934,
      "grad_norm": 0.828125,
      "learning_rate": 0.000495182909950648,
      "loss": 5.7456,
      "mean_token_accuracy": 0.1701326534152031,
      "num_tokens": 17190619.0,
      "step": 8485
    },
    {
      "entropy": 5.851893615722656,
      "epoch": 0.7419383028925981,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004951764943869117,
      "loss": 5.524,
      "mean_token_accuracy": 0.18032165318727494,
      "num_tokens": 17201041.0,
      "step": 8490
    },
    {
      "entropy": 5.883549451828003,
      "epoch": 0.7423752512453028,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004951700746000749,
      "loss": 5.643,
      "mean_token_accuracy": 0.169609172642231,
      "num_tokens": 17211779.0,
      "step": 8495
    },
    {
      "entropy": 5.900363969802856,
      "epoch": 0.7428121995980075,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004951636505902606,
      "loss": 5.5991,
      "mean_token_accuracy": 0.18183622062206267,
      "num_tokens": 17221249.0,
      "step": 8500
    },
    {
      "entropy": 5.9410741329193115,
      "epoch": 0.7432491479507122,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004951572223575922,
      "loss": 5.6456,
      "mean_token_accuracy": 0.1706314876675606,
      "num_tokens": 17231634.0,
      "step": 8505
    },
    {
      "entropy": 5.902452850341797,
      "epoch": 0.743686096303417,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004951507899021929,
      "loss": 5.6752,
      "mean_token_accuracy": 0.17234656363725662,
      "num_tokens": 17241751.0,
      "step": 8510
    },
    {
      "entropy": 5.945382165908813,
      "epoch": 0.7441230446561217,
      "grad_norm": 1.015625,
      "learning_rate": 0.000495144353224186,
      "loss": 5.6408,
      "mean_token_accuracy": 0.1761294811964035,
      "num_tokens": 17251962.0,
      "step": 8515
    },
    {
      "entropy": 5.919055652618408,
      "epoch": 0.7445599930088264,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004951379123236952,
      "loss": 5.544,
      "mean_token_accuracy": 0.1766454130411148,
      "num_tokens": 17261895.0,
      "step": 8520
    },
    {
      "entropy": 5.87232518196106,
      "epoch": 0.744996941361531,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004951314672008439,
      "loss": 5.6235,
      "mean_token_accuracy": 0.18150811344385148,
      "num_tokens": 17272025.0,
      "step": 8525
    },
    {
      "entropy": 5.926662635803223,
      "epoch": 0.7454338897142357,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004951250178557556,
      "loss": 5.6045,
      "mean_token_accuracy": 0.18201879113912584,
      "num_tokens": 17282705.0,
      "step": 8530
    },
    {
      "entropy": 5.886295938491822,
      "epoch": 0.7458708380669404,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004951185642885542,
      "loss": 5.5536,
      "mean_token_accuracy": 0.18098077774047852,
      "num_tokens": 17293045.0,
      "step": 8535
    },
    {
      "entropy": 5.945724296569824,
      "epoch": 0.7463077864196452,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004951121064993633,
      "loss": 5.6575,
      "mean_token_accuracy": 0.16938232630491257,
      "num_tokens": 17303893.0,
      "step": 8540
    },
    {
      "entropy": 5.909386205673218,
      "epoch": 0.7467447347723499,
      "grad_norm": 1.0,
      "learning_rate": 0.0004951056444883069,
      "loss": 5.615,
      "mean_token_accuracy": 0.16644756495952606,
      "num_tokens": 17314250.0,
      "step": 8545
    },
    {
      "entropy": 5.866173267364502,
      "epoch": 0.7471816831250546,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000495099178255509,
      "loss": 5.5383,
      "mean_token_accuracy": 0.17747173756361007,
      "num_tokens": 17324268.0,
      "step": 8550
    },
    {
      "entropy": 5.933831119537354,
      "epoch": 0.7476186314777593,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004950927078010934,
      "loss": 5.6571,
      "mean_token_accuracy": 0.16879514157772063,
      "num_tokens": 17334512.0,
      "step": 8555
    },
    {
      "entropy": 5.944765996932984,
      "epoch": 0.748055579830464,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004950862331251845,
      "loss": 5.6053,
      "mean_token_accuracy": 0.17522755712270738,
      "num_tokens": 17346244.0,
      "step": 8560
    },
    {
      "entropy": 5.856839752197265,
      "epoch": 0.7484925281831688,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004950797542279061,
      "loss": 5.6013,
      "mean_token_accuracy": 0.17897205650806428,
      "num_tokens": 17355832.0,
      "step": 8565
    },
    {
      "entropy": 5.931445360183716,
      "epoch": 0.7489294765358735,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004950732711093828,
      "loss": 5.6042,
      "mean_token_accuracy": 0.17676470875740052,
      "num_tokens": 17366871.0,
      "step": 8570
    },
    {
      "entropy": 5.981343412399292,
      "epoch": 0.7493664248885782,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004950667837697388,
      "loss": 5.6806,
      "mean_token_accuracy": 0.16790715605020523,
      "num_tokens": 17377242.0,
      "step": 8575
    },
    {
      "entropy": 5.951044368743896,
      "epoch": 0.7498033732412829,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004950602922090988,
      "loss": 5.649,
      "mean_token_accuracy": 0.1693328008055687,
      "num_tokens": 17387942.0,
      "step": 8580
    },
    {
      "entropy": 5.928488254547119,
      "epoch": 0.7502403215939876,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004950537964275868,
      "loss": 5.6103,
      "mean_token_accuracy": 0.16709787398576736,
      "num_tokens": 17397413.0,
      "step": 8585
    },
    {
      "entropy": 5.916830205917359,
      "epoch": 0.7506772699466923,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004950472964253278,
      "loss": 5.5551,
      "mean_token_accuracy": 0.1763176992535591,
      "num_tokens": 17408456.0,
      "step": 8590
    },
    {
      "entropy": 5.924591970443726,
      "epoch": 0.7511142182993971,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004950407922024463,
      "loss": 5.5709,
      "mean_token_accuracy": 0.17908758372068406,
      "num_tokens": 17417805.0,
      "step": 8595
    },
    {
      "entropy": 5.85422568321228,
      "epoch": 0.7515511666521018,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004950342837590671,
      "loss": 5.5202,
      "mean_token_accuracy": 0.17938392013311386,
      "num_tokens": 17426932.0,
      "step": 8600
    },
    {
      "entropy": 5.844671154022217,
      "epoch": 0.7519881150048064,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004950277710953152,
      "loss": 5.6173,
      "mean_token_accuracy": 0.17387023717164993,
      "num_tokens": 17436770.0,
      "step": 8605
    },
    {
      "entropy": 5.892503166198731,
      "epoch": 0.7524250633575111,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004950212542113153,
      "loss": 5.5435,
      "mean_token_accuracy": 0.1773211508989334,
      "num_tokens": 17447852.0,
      "step": 8610
    },
    {
      "entropy": 5.854351329803467,
      "epoch": 0.7528620117102158,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004950147331071924,
      "loss": 5.5386,
      "mean_token_accuracy": 0.17547447681427003,
      "num_tokens": 17458020.0,
      "step": 8615
    },
    {
      "entropy": 5.930532693862915,
      "epoch": 0.7532989600629205,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004950082077830717,
      "loss": 5.6556,
      "mean_token_accuracy": 0.17059237211942674,
      "num_tokens": 17468604.0,
      "step": 8620
    },
    {
      "entropy": 5.936930990219116,
      "epoch": 0.7537359084156253,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004950016782390782,
      "loss": 5.5898,
      "mean_token_accuracy": 0.1785125032067299,
      "num_tokens": 17478374.0,
      "step": 8625
    },
    {
      "entropy": 5.938373851776123,
      "epoch": 0.75417285676833,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004949951444753372,
      "loss": 5.5925,
      "mean_token_accuracy": 0.1741809979081154,
      "num_tokens": 17487894.0,
      "step": 8630
    },
    {
      "entropy": 5.890926122665405,
      "epoch": 0.7546098051210347,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004949886064919741,
      "loss": 5.5752,
      "mean_token_accuracy": 0.17204560339450836,
      "num_tokens": 17498473.0,
      "step": 8635
    },
    {
      "entropy": 5.924828386306762,
      "epoch": 0.7550467534737394,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004949820642891142,
      "loss": 5.5788,
      "mean_token_accuracy": 0.17649332880973817,
      "num_tokens": 17507796.0,
      "step": 8640
    },
    {
      "entropy": 5.836776399612427,
      "epoch": 0.7554837018264441,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004949755178668832,
      "loss": 5.5921,
      "mean_token_accuracy": 0.17796891033649445,
      "num_tokens": 17517629.0,
      "step": 8645
    },
    {
      "entropy": 5.935391998291015,
      "epoch": 0.7559206501791488,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004949689672254064,
      "loss": 5.664,
      "mean_token_accuracy": 0.16312945932149886,
      "num_tokens": 17528726.0,
      "step": 8650
    },
    {
      "entropy": 5.892144250869751,
      "epoch": 0.7563575985318536,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004949624123648096,
      "loss": 5.5503,
      "mean_token_accuracy": 0.17979073971509935,
      "num_tokens": 17539022.0,
      "step": 8655
    },
    {
      "entropy": 5.889766645431519,
      "epoch": 0.7567945468845583,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004949558532852185,
      "loss": 5.607,
      "mean_token_accuracy": 0.17835477888584136,
      "num_tokens": 17549293.0,
      "step": 8660
    },
    {
      "entropy": 5.914101409912109,
      "epoch": 0.757231495237263,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004949492899867587,
      "loss": 5.5955,
      "mean_token_accuracy": 0.1719934970140457,
      "num_tokens": 17559072.0,
      "step": 8665
    },
    {
      "entropy": 5.908097934722901,
      "epoch": 0.7576684435899677,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004949427224695564,
      "loss": 5.5765,
      "mean_token_accuracy": 0.17186656296253205,
      "num_tokens": 17568757.0,
      "step": 8670
    },
    {
      "entropy": 5.892359924316406,
      "epoch": 0.7581053919426723,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004949361507337374,
      "loss": 5.577,
      "mean_token_accuracy": 0.17804652601480483,
      "num_tokens": 17578288.0,
      "step": 8675
    },
    {
      "entropy": 5.858347225189209,
      "epoch": 0.758542340295377,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004949295747794277,
      "loss": 5.6746,
      "mean_token_accuracy": 0.16992638111114503,
      "num_tokens": 17588864.0,
      "step": 8680
    },
    {
      "entropy": 5.92705979347229,
      "epoch": 0.7589792886480818,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004949229946067537,
      "loss": 5.6902,
      "mean_token_accuracy": 0.1690803050994873,
      "num_tokens": 17599103.0,
      "step": 8685
    },
    {
      "entropy": 5.8211644172668455,
      "epoch": 0.7594162370007865,
      "grad_norm": 1.0,
      "learning_rate": 0.0004949164102158413,
      "loss": 5.5729,
      "mean_token_accuracy": 0.17779555022716523,
      "num_tokens": 17608608.0,
      "step": 8690
    },
    {
      "entropy": 5.936952209472656,
      "epoch": 0.7598531853534912,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000494909821606817,
      "loss": 5.6457,
      "mean_token_accuracy": 0.17514692693948747,
      "num_tokens": 17618604.0,
      "step": 8695
    },
    {
      "entropy": 5.966109085083008,
      "epoch": 0.7602901337061959,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004949032287798069,
      "loss": 5.6083,
      "mean_token_accuracy": 0.17379187792539597,
      "num_tokens": 17628479.0,
      "step": 8700
    },
    {
      "entropy": 5.887004947662353,
      "epoch": 0.7607270820589006,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004948966317349379,
      "loss": 5.5987,
      "mean_token_accuracy": 0.17903804332017897,
      "num_tokens": 17637519.0,
      "step": 8705
    },
    {
      "entropy": 5.89688892364502,
      "epoch": 0.7611640304116053,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004948900304723361,
      "loss": 5.5596,
      "mean_token_accuracy": 0.18254164904356002,
      "num_tokens": 17647902.0,
      "step": 8710
    },
    {
      "entropy": 5.914652013778687,
      "epoch": 0.7616009787643101,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004948834249921285,
      "loss": 5.5751,
      "mean_token_accuracy": 0.17777296602725984,
      "num_tokens": 17658576.0,
      "step": 8715
    },
    {
      "entropy": 5.903885269165039,
      "epoch": 0.7620379271170148,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004948768152944414,
      "loss": 5.6303,
      "mean_token_accuracy": 0.17258001416921614,
      "num_tokens": 17667059.0,
      "step": 8720
    },
    {
      "entropy": 5.989359664916992,
      "epoch": 0.7624748754697195,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004948702013794018,
      "loss": 5.7412,
      "mean_token_accuracy": 0.16413859874010087,
      "num_tokens": 17677447.0,
      "step": 8725
    },
    {
      "entropy": 5.933281517028808,
      "epoch": 0.7629118238224242,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004948635832471366,
      "loss": 5.6169,
      "mean_token_accuracy": 0.17940041720867156,
      "num_tokens": 17687893.0,
      "step": 8730
    },
    {
      "entropy": 5.870749807357788,
      "epoch": 0.7633487721751289,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004948569608977726,
      "loss": 5.6052,
      "mean_token_accuracy": 0.1765962228178978,
      "num_tokens": 17698599.0,
      "step": 8735
    },
    {
      "entropy": 5.926058959960938,
      "epoch": 0.7637857205278336,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004948503343314369,
      "loss": 5.6413,
      "mean_token_accuracy": 0.17233529984951018,
      "num_tokens": 17709711.0,
      "step": 8740
    },
    {
      "entropy": 5.918433380126953,
      "epoch": 0.7642226688805384,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004948437035482566,
      "loss": 5.596,
      "mean_token_accuracy": 0.18089101761579512,
      "num_tokens": 17719613.0,
      "step": 8745
    },
    {
      "entropy": 5.892623329162598,
      "epoch": 0.764659617233243,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004948370685483588,
      "loss": 5.5733,
      "mean_token_accuracy": 0.18544929325580597,
      "num_tokens": 17729829.0,
      "step": 8750
    },
    {
      "entropy": 5.885449600219727,
      "epoch": 0.7650965655859477,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004948304293318709,
      "loss": 5.5938,
      "mean_token_accuracy": 0.17445456087589264,
      "num_tokens": 17739254.0,
      "step": 8755
    },
    {
      "entropy": 5.933061122894287,
      "epoch": 0.7655335139386524,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004948237858989201,
      "loss": 5.5974,
      "mean_token_accuracy": 0.1727102130651474,
      "num_tokens": 17749294.0,
      "step": 8760
    },
    {
      "entropy": 5.88453516960144,
      "epoch": 0.7659704622913571,
      "grad_norm": 1.03125,
      "learning_rate": 0.000494817138249634,
      "loss": 5.5159,
      "mean_token_accuracy": 0.18405419439077378,
      "num_tokens": 17758866.0,
      "step": 8765
    },
    {
      "entropy": 5.9494177341461185,
      "epoch": 0.7664074106440619,
      "grad_norm": 1.03125,
      "learning_rate": 0.00049481048638414,
      "loss": 5.6955,
      "mean_token_accuracy": 0.16350459605455397,
      "num_tokens": 17769202.0,
      "step": 8770
    },
    {
      "entropy": 5.875809240341186,
      "epoch": 0.7668443589967666,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004948038303025657,
      "loss": 5.6795,
      "mean_token_accuracy": 0.1740434467792511,
      "num_tokens": 17779125.0,
      "step": 8775
    },
    {
      "entropy": 5.948620176315307,
      "epoch": 0.7672813073494713,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004947971700050387,
      "loss": 5.6182,
      "mean_token_accuracy": 0.1713486298918724,
      "num_tokens": 17788696.0,
      "step": 8780
    },
    {
      "entropy": 5.913877820968628,
      "epoch": 0.767718255702176,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004947905054916869,
      "loss": 5.6048,
      "mean_token_accuracy": 0.17460788339376448,
      "num_tokens": 17799867.0,
      "step": 8785
    },
    {
      "entropy": 5.965176725387574,
      "epoch": 0.7681552040548807,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004947838367626379,
      "loss": 5.624,
      "mean_token_accuracy": 0.17099858671426774,
      "num_tokens": 17809418.0,
      "step": 8790
    },
    {
      "entropy": 5.87390022277832,
      "epoch": 0.7685921524075854,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004947771638180198,
      "loss": 5.5153,
      "mean_token_accuracy": 0.1798015922307968,
      "num_tokens": 17818554.0,
      "step": 8795
    },
    {
      "entropy": 5.825968790054321,
      "epoch": 0.7690291007602902,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004947704866579605,
      "loss": 5.5513,
      "mean_token_accuracy": 0.17864262461662292,
      "num_tokens": 17828918.0,
      "step": 8800
    },
    {
      "entropy": 5.956258010864258,
      "epoch": 0.7694660491129949,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004947638052825883,
      "loss": 5.7137,
      "mean_token_accuracy": 0.1663025677204132,
      "num_tokens": 17840791.0,
      "step": 8805
    },
    {
      "entropy": 5.970945167541504,
      "epoch": 0.7699029974656996,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004947571196920309,
      "loss": 5.6009,
      "mean_token_accuracy": 0.17106885612010955,
      "num_tokens": 17850675.0,
      "step": 8810
    },
    {
      "entropy": 5.838865280151367,
      "epoch": 0.7703399458184043,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004947504298864169,
      "loss": 5.5707,
      "mean_token_accuracy": 0.18421148657798767,
      "num_tokens": 17860684.0,
      "step": 8815
    },
    {
      "entropy": 5.8606020450592045,
      "epoch": 0.770776894171109,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004947437358658746,
      "loss": 5.6013,
      "mean_token_accuracy": 0.17578084021806717,
      "num_tokens": 17870394.0,
      "step": 8820
    },
    {
      "entropy": 5.890427017211914,
      "epoch": 0.7712138425238136,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004947370376305321,
      "loss": 5.5613,
      "mean_token_accuracy": 0.17520854920148848,
      "num_tokens": 17881256.0,
      "step": 8825
    },
    {
      "entropy": 5.894306087493897,
      "epoch": 0.7716507908765184,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004947303351805183,
      "loss": 5.6082,
      "mean_token_accuracy": 0.17457226067781448,
      "num_tokens": 17890888.0,
      "step": 8830
    },
    {
      "entropy": 5.9256593704223635,
      "epoch": 0.7720877392292231,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004947236285159615,
      "loss": 5.5914,
      "mean_token_accuracy": 0.1744719162583351,
      "num_tokens": 17900507.0,
      "step": 8835
    },
    {
      "entropy": 5.897252178192138,
      "epoch": 0.7725246875819278,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004947169176369902,
      "loss": 5.6643,
      "mean_token_accuracy": 0.17159840017557143,
      "num_tokens": 17911333.0,
      "step": 8840
    },
    {
      "entropy": 5.970972013473511,
      "epoch": 0.7729616359346325,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004947102025437334,
      "loss": 5.5997,
      "mean_token_accuracy": 0.1767983078956604,
      "num_tokens": 17921540.0,
      "step": 8845
    },
    {
      "entropy": 5.937785816192627,
      "epoch": 0.7733985842873372,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004947034832363199,
      "loss": 5.6891,
      "mean_token_accuracy": 0.17068566828966142,
      "num_tokens": 17931820.0,
      "step": 8850
    },
    {
      "entropy": 5.899560976028442,
      "epoch": 0.7738355326400419,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004946967597148782,
      "loss": 5.6023,
      "mean_token_accuracy": 0.17295095473527908,
      "num_tokens": 17942717.0,
      "step": 8855
    },
    {
      "entropy": 5.9277972221374515,
      "epoch": 0.7742724809927467,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004946900319795377,
      "loss": 5.5902,
      "mean_token_accuracy": 0.1811741292476654,
      "num_tokens": 17953775.0,
      "step": 8860
    },
    {
      "entropy": 5.920348596572876,
      "epoch": 0.7747094293454514,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004946833000304274,
      "loss": 5.5722,
      "mean_token_accuracy": 0.1718575805425644,
      "num_tokens": 17963877.0,
      "step": 8865
    },
    {
      "entropy": 5.873974323272705,
      "epoch": 0.7751463776981561,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004946765638676761,
      "loss": 5.5725,
      "mean_token_accuracy": 0.1806543380022049,
      "num_tokens": 17974015.0,
      "step": 8870
    },
    {
      "entropy": 5.885319995880127,
      "epoch": 0.7755833260508608,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004946698234914132,
      "loss": 5.5867,
      "mean_token_accuracy": 0.18262851536273955,
      "num_tokens": 17984189.0,
      "step": 8875
    },
    {
      "entropy": 5.865130186080933,
      "epoch": 0.7760202744035655,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000494663078901768,
      "loss": 5.5441,
      "mean_token_accuracy": 0.18494847416877747,
      "num_tokens": 17994615.0,
      "step": 8880
    },
    {
      "entropy": 5.906695985794068,
      "epoch": 0.7764572227562702,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004946563300988699,
      "loss": 5.5811,
      "mean_token_accuracy": 0.17282510548830032,
      "num_tokens": 18004727.0,
      "step": 8885
    },
    {
      "entropy": 5.903292608261109,
      "epoch": 0.776894171108975,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004946495770828482,
      "loss": 5.644,
      "mean_token_accuracy": 0.1755265548825264,
      "num_tokens": 18014576.0,
      "step": 8890
    },
    {
      "entropy": 5.878237581253051,
      "epoch": 0.7773311194616797,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004946428198538326,
      "loss": 5.6427,
      "mean_token_accuracy": 0.17987878024578094,
      "num_tokens": 18024709.0,
      "step": 8895
    },
    {
      "entropy": 5.93417649269104,
      "epoch": 0.7777680678143843,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004946360584119526,
      "loss": 5.5718,
      "mean_token_accuracy": 0.1721986472606659,
      "num_tokens": 18034673.0,
      "step": 8900
    },
    {
      "entropy": 5.929644918441772,
      "epoch": 0.778205016167089,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000494629292757338,
      "loss": 5.5668,
      "mean_token_accuracy": 0.17459271848201752,
      "num_tokens": 18044774.0,
      "step": 8905
    },
    {
      "entropy": 5.857529830932617,
      "epoch": 0.7786419645197937,
      "grad_norm": 1.0,
      "learning_rate": 0.0004946225228901184,
      "loss": 5.5799,
      "mean_token_accuracy": 0.17066750228404998,
      "num_tokens": 18055244.0,
      "step": 8910
    },
    {
      "entropy": 5.923955249786377,
      "epoch": 0.7790789128724984,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004946157488104236,
      "loss": 5.6635,
      "mean_token_accuracy": 0.1686917930841446,
      "num_tokens": 18066051.0,
      "step": 8915
    },
    {
      "entropy": 5.969832849502564,
      "epoch": 0.7795158612252032,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004946089705183839,
      "loss": 5.6908,
      "mean_token_accuracy": 0.1623295560479164,
      "num_tokens": 18077693.0,
      "step": 8920
    },
    {
      "entropy": 5.928920125961303,
      "epoch": 0.7799528095779079,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004946021880141289,
      "loss": 5.6003,
      "mean_token_accuracy": 0.17926839739084244,
      "num_tokens": 18088032.0,
      "step": 8925
    },
    {
      "entropy": 5.920402956008911,
      "epoch": 0.7803897579306126,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000494595401297789,
      "loss": 5.5748,
      "mean_token_accuracy": 0.17675147205591202,
      "num_tokens": 18097998.0,
      "step": 8930
    },
    {
      "entropy": 5.87354416847229,
      "epoch": 0.7808267062833173,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004945886103694941,
      "loss": 5.5259,
      "mean_token_accuracy": 0.18091251999139785,
      "num_tokens": 18107538.0,
      "step": 8935
    },
    {
      "entropy": 5.891588306427002,
      "epoch": 0.781263654636022,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004945818152293747,
      "loss": 5.6197,
      "mean_token_accuracy": 0.17868129760026932,
      "num_tokens": 18117792.0,
      "step": 8940
    },
    {
      "entropy": 5.932331657409668,
      "epoch": 0.7817006029887267,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004945750158775609,
      "loss": 5.6247,
      "mean_token_accuracy": 0.1735410660505295,
      "num_tokens": 18128005.0,
      "step": 8945
    },
    {
      "entropy": 5.877100896835327,
      "epoch": 0.7821375513414315,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004945682123141833,
      "loss": 5.5993,
      "mean_token_accuracy": 0.1785768061876297,
      "num_tokens": 18138398.0,
      "step": 8950
    },
    {
      "entropy": 5.911321115493775,
      "epoch": 0.7825744996941362,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004945614045393724,
      "loss": 5.5704,
      "mean_token_accuracy": 0.17316683977842331,
      "num_tokens": 18147676.0,
      "step": 8955
    },
    {
      "entropy": 5.823401880264282,
      "epoch": 0.7830114480468409,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004945545925532587,
      "loss": 5.4825,
      "mean_token_accuracy": 0.18595926612615585,
      "num_tokens": 18156842.0,
      "step": 8960
    },
    {
      "entropy": 5.884054183959961,
      "epoch": 0.7834483963995456,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004945477763559728,
      "loss": 5.6203,
      "mean_token_accuracy": 0.17241388112306594,
      "num_tokens": 18166902.0,
      "step": 8965
    },
    {
      "entropy": 5.891348886489868,
      "epoch": 0.7838853447522502,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004945409559476455,
      "loss": 5.5905,
      "mean_token_accuracy": 0.17643243670463563,
      "num_tokens": 18177561.0,
      "step": 8970
    },
    {
      "entropy": 5.8535362720489506,
      "epoch": 0.784322293104955,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004945341313284077,
      "loss": 5.5772,
      "mean_token_accuracy": 0.18069988042116164,
      "num_tokens": 18187739.0,
      "step": 8975
    },
    {
      "entropy": 5.863891363143921,
      "epoch": 0.7847592414576597,
      "grad_norm": 0.875,
      "learning_rate": 0.0004945273024983902,
      "loss": 5.5053,
      "mean_token_accuracy": 0.18076965510845183,
      "num_tokens": 18198435.0,
      "step": 8980
    },
    {
      "entropy": 5.930920219421386,
      "epoch": 0.7851961898103644,
      "grad_norm": 0.953125,
      "learning_rate": 0.000494520469457724,
      "loss": 5.534,
      "mean_token_accuracy": 0.18251188099384308,
      "num_tokens": 18208374.0,
      "step": 8985
    },
    {
      "entropy": 5.827178239822388,
      "epoch": 0.7856331381630691,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004945136322065402,
      "loss": 5.6257,
      "mean_token_accuracy": 0.17918350845575332,
      "num_tokens": 18219229.0,
      "step": 8990
    },
    {
      "entropy": 5.876517581939697,
      "epoch": 0.7860700865157738,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004945067907449699,
      "loss": 5.55,
      "mean_token_accuracy": 0.17960241734981536,
      "num_tokens": 18229208.0,
      "step": 8995
    },
    {
      "entropy": 5.981864070892334,
      "epoch": 0.7865070348684785,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004944999450731442,
      "loss": 5.7064,
      "mean_token_accuracy": 0.16530189663171768,
      "num_tokens": 18238454.0,
      "step": 9000
    },
    {
      "epoch": 0.7865070348684785,
      "eval_entropy": 5.742980309328302,
      "eval_loss": 5.601479530334473,
      "eval_mean_token_accuracy": 0.1841879130933077,
      "eval_num_tokens": 18238454.0,
      "eval_runtime": 21.599,
      "eval_samples_per_second": 1713.69,
      "eval_steps_per_second": 214.223,
      "step": 9000
    },
    {
      "entropy": 5.920482110977173,
      "epoch": 0.7869439832211833,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004944930951911947,
      "loss": 5.5844,
      "mean_token_accuracy": 0.18012760877609252,
      "num_tokens": 18248845.0,
      "step": 9005
    },
    {
      "entropy": 5.972012281417847,
      "epoch": 0.787380931573888,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004944862410992526,
      "loss": 5.6022,
      "mean_token_accuracy": 0.1733531579375267,
      "num_tokens": 18258912.0,
      "step": 9010
    },
    {
      "entropy": 5.891234731674194,
      "epoch": 0.7878178799265927,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004944793827974494,
      "loss": 5.628,
      "mean_token_accuracy": 0.16904295086860657,
      "num_tokens": 18269330.0,
      "step": 9015
    },
    {
      "entropy": 5.8841194152832035,
      "epoch": 0.7882548282792974,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004944725202859165,
      "loss": 5.604,
      "mean_token_accuracy": 0.17779863625764847,
      "num_tokens": 18279721.0,
      "step": 9020
    },
    {
      "entropy": 5.865096187591552,
      "epoch": 0.7886917766320021,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004944656535647858,
      "loss": 5.5334,
      "mean_token_accuracy": 0.18496831506490707,
      "num_tokens": 18288581.0,
      "step": 9025
    },
    {
      "entropy": 5.881364345550537,
      "epoch": 0.7891287249847068,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004944587826341888,
      "loss": 5.5886,
      "mean_token_accuracy": 0.1747352510690689,
      "num_tokens": 18298179.0,
      "step": 9030
    },
    {
      "entropy": 5.904235363006592,
      "epoch": 0.7895656733374116,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004944519074942574,
      "loss": 5.5875,
      "mean_token_accuracy": 0.1727092981338501,
      "num_tokens": 18307957.0,
      "step": 9035
    },
    {
      "entropy": 5.808629941940308,
      "epoch": 0.7900026216901163,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004944450281451234,
      "loss": 5.5672,
      "mean_token_accuracy": 0.17520609945058824,
      "num_tokens": 18317799.0,
      "step": 9040
    },
    {
      "entropy": 5.973935937881469,
      "epoch": 0.790439570042821,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004944381445869188,
      "loss": 5.6348,
      "mean_token_accuracy": 0.17986664175987244,
      "num_tokens": 18327889.0,
      "step": 9045
    },
    {
      "entropy": 5.916871643066406,
      "epoch": 0.7908765183955256,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004944312568197756,
      "loss": 5.5946,
      "mean_token_accuracy": 0.18314420580863952,
      "num_tokens": 18338288.0,
      "step": 9050
    },
    {
      "entropy": 5.890463018417359,
      "epoch": 0.7913134667482303,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004944243648438258,
      "loss": 5.6468,
      "mean_token_accuracy": 0.16756970882415773,
      "num_tokens": 18349070.0,
      "step": 9055
    },
    {
      "entropy": 5.957594394683838,
      "epoch": 0.791750415100935,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004944174686592019,
      "loss": 5.572,
      "mean_token_accuracy": 0.17855478078126907,
      "num_tokens": 18359683.0,
      "step": 9060
    },
    {
      "entropy": 5.948366594314575,
      "epoch": 0.7921873634536398,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004944105682660358,
      "loss": 5.5721,
      "mean_token_accuracy": 0.17706146389245986,
      "num_tokens": 18371355.0,
      "step": 9065
    },
    {
      "entropy": 5.7725670337677,
      "epoch": 0.7926243118063445,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004944036636644601,
      "loss": 5.6081,
      "mean_token_accuracy": 0.18081346303224563,
      "num_tokens": 18381591.0,
      "step": 9070
    },
    {
      "entropy": 5.930037546157837,
      "epoch": 0.7930612601590492,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004943967548546071,
      "loss": 5.5891,
      "mean_token_accuracy": 0.17402844429016112,
      "num_tokens": 18390760.0,
      "step": 9075
    },
    {
      "entropy": 5.974299526214599,
      "epoch": 0.7934982085117539,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004943898418366096,
      "loss": 5.6344,
      "mean_token_accuracy": 0.17353388369083406,
      "num_tokens": 18400643.0,
      "step": 9080
    },
    {
      "entropy": 5.913231134414673,
      "epoch": 0.7939351568644586,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004943829246105997,
      "loss": 5.6637,
      "mean_token_accuracy": 0.1717489629983902,
      "num_tokens": 18411559.0,
      "step": 9085
    },
    {
      "entropy": 5.9020185470581055,
      "epoch": 0.7943721052171633,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004943760031767106,
      "loss": 5.5999,
      "mean_token_accuracy": 0.1801138773560524,
      "num_tokens": 18422645.0,
      "step": 9090
    },
    {
      "entropy": 5.894098472595215,
      "epoch": 0.7948090535698681,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004943690775350746,
      "loss": 5.5839,
      "mean_token_accuracy": 0.17699578702449797,
      "num_tokens": 18432369.0,
      "step": 9095
    },
    {
      "entropy": 5.918116569519043,
      "epoch": 0.7952460019225728,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004943621476858249,
      "loss": 5.5522,
      "mean_token_accuracy": 0.17962248325347902,
      "num_tokens": 18442143.0,
      "step": 9100
    },
    {
      "entropy": 5.853396368026734,
      "epoch": 0.7956829502752775,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004943552136290942,
      "loss": 5.5995,
      "mean_token_accuracy": 0.17885066866874694,
      "num_tokens": 18452161.0,
      "step": 9105
    },
    {
      "entropy": 5.954837226867676,
      "epoch": 0.7961198986279822,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004943482753650154,
      "loss": 5.7064,
      "mean_token_accuracy": 0.17203027606010438,
      "num_tokens": 18461873.0,
      "step": 9110
    },
    {
      "entropy": 5.91164002418518,
      "epoch": 0.7965568469806869,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004943413328937221,
      "loss": 5.5814,
      "mean_token_accuracy": 0.17650333940982818,
      "num_tokens": 18472584.0,
      "step": 9115
    },
    {
      "entropy": 5.9386616230010985,
      "epoch": 0.7969937953333915,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004943343862153467,
      "loss": 5.5403,
      "mean_token_accuracy": 0.17396142482757568,
      "num_tokens": 18482174.0,
      "step": 9120
    },
    {
      "entropy": 5.932789468765259,
      "epoch": 0.7974307436860963,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004943274353300231,
      "loss": 5.7021,
      "mean_token_accuracy": 0.1708983972668648,
      "num_tokens": 18492861.0,
      "step": 9125
    },
    {
      "entropy": 5.935539817810058,
      "epoch": 0.797867692038801,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004943204802378843,
      "loss": 5.5925,
      "mean_token_accuracy": 0.17776966243982315,
      "num_tokens": 18502830.0,
      "step": 9130
    },
    {
      "entropy": 5.950483751296997,
      "epoch": 0.7983046403915057,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004943135209390638,
      "loss": 5.5892,
      "mean_token_accuracy": 0.1738123267889023,
      "num_tokens": 18513063.0,
      "step": 9135
    },
    {
      "entropy": 5.823931694030762,
      "epoch": 0.7987415887442104,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004943065574336949,
      "loss": 5.5714,
      "mean_token_accuracy": 0.18051699697971343,
      "num_tokens": 18523773.0,
      "step": 9140
    },
    {
      "entropy": 5.942397975921631,
      "epoch": 0.7991785370969151,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004942995897219115,
      "loss": 5.5836,
      "mean_token_accuracy": 0.1774800091981888,
      "num_tokens": 18533101.0,
      "step": 9145
    },
    {
      "entropy": 5.947246551513672,
      "epoch": 0.7996154854496198,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004942926178038469,
      "loss": 5.6257,
      "mean_token_accuracy": 0.17890289574861526,
      "num_tokens": 18543863.0,
      "step": 9150
    },
    {
      "entropy": 5.865755414962768,
      "epoch": 0.8000524338023246,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004942856416796352,
      "loss": 5.6221,
      "mean_token_accuracy": 0.17422727942466737,
      "num_tokens": 18554249.0,
      "step": 9155
    },
    {
      "entropy": 5.903285169601441,
      "epoch": 0.8004893821550293,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004942786613494098,
      "loss": 5.587,
      "mean_token_accuracy": 0.17542706429958344,
      "num_tokens": 18563525.0,
      "step": 9160
    },
    {
      "entropy": 5.888577842712403,
      "epoch": 0.800926330507734,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004942716768133048,
      "loss": 5.5641,
      "mean_token_accuracy": 0.17859871536493302,
      "num_tokens": 18572734.0,
      "step": 9165
    },
    {
      "entropy": 5.875475883483887,
      "epoch": 0.8013632788604387,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004942646880714543,
      "loss": 5.6315,
      "mean_token_accuracy": 0.17567993104457855,
      "num_tokens": 18581939.0,
      "step": 9170
    },
    {
      "entropy": 5.951259899139404,
      "epoch": 0.8018002272131434,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000494257695123992,
      "loss": 5.6163,
      "mean_token_accuracy": 0.17827687561511993,
      "num_tokens": 18591635.0,
      "step": 9175
    },
    {
      "entropy": 5.935528087615967,
      "epoch": 0.8022371755658482,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004942506979710524,
      "loss": 5.5125,
      "mean_token_accuracy": 0.17725635319948196,
      "num_tokens": 18602060.0,
      "step": 9180
    },
    {
      "entropy": 5.872634220123291,
      "epoch": 0.8026741239185529,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004942436966127693,
      "loss": 5.5711,
      "mean_token_accuracy": 0.17681765109300612,
      "num_tokens": 18611187.0,
      "step": 9185
    },
    {
      "entropy": 5.919779682159424,
      "epoch": 0.8031110722712576,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004942366910492775,
      "loss": 5.5155,
      "mean_token_accuracy": 0.18545909225940704,
      "num_tokens": 18620329.0,
      "step": 9190
    },
    {
      "entropy": 5.967359304428101,
      "epoch": 0.8035480206239622,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004942296812807109,
      "loss": 5.5951,
      "mean_token_accuracy": 0.17340246140956878,
      "num_tokens": 18629600.0,
      "step": 9195
    },
    {
      "entropy": 5.877811002731323,
      "epoch": 0.8039849689766669,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004942226673072041,
      "loss": 5.6507,
      "mean_token_accuracy": 0.1684898167848587,
      "num_tokens": 18640771.0,
      "step": 9200
    },
    {
      "entropy": 5.813657188415528,
      "epoch": 0.8044219173293716,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004942156491288917,
      "loss": 5.5548,
      "mean_token_accuracy": 0.17577370703220369,
      "num_tokens": 18650606.0,
      "step": 9205
    },
    {
      "entropy": 5.906002473831177,
      "epoch": 0.8048588656820764,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004942086267459083,
      "loss": 5.5962,
      "mean_token_accuracy": 0.17708454430103301,
      "num_tokens": 18660487.0,
      "step": 9210
    },
    {
      "entropy": 5.934459161758423,
      "epoch": 0.8052958140347811,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004942016001583886,
      "loss": 5.5878,
      "mean_token_accuracy": 0.17897555679082872,
      "num_tokens": 18670987.0,
      "step": 9215
    },
    {
      "entropy": 5.892013645172119,
      "epoch": 0.8057327623874858,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004941945693664674,
      "loss": 5.6626,
      "mean_token_accuracy": 0.1700621321797371,
      "num_tokens": 18682560.0,
      "step": 9220
    },
    {
      "entropy": 5.925118780136108,
      "epoch": 0.8061697107401905,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004941875343702795,
      "loss": 5.6112,
      "mean_token_accuracy": 0.17602229118347168,
      "num_tokens": 18692743.0,
      "step": 9225
    },
    {
      "entropy": 5.921692609786987,
      "epoch": 0.8066066590928952,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004941804951699597,
      "loss": 5.5722,
      "mean_token_accuracy": 0.17854295223951339,
      "num_tokens": 18703039.0,
      "step": 9230
    },
    {
      "entropy": 5.849697494506836,
      "epoch": 0.8070436074455999,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004941734517656432,
      "loss": 5.6038,
      "mean_token_accuracy": 0.17764324247837066,
      "num_tokens": 18713777.0,
      "step": 9235
    },
    {
      "entropy": 5.8858050346374515,
      "epoch": 0.8074805557983047,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004941664041574651,
      "loss": 5.6177,
      "mean_token_accuracy": 0.17466839998960496,
      "num_tokens": 18723422.0,
      "step": 9240
    },
    {
      "entropy": 5.921637392044067,
      "epoch": 0.8079175041510094,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004941593523455605,
      "loss": 5.5554,
      "mean_token_accuracy": 0.17754586338996886,
      "num_tokens": 18734369.0,
      "step": 9245
    },
    {
      "entropy": 5.976261043548584,
      "epoch": 0.8083544525037141,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004941522963300647,
      "loss": 5.6174,
      "mean_token_accuracy": 0.1757915735244751,
      "num_tokens": 18744647.0,
      "step": 9250
    },
    {
      "entropy": 5.95689959526062,
      "epoch": 0.8087914008564188,
      "grad_norm": 1.015625,
      "learning_rate": 0.000494145236111113,
      "loss": 5.5672,
      "mean_token_accuracy": 0.1740198627114296,
      "num_tokens": 18753929.0,
      "step": 9255
    },
    {
      "entropy": 5.896379804611206,
      "epoch": 0.8092283492091235,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004941381716888409,
      "loss": 5.5855,
      "mean_token_accuracy": 0.17592449337244034,
      "num_tokens": 18764103.0,
      "step": 9260
    },
    {
      "entropy": 5.934111881256103,
      "epoch": 0.8096652975618281,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004941311030633838,
      "loss": 5.6275,
      "mean_token_accuracy": 0.1726614385843277,
      "num_tokens": 18773756.0,
      "step": 9265
    },
    {
      "entropy": 5.855281162261963,
      "epoch": 0.810102245914533,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004941240302348773,
      "loss": 5.4906,
      "mean_token_accuracy": 0.18545803278684617,
      "num_tokens": 18783778.0,
      "step": 9270
    },
    {
      "entropy": 5.867553091049194,
      "epoch": 0.8105391942672376,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004941169532034571,
      "loss": 5.6232,
      "mean_token_accuracy": 0.17448436468839645,
      "num_tokens": 18794664.0,
      "step": 9275
    },
    {
      "entropy": 5.900312423706055,
      "epoch": 0.8109761426199423,
      "grad_norm": 1.0,
      "learning_rate": 0.0004941098719692589,
      "loss": 5.5805,
      "mean_token_accuracy": 0.16826996505260466,
      "num_tokens": 18803773.0,
      "step": 9280
    },
    {
      "entropy": 5.88424129486084,
      "epoch": 0.811413090972647,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004941027865324186,
      "loss": 5.6379,
      "mean_token_accuracy": 0.17299726009368896,
      "num_tokens": 18814419.0,
      "step": 9285
    },
    {
      "entropy": 5.96412205696106,
      "epoch": 0.8118500393253517,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000494095696893072,
      "loss": 5.5901,
      "mean_token_accuracy": 0.1740448758006096,
      "num_tokens": 18825011.0,
      "step": 9290
    },
    {
      "entropy": 5.883392953872681,
      "epoch": 0.8122869876780564,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004940886030513551,
      "loss": 5.5076,
      "mean_token_accuracy": 0.1810905158519745,
      "num_tokens": 18835003.0,
      "step": 9295
    },
    {
      "entropy": 5.871304321289062,
      "epoch": 0.8127239360307612,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004940815050074041,
      "loss": 5.5889,
      "mean_token_accuracy": 0.17430961430072783,
      "num_tokens": 18845072.0,
      "step": 9300
    },
    {
      "entropy": 5.9592287063598635,
      "epoch": 0.8131608843834659,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004940744027613549,
      "loss": 5.5875,
      "mean_token_accuracy": 0.17810352742671967,
      "num_tokens": 18854583.0,
      "step": 9305
    },
    {
      "entropy": 5.9458740711212155,
      "epoch": 0.8135978327361706,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004940672963133439,
      "loss": 5.5861,
      "mean_token_accuracy": 0.18217318505048752,
      "num_tokens": 18865460.0,
      "step": 9310
    },
    {
      "entropy": 5.861312770843506,
      "epoch": 0.8140347810888753,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004940601856635074,
      "loss": 5.6004,
      "mean_token_accuracy": 0.17715551406145097,
      "num_tokens": 18875568.0,
      "step": 9315
    },
    {
      "entropy": 5.833522367477417,
      "epoch": 0.81447172944158,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004940530708119818,
      "loss": 5.5038,
      "mean_token_accuracy": 0.1808427318930626,
      "num_tokens": 18884894.0,
      "step": 9320
    },
    {
      "entropy": 5.8921302318572994,
      "epoch": 0.8149086777942847,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004940459517589034,
      "loss": 5.5539,
      "mean_token_accuracy": 0.17703500688076018,
      "num_tokens": 18895353.0,
      "step": 9325
    },
    {
      "entropy": 5.907979297637939,
      "epoch": 0.8153456261469895,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000494038828504409,
      "loss": 5.5972,
      "mean_token_accuracy": 0.1810010105371475,
      "num_tokens": 18905474.0,
      "step": 9330
    },
    {
      "entropy": 5.891048812866211,
      "epoch": 0.8157825744996942,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004940317010486351,
      "loss": 5.5357,
      "mean_token_accuracy": 0.18218014985322953,
      "num_tokens": 18914797.0,
      "step": 9335
    },
    {
      "entropy": 5.866167211532593,
      "epoch": 0.8162195228523988,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004940245693917183,
      "loss": 5.6059,
      "mean_token_accuracy": 0.17374850064516068,
      "num_tokens": 18924642.0,
      "step": 9340
    },
    {
      "entropy": 5.831859731674195,
      "epoch": 0.8166564712051035,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004940174335337955,
      "loss": 5.4962,
      "mean_token_accuracy": 0.1884779304265976,
      "num_tokens": 18934026.0,
      "step": 9345
    },
    {
      "entropy": 5.905361270904541,
      "epoch": 0.8170934195578082,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004940102934750037,
      "loss": 5.5024,
      "mean_token_accuracy": 0.18683067858219146,
      "num_tokens": 18943262.0,
      "step": 9350
    },
    {
      "entropy": 5.877033662796021,
      "epoch": 0.8175303679105129,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004940031492154795,
      "loss": 5.5478,
      "mean_token_accuracy": 0.17228838801383972,
      "num_tokens": 18953082.0,
      "step": 9355
    },
    {
      "entropy": 5.8511982440948485,
      "epoch": 0.8179673162632177,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004939960007553603,
      "loss": 5.5724,
      "mean_token_accuracy": 0.1823079064488411,
      "num_tokens": 18962408.0,
      "step": 9360
    },
    {
      "entropy": 5.954546308517456,
      "epoch": 0.8184042646159224,
      "grad_norm": 0.90625,
      "learning_rate": 0.000493988848094783,
      "loss": 5.6536,
      "mean_token_accuracy": 0.16534460932016373,
      "num_tokens": 18972918.0,
      "step": 9365
    },
    {
      "entropy": 5.940453147888183,
      "epoch": 0.8188412129686271,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004939816912338848,
      "loss": 5.6002,
      "mean_token_accuracy": 0.17439760267734528,
      "num_tokens": 18983152.0,
      "step": 9370
    },
    {
      "entropy": 5.837433528900147,
      "epoch": 0.8192781613213318,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004939745301728031,
      "loss": 5.6022,
      "mean_token_accuracy": 0.1728952407836914,
      "num_tokens": 18993613.0,
      "step": 9375
    },
    {
      "entropy": 5.8522376537323,
      "epoch": 0.8197151096740365,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004939673649116752,
      "loss": 5.5094,
      "mean_token_accuracy": 0.17991740852594376,
      "num_tokens": 19004079.0,
      "step": 9380
    },
    {
      "entropy": 5.882761430740357,
      "epoch": 0.8201520580267413,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004939601954506383,
      "loss": 5.6543,
      "mean_token_accuracy": 0.1756019800901413,
      "num_tokens": 19013994.0,
      "step": 9385
    },
    {
      "entropy": 5.9017080783844,
      "epoch": 0.820589006379446,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004939530217898303,
      "loss": 5.6448,
      "mean_token_accuracy": 0.17020079344511033,
      "num_tokens": 19024674.0,
      "step": 9390
    },
    {
      "entropy": 5.951020050048828,
      "epoch": 0.8210259547321507,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004939458439293886,
      "loss": 5.6038,
      "mean_token_accuracy": 0.17405484914779662,
      "num_tokens": 19035333.0,
      "step": 9395
    },
    {
      "entropy": 5.989636850357056,
      "epoch": 0.8214629030848554,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004939386618694508,
      "loss": 5.6532,
      "mean_token_accuracy": 0.17245471328496934,
      "num_tokens": 19046279.0,
      "step": 9400
    },
    {
      "entropy": 5.854677820205689,
      "epoch": 0.8218998514375601,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004939314756101547,
      "loss": 5.5931,
      "mean_token_accuracy": 0.173660010099411,
      "num_tokens": 19056506.0,
      "step": 9405
    },
    {
      "entropy": 5.914020204544068,
      "epoch": 0.8223367997902647,
      "grad_norm": 1.0,
      "learning_rate": 0.0004939242851516384,
      "loss": 5.5644,
      "mean_token_accuracy": 0.1798592835664749,
      "num_tokens": 19067217.0,
      "step": 9410
    },
    {
      "entropy": 5.909345388412476,
      "epoch": 0.8227737481429696,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004939170904940395,
      "loss": 5.6357,
      "mean_token_accuracy": 0.17441294193267823,
      "num_tokens": 19079028.0,
      "step": 9415
    },
    {
      "entropy": 5.83937702178955,
      "epoch": 0.8232106964956742,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000493909891637496,
      "loss": 5.6011,
      "mean_token_accuracy": 0.18219101876020433,
      "num_tokens": 19088236.0,
      "step": 9420
    },
    {
      "entropy": 5.892426919937134,
      "epoch": 0.8236476448483789,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004939026885821461,
      "loss": 5.5759,
      "mean_token_accuracy": 0.1818554624915123,
      "num_tokens": 19099598.0,
      "step": 9425
    },
    {
      "entropy": 5.9143633365631105,
      "epoch": 0.8240845932010836,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000493895481328128,
      "loss": 5.6141,
      "mean_token_accuracy": 0.18086249977350236,
      "num_tokens": 19110347.0,
      "step": 9430
    },
    {
      "entropy": 5.9694703102111815,
      "epoch": 0.8245215415537883,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004938882698755798,
      "loss": 5.5911,
      "mean_token_accuracy": 0.17643582820892334,
      "num_tokens": 19120643.0,
      "step": 9435
    },
    {
      "entropy": 5.856112623214722,
      "epoch": 0.824958489906493,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00049388105422464,
      "loss": 5.5634,
      "mean_token_accuracy": 0.18109374344348908,
      "num_tokens": 19130209.0,
      "step": 9440
    },
    {
      "entropy": 5.8979089736938475,
      "epoch": 0.8253954382591978,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004938738343754467,
      "loss": 5.5505,
      "mean_token_accuracy": 0.17858443558216094,
      "num_tokens": 19139196.0,
      "step": 9445
    },
    {
      "entropy": 5.932520532608033,
      "epoch": 0.8258323866119025,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004938666103281389,
      "loss": 5.5543,
      "mean_token_accuracy": 0.1779303580522537,
      "num_tokens": 19149618.0,
      "step": 9450
    },
    {
      "entropy": 5.910298776626587,
      "epoch": 0.8262693349646072,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004938593820828545,
      "loss": 5.5421,
      "mean_token_accuracy": 0.18083546906709672,
      "num_tokens": 19159875.0,
      "step": 9455
    },
    {
      "entropy": 5.8893217086792,
      "epoch": 0.8267062833173119,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004938521496397326,
      "loss": 5.5811,
      "mean_token_accuracy": 0.17831599712371826,
      "num_tokens": 19170403.0,
      "step": 9460
    },
    {
      "entropy": 5.892542028427124,
      "epoch": 0.8271432316700166,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004938449129989117,
      "loss": 5.5511,
      "mean_token_accuracy": 0.18554149568080902,
      "num_tokens": 19181187.0,
      "step": 9465
    },
    {
      "entropy": 5.892080545425415,
      "epoch": 0.8275801800227213,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004938376721605308,
      "loss": 5.6089,
      "mean_token_accuracy": 0.1745896816253662,
      "num_tokens": 19191328.0,
      "step": 9470
    },
    {
      "entropy": 5.8466956615448,
      "epoch": 0.8280171283754261,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004938304271247286,
      "loss": 5.5501,
      "mean_token_accuracy": 0.17977276742458342,
      "num_tokens": 19202041.0,
      "step": 9475
    },
    {
      "entropy": 5.899812793731689,
      "epoch": 0.8284540767281308,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004938231778916442,
      "loss": 5.5516,
      "mean_token_accuracy": 0.17843585312366486,
      "num_tokens": 19212016.0,
      "step": 9480
    },
    {
      "entropy": 5.8975804328918455,
      "epoch": 0.8288910250808355,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004938159244614167,
      "loss": 5.5397,
      "mean_token_accuracy": 0.18016014993190765,
      "num_tokens": 19222111.0,
      "step": 9485
    },
    {
      "entropy": 5.959759330749511,
      "epoch": 0.8293279734335401,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004938086668341849,
      "loss": 5.5254,
      "mean_token_accuracy": 0.18121186941862105,
      "num_tokens": 19231865.0,
      "step": 9490
    },
    {
      "entropy": 5.845223426818848,
      "epoch": 0.8297649217862448,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004938014050100884,
      "loss": 5.5349,
      "mean_token_accuracy": 0.18395625054836273,
      "num_tokens": 19242993.0,
      "step": 9495
    },
    {
      "entropy": 5.843985462188721,
      "epoch": 0.8302018701389495,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004937941389892662,
      "loss": 5.6041,
      "mean_token_accuracy": 0.17346795946359633,
      "num_tokens": 19252776.0,
      "step": 9500
    },
    {
      "entropy": 5.845558166503906,
      "epoch": 0.8306388184916543,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004937868687718577,
      "loss": 5.5453,
      "mean_token_accuracy": 0.17791442722082138,
      "num_tokens": 19263022.0,
      "step": 9505
    },
    {
      "entropy": 5.921742534637451,
      "epoch": 0.831075766844359,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004937795943580026,
      "loss": 5.5736,
      "mean_token_accuracy": 0.18393744081258773,
      "num_tokens": 19271859.0,
      "step": 9510
    },
    {
      "entropy": 5.834502410888672,
      "epoch": 0.8315127151970637,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00049377231574784,
      "loss": 5.5171,
      "mean_token_accuracy": 0.18394780158996582,
      "num_tokens": 19282771.0,
      "step": 9515
    },
    {
      "entropy": 5.878643035888672,
      "epoch": 0.8319496635497684,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004937650329415099,
      "loss": 5.5769,
      "mean_token_accuracy": 0.1726685032248497,
      "num_tokens": 19292985.0,
      "step": 9520
    },
    {
      "entropy": 5.8757857322692875,
      "epoch": 0.8323866119024731,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004937577459391519,
      "loss": 5.5869,
      "mean_token_accuracy": 0.18099523484706878,
      "num_tokens": 19303409.0,
      "step": 9525
    },
    {
      "entropy": 5.914814233779907,
      "epoch": 0.8328235602551778,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004937504547409056,
      "loss": 5.5847,
      "mean_token_accuracy": 0.17738027423620223,
      "num_tokens": 19312632.0,
      "step": 9530
    },
    {
      "entropy": 5.871352958679199,
      "epoch": 0.8332605086078826,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004937431593469109,
      "loss": 5.4756,
      "mean_token_accuracy": 0.18461024016141891,
      "num_tokens": 19322759.0,
      "step": 9535
    },
    {
      "entropy": 5.856506824493408,
      "epoch": 0.8336974569605873,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004937358597573079,
      "loss": 5.5193,
      "mean_token_accuracy": 0.18838415592908858,
      "num_tokens": 19333307.0,
      "step": 9540
    },
    {
      "entropy": 5.836728048324585,
      "epoch": 0.834134405313292,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004937285559722364,
      "loss": 5.6253,
      "mean_token_accuracy": 0.1746942549943924,
      "num_tokens": 19342528.0,
      "step": 9545
    },
    {
      "entropy": 5.923878812789917,
      "epoch": 0.8345713536659967,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004937212479918367,
      "loss": 5.58,
      "mean_token_accuracy": 0.17855777144432067,
      "num_tokens": 19352640.0,
      "step": 9550
    },
    {
      "entropy": 5.912781667709351,
      "epoch": 0.8350083020187014,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004937139358162488,
      "loss": 5.5461,
      "mean_token_accuracy": 0.17982480973005294,
      "num_tokens": 19363285.0,
      "step": 9555
    },
    {
      "entropy": 5.90484848022461,
      "epoch": 0.835445250371406,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000493706619445613,
      "loss": 5.5408,
      "mean_token_accuracy": 0.18171631544828415,
      "num_tokens": 19373141.0,
      "step": 9560
    },
    {
      "entropy": 5.816081142425537,
      "epoch": 0.8358821987241108,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004936992988800697,
      "loss": 5.5863,
      "mean_token_accuracy": 0.17704508751630782,
      "num_tokens": 19383091.0,
      "step": 9565
    },
    {
      "entropy": 5.906300449371338,
      "epoch": 0.8363191470768155,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004936919741197591,
      "loss": 5.5281,
      "mean_token_accuracy": 0.1786302626132965,
      "num_tokens": 19393163.0,
      "step": 9570
    },
    {
      "entropy": 5.848725080490112,
      "epoch": 0.8367560954295202,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004936846451648219,
      "loss": 5.4818,
      "mean_token_accuracy": 0.1840616539120674,
      "num_tokens": 19403475.0,
      "step": 9575
    },
    {
      "entropy": 5.782735157012939,
      "epoch": 0.8371930437822249,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004936773120153987,
      "loss": 5.5232,
      "mean_token_accuracy": 0.18107084333896636,
      "num_tokens": 19413302.0,
      "step": 9580
    },
    {
      "entropy": 5.890137672424316,
      "epoch": 0.8376299921349296,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00049366997467163,
      "loss": 5.5089,
      "mean_token_accuracy": 0.186966472864151,
      "num_tokens": 19423473.0,
      "step": 9585
    },
    {
      "entropy": 5.967378187179565,
      "epoch": 0.8380669404876344,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004936626331336567,
      "loss": 5.6276,
      "mean_token_accuracy": 0.17034847140312195,
      "num_tokens": 19433289.0,
      "step": 9590
    },
    {
      "entropy": 5.929351949691773,
      "epoch": 0.8385038888403391,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004936552874016194,
      "loss": 5.6052,
      "mean_token_accuracy": 0.17047603130340577,
      "num_tokens": 19442188.0,
      "step": 9595
    },
    {
      "entropy": 5.887839126586914,
      "epoch": 0.8389408371930438,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004936479374756592,
      "loss": 5.5978,
      "mean_token_accuracy": 0.17449402958154678,
      "num_tokens": 19452971.0,
      "step": 9600
    },
    {
      "entropy": 5.930010032653809,
      "epoch": 0.8393777855457485,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004936405833559169,
      "loss": 5.5859,
      "mean_token_accuracy": 0.1710500657558441,
      "num_tokens": 19463263.0,
      "step": 9605
    },
    {
      "entropy": 5.970377111434937,
      "epoch": 0.8398147338984532,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004936332250425337,
      "loss": 5.6083,
      "mean_token_accuracy": 0.17238924503326417,
      "num_tokens": 19473234.0,
      "step": 9610
    },
    {
      "entropy": 5.824743986129761,
      "epoch": 0.8402516822511579,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004936258625356507,
      "loss": 5.5719,
      "mean_token_accuracy": 0.17583155184984206,
      "num_tokens": 19483714.0,
      "step": 9615
    },
    {
      "entropy": 5.8039875507354735,
      "epoch": 0.8406886306038627,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004936184958354091,
      "loss": 5.47,
      "mean_token_accuracy": 0.19078062623739242,
      "num_tokens": 19493128.0,
      "step": 9620
    },
    {
      "entropy": 5.940363645553589,
      "epoch": 0.8411255789565674,
      "grad_norm": 1.75,
      "learning_rate": 0.00049361112494195,
      "loss": 5.6187,
      "mean_token_accuracy": 0.17033471167087555,
      "num_tokens": 19503401.0,
      "step": 9625
    },
    {
      "entropy": 5.892543363571167,
      "epoch": 0.8415625273092721,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004936037498554153,
      "loss": 5.6479,
      "mean_token_accuracy": 0.1778810828924179,
      "num_tokens": 19513050.0,
      "step": 9630
    },
    {
      "entropy": 5.88714747428894,
      "epoch": 0.8419994756619767,
      "grad_norm": 1.015625,
      "learning_rate": 0.000493596370575946,
      "loss": 5.5451,
      "mean_token_accuracy": 0.18293933868408202,
      "num_tokens": 19524534.0,
      "step": 9635
    },
    {
      "entropy": 5.914581441879273,
      "epoch": 0.8424364240146814,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004935889871036837,
      "loss": 5.5192,
      "mean_token_accuracy": 0.18300466239452362,
      "num_tokens": 19534863.0,
      "step": 9640
    },
    {
      "entropy": 5.920818185806274,
      "epoch": 0.8428733723673861,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004935815994387701,
      "loss": 5.6187,
      "mean_token_accuracy": 0.17816988676786422,
      "num_tokens": 19545037.0,
      "step": 9645
    },
    {
      "entropy": 5.884091472625732,
      "epoch": 0.8433103207200909,
      "grad_norm": 1.0,
      "learning_rate": 0.000493574207581347,
      "loss": 5.6043,
      "mean_token_accuracy": 0.1671932891011238,
      "num_tokens": 19555377.0,
      "step": 9650
    },
    {
      "entropy": 5.947559356689453,
      "epoch": 0.8437472690727956,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004935668115315559,
      "loss": 5.5967,
      "mean_token_accuracy": 0.1737657144665718,
      "num_tokens": 19565874.0,
      "step": 9655
    },
    {
      "entropy": 5.774116706848145,
      "epoch": 0.8441842174255003,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000493559411289539,
      "loss": 5.4556,
      "mean_token_accuracy": 0.19218890368938446,
      "num_tokens": 19575002.0,
      "step": 9660
    },
    {
      "entropy": 5.838811254501342,
      "epoch": 0.844621165778205,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000493552006855438,
      "loss": 5.5509,
      "mean_token_accuracy": 0.18497721403837203,
      "num_tokens": 19585863.0,
      "step": 9665
    },
    {
      "entropy": 5.925429010391236,
      "epoch": 0.8450581141309097,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004935445982293949,
      "loss": 5.5263,
      "mean_token_accuracy": 0.1784050092101097,
      "num_tokens": 19596787.0,
      "step": 9670
    },
    {
      "entropy": 5.90249547958374,
      "epoch": 0.8454950624836144,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004935371854115519,
      "loss": 5.6074,
      "mean_token_accuracy": 0.17342006266117097,
      "num_tokens": 19607114.0,
      "step": 9675
    },
    {
      "entropy": 5.856340074539185,
      "epoch": 0.8459320108363192,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004935297684020513,
      "loss": 5.4916,
      "mean_token_accuracy": 0.19011710584163666,
      "num_tokens": 19617334.0,
      "step": 9680
    },
    {
      "entropy": 5.892682266235352,
      "epoch": 0.8463689591890239,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004935223472010352,
      "loss": 5.5641,
      "mean_token_accuracy": 0.1784436285495758,
      "num_tokens": 19626844.0,
      "step": 9685
    },
    {
      "entropy": 5.865726327896118,
      "epoch": 0.8468059075417286,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004935149218086459,
      "loss": 5.5838,
      "mean_token_accuracy": 0.1768082559108734,
      "num_tokens": 19637897.0,
      "step": 9690
    },
    {
      "entropy": 5.883272504806518,
      "epoch": 0.8472428558944333,
      "grad_norm": 0.984375,
      "learning_rate": 0.000493507492225026,
      "loss": 5.6623,
      "mean_token_accuracy": 0.17329731583595276,
      "num_tokens": 19649161.0,
      "step": 9695
    },
    {
      "entropy": 5.937921762466431,
      "epoch": 0.847679804247138,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004935000584503178,
      "loss": 5.5637,
      "mean_token_accuracy": 0.17623288780450821,
      "num_tokens": 19659328.0,
      "step": 9700
    },
    {
      "entropy": 5.911883735656739,
      "epoch": 0.8481167525998426,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004934926204846641,
      "loss": 5.638,
      "mean_token_accuracy": 0.1751073494553566,
      "num_tokens": 19669196.0,
      "step": 9705
    },
    {
      "entropy": 5.835870695114136,
      "epoch": 0.8485537009525475,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004934851783282073,
      "loss": 5.5183,
      "mean_token_accuracy": 0.17718424797058105,
      "num_tokens": 19680019.0,
      "step": 9710
    },
    {
      "entropy": 5.897028875350952,
      "epoch": 0.8489906493052521,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004934777319810904,
      "loss": 5.6493,
      "mean_token_accuracy": 0.17690180540084838,
      "num_tokens": 19689391.0,
      "step": 9715
    },
    {
      "entropy": 5.9297308921813965,
      "epoch": 0.8494275976579568,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004934702814434561,
      "loss": 5.6268,
      "mean_token_accuracy": 0.17157294750213622,
      "num_tokens": 19699683.0,
      "step": 9720
    },
    {
      "entropy": 5.964580535888672,
      "epoch": 0.8498645460106615,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004934628267154474,
      "loss": 5.676,
      "mean_token_accuracy": 0.16948022544384003,
      "num_tokens": 19709844.0,
      "step": 9725
    },
    {
      "entropy": 5.967657136917114,
      "epoch": 0.8503014943633662,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000493455367797207,
      "loss": 5.6104,
      "mean_token_accuracy": 0.17415632009506227,
      "num_tokens": 19719926.0,
      "step": 9730
    },
    {
      "entropy": 5.872834825515747,
      "epoch": 0.8507384427160709,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004934479046888784,
      "loss": 5.5419,
      "mean_token_accuracy": 0.176617182046175,
      "num_tokens": 19730548.0,
      "step": 9735
    },
    {
      "entropy": 5.905636215209961,
      "epoch": 0.8511753910687757,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004934404373906045,
      "loss": 5.6157,
      "mean_token_accuracy": 0.1763654977083206,
      "num_tokens": 19740471.0,
      "step": 9740
    },
    {
      "entropy": 5.949840402603149,
      "epoch": 0.8516123394214804,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004934329659025283,
      "loss": 5.5375,
      "mean_token_accuracy": 0.1797020837664604,
      "num_tokens": 19749395.0,
      "step": 9745
    },
    {
      "entropy": 5.92762598991394,
      "epoch": 0.8520492877741851,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004934254902247936,
      "loss": 5.6156,
      "mean_token_accuracy": 0.16721540242433547,
      "num_tokens": 19759203.0,
      "step": 9750
    },
    {
      "entropy": 5.918326807022095,
      "epoch": 0.8524862361268898,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004934180103575436,
      "loss": 5.4751,
      "mean_token_accuracy": 0.1902475968003273,
      "num_tokens": 19768977.0,
      "step": 9755
    },
    {
      "entropy": 5.851624584197998,
      "epoch": 0.8529231844795945,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004934105263009216,
      "loss": 5.4865,
      "mean_token_accuracy": 0.17970294803380965,
      "num_tokens": 19778569.0,
      "step": 9760
    },
    {
      "entropy": 5.909655904769897,
      "epoch": 0.8533601328322992,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004934030380550713,
      "loss": 5.5858,
      "mean_token_accuracy": 0.18022527545690536,
      "num_tokens": 19787939.0,
      "step": 9765
    },
    {
      "entropy": 5.882731246948242,
      "epoch": 0.853797081185004,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004933955456201362,
      "loss": 5.5634,
      "mean_token_accuracy": 0.17719409018754959,
      "num_tokens": 19798407.0,
      "step": 9770
    },
    {
      "entropy": 5.958160543441773,
      "epoch": 0.8542340295377087,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004933880489962602,
      "loss": 5.5946,
      "mean_token_accuracy": 0.1775614559650421,
      "num_tokens": 19808495.0,
      "step": 9775
    },
    {
      "entropy": 5.870166015625,
      "epoch": 0.8546709778904134,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004933805481835868,
      "loss": 5.5451,
      "mean_token_accuracy": 0.17675618529319764,
      "num_tokens": 19818709.0,
      "step": 9780
    },
    {
      "entropy": 5.929715490341186,
      "epoch": 0.855107926243118,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004933730431822603,
      "loss": 5.6191,
      "mean_token_accuracy": 0.17871429324150084,
      "num_tokens": 19828865.0,
      "step": 9785
    },
    {
      "entropy": 5.8710166931152346,
      "epoch": 0.8555448745958227,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004933655339924243,
      "loss": 5.5355,
      "mean_token_accuracy": 0.1772872105240822,
      "num_tokens": 19838757.0,
      "step": 9790
    },
    {
      "entropy": 5.818740797042847,
      "epoch": 0.8559818229485275,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000493358020614223,
      "loss": 5.5716,
      "mean_token_accuracy": 0.18378417789936066,
      "num_tokens": 19848967.0,
      "step": 9795
    },
    {
      "entropy": 5.92359471321106,
      "epoch": 0.8564187713012322,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004933505030478003,
      "loss": 5.5907,
      "mean_token_accuracy": 0.17691970020532607,
      "num_tokens": 19859487.0,
      "step": 9800
    },
    {
      "entropy": 5.957725095748901,
      "epoch": 0.8568557196539369,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004933429812933005,
      "loss": 5.6482,
      "mean_token_accuracy": 0.175689797103405,
      "num_tokens": 19870170.0,
      "step": 9805
    },
    {
      "entropy": 5.834233999252319,
      "epoch": 0.8572926680066416,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004933354553508681,
      "loss": 5.527,
      "mean_token_accuracy": 0.18239171504974366,
      "num_tokens": 19881031.0,
      "step": 9810
    },
    {
      "entropy": 5.9284279346466064,
      "epoch": 0.8577296163593463,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004933279252206472,
      "loss": 5.5827,
      "mean_token_accuracy": 0.17529007494449617,
      "num_tokens": 19891082.0,
      "step": 9815
    },
    {
      "entropy": 5.954549932479859,
      "epoch": 0.858166564712051,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004933203909027823,
      "loss": 5.5532,
      "mean_token_accuracy": 0.17859879434108733,
      "num_tokens": 19900626.0,
      "step": 9820
    },
    {
      "entropy": 5.858586692810059,
      "epoch": 0.8586035130647558,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004933128523974179,
      "loss": 5.5565,
      "mean_token_accuracy": 0.18286827504634856,
      "num_tokens": 19910660.0,
      "step": 9825
    },
    {
      "entropy": 5.852932977676391,
      "epoch": 0.8590404614174605,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004933053097046986,
      "loss": 5.5288,
      "mean_token_accuracy": 0.17911264747381211,
      "num_tokens": 19920476.0,
      "step": 9830
    },
    {
      "entropy": 6.0233715057373045,
      "epoch": 0.8594774097701652,
      "grad_norm": 0.953125,
      "learning_rate": 0.000493297762824769,
      "loss": 5.6591,
      "mean_token_accuracy": 0.1716077819466591,
      "num_tokens": 19931151.0,
      "step": 9835
    },
    {
      "entropy": 5.92575888633728,
      "epoch": 0.8599143581228699,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000493290211757774,
      "loss": 5.5518,
      "mean_token_accuracy": 0.1709921345114708,
      "num_tokens": 19942469.0,
      "step": 9840
    },
    {
      "entropy": 5.8174220561981205,
      "epoch": 0.8603513064755746,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004932826565038584,
      "loss": 5.6248,
      "mean_token_accuracy": 0.17683601081371308,
      "num_tokens": 19953055.0,
      "step": 9845
    },
    {
      "entropy": 5.815469884872437,
      "epoch": 0.8607882548282793,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004932750970631669,
      "loss": 5.5762,
      "mean_token_accuracy": 0.17882972806692124,
      "num_tokens": 19963944.0,
      "step": 9850
    },
    {
      "entropy": 5.903191328048706,
      "epoch": 0.8612252031809841,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004932675334358448,
      "loss": 5.499,
      "mean_token_accuracy": 0.18087289482355118,
      "num_tokens": 19973913.0,
      "step": 9855
    },
    {
      "entropy": 5.887454843521118,
      "epoch": 0.8616621515336887,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000493259965622037,
      "loss": 5.5722,
      "mean_token_accuracy": 0.1793134927749634,
      "num_tokens": 19984130.0,
      "step": 9860
    },
    {
      "entropy": 5.862949466705322,
      "epoch": 0.8620990998863934,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004932523936218887,
      "loss": 5.4727,
      "mean_token_accuracy": 0.1815623626112938,
      "num_tokens": 19993840.0,
      "step": 9865
    },
    {
      "entropy": 5.867349910736084,
      "epoch": 0.8625360482390981,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000493244817435545,
      "loss": 5.5603,
      "mean_token_accuracy": 0.1786809518933296,
      "num_tokens": 20004228.0,
      "step": 9870
    },
    {
      "entropy": 5.910563564300537,
      "epoch": 0.8629729965918028,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004932372370631515,
      "loss": 5.5683,
      "mean_token_accuracy": 0.18252712488174438,
      "num_tokens": 20012620.0,
      "step": 9875
    },
    {
      "entropy": 5.89579496383667,
      "epoch": 0.8634099449445075,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004932296525048535,
      "loss": 5.655,
      "mean_token_accuracy": 0.1779656082391739,
      "num_tokens": 20022892.0,
      "step": 9880
    },
    {
      "entropy": 5.930139255523682,
      "epoch": 0.8638468932972123,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004932220637607962,
      "loss": 5.5254,
      "mean_token_accuracy": 0.18499893099069595,
      "num_tokens": 20032788.0,
      "step": 9885
    },
    {
      "entropy": 5.8371728420257565,
      "epoch": 0.864283841649917,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004932144708311256,
      "loss": 5.5572,
      "mean_token_accuracy": 0.18030902594327927,
      "num_tokens": 20043176.0,
      "step": 9890
    },
    {
      "entropy": 5.896143627166748,
      "epoch": 0.8647207900026217,
      "grad_norm": 0.9375,
      "learning_rate": 0.000493206873715987,
      "loss": 5.5705,
      "mean_token_accuracy": 0.17639719396829606,
      "num_tokens": 20052298.0,
      "step": 9895
    },
    {
      "entropy": 5.915764951705933,
      "epoch": 0.8651577383553264,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004931992724155262,
      "loss": 5.5125,
      "mean_token_accuracy": 0.17809111773967742,
      "num_tokens": 20063163.0,
      "step": 9900
    },
    {
      "entropy": 5.92968487739563,
      "epoch": 0.8655946867080311,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000493191666929889,
      "loss": 5.6021,
      "mean_token_accuracy": 0.17220424711704255,
      "num_tokens": 20073644.0,
      "step": 9905
    },
    {
      "entropy": 5.875770568847656,
      "epoch": 0.8660316350607358,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004931840572592214,
      "loss": 5.5322,
      "mean_token_accuracy": 0.18254094272851945,
      "num_tokens": 20084457.0,
      "step": 9910
    },
    {
      "entropy": 5.80830945968628,
      "epoch": 0.8664685834134406,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004931764434036692,
      "loss": 5.4293,
      "mean_token_accuracy": 0.1866819068789482,
      "num_tokens": 20094153.0,
      "step": 9915
    },
    {
      "entropy": 5.890798091888428,
      "epoch": 0.8669055317661453,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004931688253633784,
      "loss": 5.6241,
      "mean_token_accuracy": 0.17258946746587753,
      "num_tokens": 20104031.0,
      "step": 9920
    },
    {
      "entropy": 5.903915596008301,
      "epoch": 0.86734248011885,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004931612031384953,
      "loss": 5.5387,
      "mean_token_accuracy": 0.18548626750707625,
      "num_tokens": 20114431.0,
      "step": 9925
    },
    {
      "entropy": 5.8377320766448975,
      "epoch": 0.8677794284715546,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000493153576729166,
      "loss": 5.4773,
      "mean_token_accuracy": 0.18269651979207993,
      "num_tokens": 20124161.0,
      "step": 9930
    },
    {
      "entropy": 5.873267602920532,
      "epoch": 0.8682163768242593,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004931459461355368,
      "loss": 5.5143,
      "mean_token_accuracy": 0.18479189276695251,
      "num_tokens": 20134227.0,
      "step": 9935
    },
    {
      "entropy": 5.83510308265686,
      "epoch": 0.868653325176964,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004931383113577541,
      "loss": 5.5321,
      "mean_token_accuracy": 0.17842700332403183,
      "num_tokens": 20143912.0,
      "step": 9940
    },
    {
      "entropy": 5.823039817810058,
      "epoch": 0.8690902735296688,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004931306723959641,
      "loss": 5.4599,
      "mean_token_accuracy": 0.18060714453458787,
      "num_tokens": 20153858.0,
      "step": 9945
    },
    {
      "entropy": 5.933502054214477,
      "epoch": 0.8695272218823735,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004931230292503137,
      "loss": 5.5777,
      "mean_token_accuracy": 0.17402014136314392,
      "num_tokens": 20164309.0,
      "step": 9950
    },
    {
      "entropy": 5.845436096191406,
      "epoch": 0.8699641702350782,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004931153819209492,
      "loss": 5.5618,
      "mean_token_accuracy": 0.1801157847046852,
      "num_tokens": 20173949.0,
      "step": 9955
    },
    {
      "entropy": 5.861681413650513,
      "epoch": 0.8704011185877829,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004931077304080175,
      "loss": 5.5413,
      "mean_token_accuracy": 0.17875294685363768,
      "num_tokens": 20184206.0,
      "step": 9960
    },
    {
      "entropy": 5.981503582000732,
      "epoch": 0.8708380669404876,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004931000747116651,
      "loss": 5.6613,
      "mean_token_accuracy": 0.17332300543785095,
      "num_tokens": 20194430.0,
      "step": 9965
    },
    {
      "entropy": 5.939148807525635,
      "epoch": 0.8712750152931923,
      "grad_norm": 1.03125,
      "learning_rate": 0.000493092414832039,
      "loss": 5.5297,
      "mean_token_accuracy": 0.18071459978818893,
      "num_tokens": 20205004.0,
      "step": 9970
    },
    {
      "entropy": 5.799662017822266,
      "epoch": 0.8717119636458971,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004930847507692862,
      "loss": 5.5117,
      "mean_token_accuracy": 0.18048176318407058,
      "num_tokens": 20214698.0,
      "step": 9975
    },
    {
      "entropy": 5.918452548980713,
      "epoch": 0.8721489119986018,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004930770825235536,
      "loss": 5.6035,
      "mean_token_accuracy": 0.17404989451169967,
      "num_tokens": 20226534.0,
      "step": 9980
    },
    {
      "entropy": 5.9507581233978275,
      "epoch": 0.8725858603513065,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004930694100949882,
      "loss": 5.5843,
      "mean_token_accuracy": 0.17571040540933608,
      "num_tokens": 20237414.0,
      "step": 9985
    },
    {
      "entropy": 5.825314569473266,
      "epoch": 0.8730228087040112,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004930617334837373,
      "loss": 5.4104,
      "mean_token_accuracy": 0.18649402409791946,
      "num_tokens": 20246948.0,
      "step": 9990
    },
    {
      "entropy": 5.8454193592071535,
      "epoch": 0.8734597570567159,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000493054052689948,
      "loss": 5.6312,
      "mean_token_accuracy": 0.17353965789079667,
      "num_tokens": 20257281.0,
      "step": 9995
    },
    {
      "entropy": 5.869400548934936,
      "epoch": 0.8738967054094207,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004930463677137679,
      "loss": 5.6136,
      "mean_token_accuracy": 0.17321180552244186,
      "num_tokens": 20269041.0,
      "step": 10000
    },
    {
      "entropy": 5.872251319885254,
      "epoch": 0.8743336537621254,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004930386785553441,
      "loss": 5.6523,
      "mean_token_accuracy": 0.1701941519975662,
      "num_tokens": 20279688.0,
      "step": 10005
    },
    {
      "entropy": 5.916024923324585,
      "epoch": 0.87477060211483,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004930309852148242,
      "loss": 5.5542,
      "mean_token_accuracy": 0.1785753145813942,
      "num_tokens": 20288913.0,
      "step": 10010
    },
    {
      "entropy": 5.959272241592407,
      "epoch": 0.8752075504675347,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004930232876923558,
      "loss": 5.5909,
      "mean_token_accuracy": 0.17882099598646164,
      "num_tokens": 20299009.0,
      "step": 10015
    },
    {
      "entropy": 5.852214002609253,
      "epoch": 0.8756444988202394,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004930155859880865,
      "loss": 5.5676,
      "mean_token_accuracy": 0.17179852724075317,
      "num_tokens": 20309478.0,
      "step": 10020
    },
    {
      "entropy": 5.889416790008545,
      "epoch": 0.8760814471729441,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000493007880102164,
      "loss": 5.5803,
      "mean_token_accuracy": 0.1810441195964813,
      "num_tokens": 20319451.0,
      "step": 10025
    },
    {
      "entropy": 5.898276710510254,
      "epoch": 0.8765183955256489,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000493000170034736,
      "loss": 5.5416,
      "mean_token_accuracy": 0.18051966428756713,
      "num_tokens": 20328497.0,
      "step": 10030
    },
    {
      "entropy": 5.836324644088745,
      "epoch": 0.8769553438783536,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004929924557859507,
      "loss": 5.4538,
      "mean_token_accuracy": 0.1915641412138939,
      "num_tokens": 20338584.0,
      "step": 10035
    },
    {
      "entropy": 5.846433305740357,
      "epoch": 0.8773922922310583,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004929847373559557,
      "loss": 5.5286,
      "mean_token_accuracy": 0.17832885682582855,
      "num_tokens": 20348808.0,
      "step": 10040
    },
    {
      "entropy": 5.850480270385742,
      "epoch": 0.877829240583763,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004929770147448992,
      "loss": 5.5208,
      "mean_token_accuracy": 0.18457551598548888,
      "num_tokens": 20358301.0,
      "step": 10045
    },
    {
      "entropy": 5.8839273929595945,
      "epoch": 0.8782661889364677,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004929692879529295,
      "loss": 5.5347,
      "mean_token_accuracy": 0.17698329985141753,
      "num_tokens": 20369032.0,
      "step": 10050
    },
    {
      "entropy": 5.894958114624023,
      "epoch": 0.8787031372891724,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004929615569801946,
      "loss": 5.5742,
      "mean_token_accuracy": 0.18061530888080596,
      "num_tokens": 20379101.0,
      "step": 10055
    },
    {
      "entropy": 5.813940048217773,
      "epoch": 0.8791400856418772,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004929538218268428,
      "loss": 5.5121,
      "mean_token_accuracy": 0.18003027588129045,
      "num_tokens": 20388864.0,
      "step": 10060
    },
    {
      "entropy": 5.956840515136719,
      "epoch": 0.8795770339945819,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004929460824930226,
      "loss": 5.6486,
      "mean_token_accuracy": 0.17600364238023758,
      "num_tokens": 20398402.0,
      "step": 10065
    },
    {
      "entropy": 5.826817798614502,
      "epoch": 0.8800139823472866,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004929383389788821,
      "loss": 5.5314,
      "mean_token_accuracy": 0.1770249530673027,
      "num_tokens": 20408209.0,
      "step": 10070
    },
    {
      "entropy": 5.886365842819214,
      "epoch": 0.8804509306999913,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004929305912845703,
      "loss": 5.5618,
      "mean_token_accuracy": 0.17938753515481948,
      "num_tokens": 20418117.0,
      "step": 10075
    },
    {
      "entropy": 6.007940435409546,
      "epoch": 0.8808878790526959,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004929228394102355,
      "loss": 5.594,
      "mean_token_accuracy": 0.17499574273824692,
      "num_tokens": 20428347.0,
      "step": 10080
    },
    {
      "entropy": 5.926840353012085,
      "epoch": 0.8813248274054006,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004929150833560263,
      "loss": 5.54,
      "mean_token_accuracy": 0.17692749947309494,
      "num_tokens": 20437804.0,
      "step": 10085
    },
    {
      "entropy": 5.9140397071838375,
      "epoch": 0.8817617757581054,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004929073231220917,
      "loss": 5.557,
      "mean_token_accuracy": 0.1728794515132904,
      "num_tokens": 20447600.0,
      "step": 10090
    },
    {
      "entropy": 5.927258539199829,
      "epoch": 0.8821987241108101,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004928995587085806,
      "loss": 5.5637,
      "mean_token_accuracy": 0.17810529321432114,
      "num_tokens": 20458023.0,
      "step": 10095
    },
    {
      "entropy": 5.900887680053711,
      "epoch": 0.8826356724635148,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004928917901156416,
      "loss": 5.5075,
      "mean_token_accuracy": 0.19023204743862152,
      "num_tokens": 20468195.0,
      "step": 10100
    },
    {
      "entropy": 5.882900190353394,
      "epoch": 0.8830726208162195,
      "grad_norm": 0.9375,
      "learning_rate": 0.000492884017343424,
      "loss": 5.6587,
      "mean_token_accuracy": 0.17208839058876038,
      "num_tokens": 20479250.0,
      "step": 10105
    },
    {
      "entropy": 5.937636613845825,
      "epoch": 0.8835095691689242,
      "grad_norm": 1.0,
      "learning_rate": 0.0004928762403920767,
      "loss": 5.5395,
      "mean_token_accuracy": 0.17951249778270723,
      "num_tokens": 20489587.0,
      "step": 10110
    },
    {
      "entropy": 5.888387250900268,
      "epoch": 0.8839465175216289,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000492868459261749,
      "loss": 5.5532,
      "mean_token_accuracy": 0.18695807456970215,
      "num_tokens": 20499788.0,
      "step": 10115
    },
    {
      "entropy": 5.842326021194458,
      "epoch": 0.8843834658743337,
      "grad_norm": 1.046875,
      "learning_rate": 0.00049286067395259,
      "loss": 5.5757,
      "mean_token_accuracy": 0.17764627784490586,
      "num_tokens": 20509716.0,
      "step": 10120
    },
    {
      "entropy": 5.880858421325684,
      "epoch": 0.8848204142270384,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004928528844647493,
      "loss": 5.5148,
      "mean_token_accuracy": 0.1812182381749153,
      "num_tokens": 20519020.0,
      "step": 10125
    },
    {
      "entropy": 5.9112848281860355,
      "epoch": 0.8852573625797431,
      "grad_norm": 1.015625,
      "learning_rate": 0.000492845090798376,
      "loss": 5.5795,
      "mean_token_accuracy": 0.17597134560346603,
      "num_tokens": 20528997.0,
      "step": 10130
    },
    {
      "entropy": 5.914076328277588,
      "epoch": 0.8856943109324478,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004928372929536197,
      "loss": 5.5841,
      "mean_token_accuracy": 0.17958904057741165,
      "num_tokens": 20538599.0,
      "step": 10135
    },
    {
      "entropy": 5.92730565071106,
      "epoch": 0.8861312592851525,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004928294909306301,
      "loss": 5.5173,
      "mean_token_accuracy": 0.1841164708137512,
      "num_tokens": 20547664.0,
      "step": 10140
    },
    {
      "entropy": 5.896369171142578,
      "epoch": 0.8865682076378572,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004928216847295565,
      "loss": 5.6245,
      "mean_token_accuracy": 0.17157290875911713,
      "num_tokens": 20557860.0,
      "step": 10145
    },
    {
      "entropy": 5.875310325622559,
      "epoch": 0.887005155990562,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000492813874350549,
      "loss": 5.5722,
      "mean_token_accuracy": 0.17932030856609343,
      "num_tokens": 20569003.0,
      "step": 10150
    },
    {
      "entropy": 5.912645292282105,
      "epoch": 0.8874421043432666,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004928060597937573,
      "loss": 5.5513,
      "mean_token_accuracy": 0.18090423941612244,
      "num_tokens": 20578402.0,
      "step": 10155
    },
    {
      "entropy": 5.907726144790649,
      "epoch": 0.8878790526959713,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004927982410593312,
      "loss": 5.4621,
      "mean_token_accuracy": 0.1850451111793518,
      "num_tokens": 20587874.0,
      "step": 10160
    },
    {
      "entropy": 5.8861829280853275,
      "epoch": 0.888316001048676,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004927904181474207,
      "loss": 5.489,
      "mean_token_accuracy": 0.17698864340782167,
      "num_tokens": 20597270.0,
      "step": 10165
    },
    {
      "entropy": 5.888202047348022,
      "epoch": 0.8887529494013807,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004927825910581758,
      "loss": 5.5973,
      "mean_token_accuracy": 0.18423565924167634,
      "num_tokens": 20607898.0,
      "step": 10170
    },
    {
      "entropy": 5.871435117721558,
      "epoch": 0.8891898977540854,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004927747597917467,
      "loss": 5.4961,
      "mean_token_accuracy": 0.18824073523283005,
      "num_tokens": 20617300.0,
      "step": 10175
    },
    {
      "entropy": 5.920030164718628,
      "epoch": 0.8896268461067902,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004927669243482836,
      "loss": 5.5922,
      "mean_token_accuracy": 0.17418672442436217,
      "num_tokens": 20626582.0,
      "step": 10180
    },
    {
      "entropy": 5.887486839294434,
      "epoch": 0.8900637944594949,
      "grad_norm": 1.0,
      "learning_rate": 0.0004927590847279369,
      "loss": 5.5486,
      "mean_token_accuracy": 0.1831202581524849,
      "num_tokens": 20635802.0,
      "step": 10185
    },
    {
      "entropy": 5.908946228027344,
      "epoch": 0.8905007428121996,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004927512409308568,
      "loss": 5.5336,
      "mean_token_accuracy": 0.18365781456232072,
      "num_tokens": 20646057.0,
      "step": 10190
    },
    {
      "entropy": 5.91832160949707,
      "epoch": 0.8909376911649043,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004927433929571937,
      "loss": 5.6493,
      "mean_token_accuracy": 0.17197972983121873,
      "num_tokens": 20656511.0,
      "step": 10195
    },
    {
      "entropy": 5.928015565872192,
      "epoch": 0.891374639517609,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004927355408070982,
      "loss": 5.5426,
      "mean_token_accuracy": 0.1826781988143921,
      "num_tokens": 20666479.0,
      "step": 10200
    },
    {
      "entropy": 5.870953607559204,
      "epoch": 0.8918115878703138,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004927276844807211,
      "loss": 5.4702,
      "mean_token_accuracy": 0.1830660343170166,
      "num_tokens": 20676830.0,
      "step": 10205
    },
    {
      "entropy": 5.806000185012818,
      "epoch": 0.8922485362230185,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004927198239782129,
      "loss": 5.5199,
      "mean_token_accuracy": 0.18208872228860856,
      "num_tokens": 20687316.0,
      "step": 10210
    },
    {
      "entropy": 5.921666479110717,
      "epoch": 0.8926854845757232,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004927119592997242,
      "loss": 5.6046,
      "mean_token_accuracy": 0.17423308789730071,
      "num_tokens": 20697901.0,
      "step": 10215
    },
    {
      "entropy": 5.909894895553589,
      "epoch": 0.8931224329284279,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004927040904454062,
      "loss": 5.512,
      "mean_token_accuracy": 0.18528199791908265,
      "num_tokens": 20707468.0,
      "step": 10220
    },
    {
      "entropy": 5.899530410766602,
      "epoch": 0.8935593812811325,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004926962174154097,
      "loss": 5.5751,
      "mean_token_accuracy": 0.1781720846891403,
      "num_tokens": 20717098.0,
      "step": 10225
    },
    {
      "entropy": 5.968518352508545,
      "epoch": 0.8939963296338372,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004926883402098857,
      "loss": 5.6619,
      "mean_token_accuracy": 0.17226873114705085,
      "num_tokens": 20727217.0,
      "step": 10230
    },
    {
      "entropy": 5.9176122665405275,
      "epoch": 0.894433277986542,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004926804588289852,
      "loss": 5.6377,
      "mean_token_accuracy": 0.17419704794883728,
      "num_tokens": 20737397.0,
      "step": 10235
    },
    {
      "entropy": 5.804396343231201,
      "epoch": 0.8948702263392467,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004926725732728595,
      "loss": 5.394,
      "mean_token_accuracy": 0.19114880710840226,
      "num_tokens": 20746712.0,
      "step": 10240
    },
    {
      "entropy": 5.865988874435425,
      "epoch": 0.8953071746919514,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004926646835416597,
      "loss": 5.5715,
      "mean_token_accuracy": 0.17788508981466294,
      "num_tokens": 20757161.0,
      "step": 10245
    },
    {
      "entropy": 5.901945400238037,
      "epoch": 0.8957441230446561,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004926567896355372,
      "loss": 5.4869,
      "mean_token_accuracy": 0.18968162536621094,
      "num_tokens": 20766478.0,
      "step": 10250
    },
    {
      "entropy": 5.839264154434204,
      "epoch": 0.8961810713973608,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004926488915546435,
      "loss": 5.533,
      "mean_token_accuracy": 0.18148647546768187,
      "num_tokens": 20776317.0,
      "step": 10255
    },
    {
      "entropy": 5.868514776229858,
      "epoch": 0.8966180197500655,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00049264098929913,
      "loss": 5.474,
      "mean_token_accuracy": 0.18167668133974074,
      "num_tokens": 20786534.0,
      "step": 10260
    },
    {
      "entropy": 5.8360902786254885,
      "epoch": 0.8970549681027703,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004926330828691482,
      "loss": 5.4974,
      "mean_token_accuracy": 0.18105001747608185,
      "num_tokens": 20796791.0,
      "step": 10265
    },
    {
      "entropy": 5.896015310287476,
      "epoch": 0.897491916455475,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004926251722648499,
      "loss": 5.522,
      "mean_token_accuracy": 0.1834713786840439,
      "num_tokens": 20806834.0,
      "step": 10270
    },
    {
      "entropy": 5.8043347835540775,
      "epoch": 0.8979288648081797,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004926172574863868,
      "loss": 5.5261,
      "mean_token_accuracy": 0.18029796332120895,
      "num_tokens": 20816664.0,
      "step": 10275
    },
    {
      "entropy": 5.849050521850586,
      "epoch": 0.8983658131608844,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004926093385339105,
      "loss": 5.6602,
      "mean_token_accuracy": 0.17523777186870576,
      "num_tokens": 20826936.0,
      "step": 10280
    },
    {
      "entropy": 5.912503242492676,
      "epoch": 0.8988027615135891,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004926014154075732,
      "loss": 5.5062,
      "mean_token_accuracy": 0.18557548373937607,
      "num_tokens": 20836732.0,
      "step": 10285
    },
    {
      "entropy": 5.9119956493377686,
      "epoch": 0.8992397098662938,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004925934881075268,
      "loss": 5.5533,
      "mean_token_accuracy": 0.1848953276872635,
      "num_tokens": 20848093.0,
      "step": 10290
    },
    {
      "entropy": 5.902449321746826,
      "epoch": 0.8996766582189986,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004925855566339232,
      "loss": 5.5157,
      "mean_token_accuracy": 0.17692889720201493,
      "num_tokens": 20858678.0,
      "step": 10295
    },
    {
      "entropy": 5.819490003585815,
      "epoch": 0.9001136065717033,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004925776209869146,
      "loss": 5.5026,
      "mean_token_accuracy": 0.1805671125650406,
      "num_tokens": 20869319.0,
      "step": 10300
    },
    {
      "entropy": 5.852482509613037,
      "epoch": 0.9005505549244079,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004925696811666532,
      "loss": 5.4729,
      "mean_token_accuracy": 0.18739187717437744,
      "num_tokens": 20879475.0,
      "step": 10305
    },
    {
      "entropy": 5.840228080749512,
      "epoch": 0.9009875032771126,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004925617371732914,
      "loss": 5.4968,
      "mean_token_accuracy": 0.18502204418182372,
      "num_tokens": 20889479.0,
      "step": 10310
    },
    {
      "entropy": 5.862395095825195,
      "epoch": 0.9014244516298173,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004925537890069815,
      "loss": 5.5841,
      "mean_token_accuracy": 0.17851855605840683,
      "num_tokens": 20899864.0,
      "step": 10315
    },
    {
      "entropy": 5.9003345489501955,
      "epoch": 0.901861399982522,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004925458366678758,
      "loss": 5.5603,
      "mean_token_accuracy": 0.18014353960752488,
      "num_tokens": 20909006.0,
      "step": 10320
    },
    {
      "entropy": 5.8858787536621096,
      "epoch": 0.9022983483352268,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004925378801561269,
      "loss": 5.512,
      "mean_token_accuracy": 0.18125748932361602,
      "num_tokens": 20917998.0,
      "step": 10325
    },
    {
      "entropy": 5.8430403709411625,
      "epoch": 0.9027352966879315,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004925299194718876,
      "loss": 5.5288,
      "mean_token_accuracy": 0.18188500106334687,
      "num_tokens": 20928407.0,
      "step": 10330
    },
    {
      "entropy": 5.807545852661133,
      "epoch": 0.9031722450406362,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004925219546153105,
      "loss": 5.5508,
      "mean_token_accuracy": 0.1774570107460022,
      "num_tokens": 20938016.0,
      "step": 10335
    },
    {
      "entropy": 5.926233005523682,
      "epoch": 0.9036091933933409,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004925139855865483,
      "loss": 5.6108,
      "mean_token_accuracy": 0.16926117986440659,
      "num_tokens": 20948578.0,
      "step": 10340
    },
    {
      "entropy": 5.916742038726807,
      "epoch": 0.9040461417460456,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004925060123857539,
      "loss": 5.4602,
      "mean_token_accuracy": 0.18140792101621628,
      "num_tokens": 20958193.0,
      "step": 10345
    },
    {
      "entropy": 5.856251430511475,
      "epoch": 0.9044830900987503,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004924980350130802,
      "loss": 5.5167,
      "mean_token_accuracy": 0.18984605222940446,
      "num_tokens": 20968092.0,
      "step": 10350
    },
    {
      "entropy": 5.826005697250366,
      "epoch": 0.9049200384514551,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004924900534686801,
      "loss": 5.549,
      "mean_token_accuracy": 0.1740780144929886,
      "num_tokens": 20978303.0,
      "step": 10355
    },
    {
      "entropy": 5.917559385299683,
      "epoch": 0.9053569868041598,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004924820677527069,
      "loss": 5.5677,
      "mean_token_accuracy": 0.17436209172010422,
      "num_tokens": 20988331.0,
      "step": 10360
    },
    {
      "entropy": 5.80744047164917,
      "epoch": 0.9057939351568645,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004924740778653137,
      "loss": 5.4808,
      "mean_token_accuracy": 0.18194967955350877,
      "num_tokens": 20997785.0,
      "step": 10365
    },
    {
      "entropy": 5.888814067840576,
      "epoch": 0.9062308835095692,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004924660838066538,
      "loss": 5.581,
      "mean_token_accuracy": 0.1825806528329849,
      "num_tokens": 21008026.0,
      "step": 10370
    },
    {
      "entropy": 5.864372444152832,
      "epoch": 0.9066678318622738,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004924580855768803,
      "loss": 5.523,
      "mean_token_accuracy": 0.18474316745996475,
      "num_tokens": 21017993.0,
      "step": 10375
    },
    {
      "entropy": 5.832047319412231,
      "epoch": 0.9071047802149785,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004924500831761468,
      "loss": 5.5426,
      "mean_token_accuracy": 0.1832178384065628,
      "num_tokens": 21028149.0,
      "step": 10380
    },
    {
      "entropy": 5.992540121078491,
      "epoch": 0.9075417285676833,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004924420766046069,
      "loss": 5.6461,
      "mean_token_accuracy": 0.1720384880900383,
      "num_tokens": 21039229.0,
      "step": 10385
    },
    {
      "entropy": 5.929101657867432,
      "epoch": 0.907978676920388,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004924340658624139,
      "loss": 5.5891,
      "mean_token_accuracy": 0.17733438611030578,
      "num_tokens": 21049358.0,
      "step": 10390
    },
    {
      "entropy": 5.8900596618652346,
      "epoch": 0.9084156252730927,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004924260509497217,
      "loss": 5.6205,
      "mean_token_accuracy": 0.1757276862859726,
      "num_tokens": 21060385.0,
      "step": 10395
    },
    {
      "entropy": 5.925390625,
      "epoch": 0.9088525736257974,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004924180318666838,
      "loss": 5.6055,
      "mean_token_accuracy": 0.17468367218971254,
      "num_tokens": 21070423.0,
      "step": 10400
    },
    {
      "entropy": 6.004812240600586,
      "epoch": 0.9092895219785021,
      "grad_norm": 1.0,
      "learning_rate": 0.0004924100086134541,
      "loss": 5.5865,
      "mean_token_accuracy": 0.17406658381223677,
      "num_tokens": 21080505.0,
      "step": 10405
    },
    {
      "entropy": 5.85016360282898,
      "epoch": 0.9097264703312069,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004924019811901866,
      "loss": 5.5616,
      "mean_token_accuracy": 0.17591047137975693,
      "num_tokens": 21090875.0,
      "step": 10410
    },
    {
      "entropy": 5.873072910308838,
      "epoch": 0.9101634186839116,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004923939495970352,
      "loss": 5.4833,
      "mean_token_accuracy": 0.18095023185014725,
      "num_tokens": 21101427.0,
      "step": 10415
    },
    {
      "entropy": 5.971972990036011,
      "epoch": 0.9106003670366163,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004923859138341539,
      "loss": 5.6462,
      "mean_token_accuracy": 0.17218540161848067,
      "num_tokens": 21112691.0,
      "step": 10420
    },
    {
      "entropy": 5.933805799484253,
      "epoch": 0.911037315389321,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004923778739016969,
      "loss": 5.6226,
      "mean_token_accuracy": 0.17266084402799606,
      "num_tokens": 21123932.0,
      "step": 10425
    },
    {
      "entropy": 5.933885145187378,
      "epoch": 0.9114742637420257,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004923698297998183,
      "loss": 5.6316,
      "mean_token_accuracy": 0.16780108362436294,
      "num_tokens": 21135380.0,
      "step": 10430
    },
    {
      "entropy": 5.909259605407715,
      "epoch": 0.9119112120947304,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004923617815286725,
      "loss": 5.5283,
      "mean_token_accuracy": 0.1823616236448288,
      "num_tokens": 21144951.0,
      "step": 10435
    },
    {
      "entropy": 5.858248853683472,
      "epoch": 0.9123481604474352,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004923537290884138,
      "loss": 5.4362,
      "mean_token_accuracy": 0.18482770770788193,
      "num_tokens": 21155919.0,
      "step": 10440
    },
    {
      "entropy": 5.84056715965271,
      "epoch": 0.9127851088001399,
      "grad_norm": 1.0,
      "learning_rate": 0.0004923456724791967,
      "loss": 5.4964,
      "mean_token_accuracy": 0.18173424601554872,
      "num_tokens": 21166186.0,
      "step": 10445
    },
    {
      "entropy": 5.812300062179565,
      "epoch": 0.9132220571528445,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004923376117011758,
      "loss": 5.4731,
      "mean_token_accuracy": 0.18686916530132294,
      "num_tokens": 21176043.0,
      "step": 10450
    },
    {
      "entropy": 5.821030902862549,
      "epoch": 0.9136590055055492,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004923295467545055,
      "loss": 5.4827,
      "mean_token_accuracy": 0.18539202958345413,
      "num_tokens": 21185677.0,
      "step": 10455
    },
    {
      "entropy": 5.931668519973755,
      "epoch": 0.9140959538582539,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004923214776393407,
      "loss": 5.6182,
      "mean_token_accuracy": 0.17387147545814513,
      "num_tokens": 21195977.0,
      "step": 10460
    },
    {
      "entropy": 5.955788278579712,
      "epoch": 0.9145329022109586,
      "grad_norm": 0.96875,
      "learning_rate": 0.000492313404355836,
      "loss": 5.5692,
      "mean_token_accuracy": 0.17782811373472213,
      "num_tokens": 21205454.0,
      "step": 10465
    },
    {
      "entropy": 5.877263498306275,
      "epoch": 0.9149698505636634,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004923053269041464,
      "loss": 5.4674,
      "mean_token_accuracy": 0.18826386332511902,
      "num_tokens": 21215700.0,
      "step": 10470
    },
    {
      "entropy": 5.84747543334961,
      "epoch": 0.9154067989163681,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004922972452844267,
      "loss": 5.5588,
      "mean_token_accuracy": 0.1823249265551567,
      "num_tokens": 21226544.0,
      "step": 10475
    },
    {
      "entropy": 5.9302825927734375,
      "epoch": 0.9158437472690728,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004922891594968321,
      "loss": 5.6131,
      "mean_token_accuracy": 0.17311323434114456,
      "num_tokens": 21236792.0,
      "step": 10480
    },
    {
      "entropy": 5.898540210723877,
      "epoch": 0.9162806956217775,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004922810695415175,
      "loss": 5.511,
      "mean_token_accuracy": 0.1847800001502037,
      "num_tokens": 21247039.0,
      "step": 10485
    },
    {
      "entropy": 5.855885648727417,
      "epoch": 0.9167176439744822,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004922729754186381,
      "loss": 5.5178,
      "mean_token_accuracy": 0.18622775226831437,
      "num_tokens": 21257008.0,
      "step": 10490
    },
    {
      "entropy": 5.868865346908569,
      "epoch": 0.9171545923271869,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004922648771283492,
      "loss": 5.6512,
      "mean_token_accuracy": 0.177496300637722,
      "num_tokens": 21267638.0,
      "step": 10495
    },
    {
      "entropy": 5.953221797943115,
      "epoch": 0.9175915406798917,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004922567746708062,
      "loss": 5.4973,
      "mean_token_accuracy": 0.18335384279489517,
      "num_tokens": 21277327.0,
      "step": 10500
    },
    {
      "entropy": 5.8986701488494875,
      "epoch": 0.9180284890325964,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004922486680461642,
      "loss": 5.5798,
      "mean_token_accuracy": 0.174693563580513,
      "num_tokens": 21287579.0,
      "step": 10505
    },
    {
      "entropy": 5.836841440200805,
      "epoch": 0.9184654373853011,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004922405572545792,
      "loss": 5.4846,
      "mean_token_accuracy": 0.1809658244252205,
      "num_tokens": 21297888.0,
      "step": 10510
    },
    {
      "entropy": 5.922271060943603,
      "epoch": 0.9189023857380058,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004922324422962063,
      "loss": 5.582,
      "mean_token_accuracy": 0.17963477224111557,
      "num_tokens": 21308506.0,
      "step": 10515
    },
    {
      "entropy": 5.945352268218994,
      "epoch": 0.9193393340907104,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004922243231712013,
      "loss": 5.6413,
      "mean_token_accuracy": 0.166617713868618,
      "num_tokens": 21319252.0,
      "step": 10520
    },
    {
      "entropy": 5.8993336200714115,
      "epoch": 0.9197762824434151,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00049221619987972,
      "loss": 5.5595,
      "mean_token_accuracy": 0.18465058803558348,
      "num_tokens": 21328829.0,
      "step": 10525
    },
    {
      "entropy": 5.831767940521241,
      "epoch": 0.9202132307961199,
      "grad_norm": 1.0,
      "learning_rate": 0.0004922080724219182,
      "loss": 5.3832,
      "mean_token_accuracy": 0.18810685575008393,
      "num_tokens": 21338524.0,
      "step": 10530
    },
    {
      "entropy": 5.89189076423645,
      "epoch": 0.9206501791488246,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004921999407979518,
      "loss": 5.5383,
      "mean_token_accuracy": 0.17759019434452056,
      "num_tokens": 21348559.0,
      "step": 10535
    },
    {
      "entropy": 5.879982614517212,
      "epoch": 0.9210871275015293,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004921918050079766,
      "loss": 5.5601,
      "mean_token_accuracy": 0.17813540995121002,
      "num_tokens": 21360763.0,
      "step": 10540
    },
    {
      "entropy": 5.90746283531189,
      "epoch": 0.921524075854234,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004921836650521489,
      "loss": 5.5787,
      "mean_token_accuracy": 0.17448193728923797,
      "num_tokens": 21370048.0,
      "step": 10545
    },
    {
      "entropy": 5.8857136249542235,
      "epoch": 0.9219610242069387,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004921755209306247,
      "loss": 5.5533,
      "mean_token_accuracy": 0.18427959978580474,
      "num_tokens": 21379760.0,
      "step": 10550
    },
    {
      "entropy": 5.928763723373413,
      "epoch": 0.9223979725596434,
      "grad_norm": 1.0,
      "learning_rate": 0.0004921673726435601,
      "loss": 5.5131,
      "mean_token_accuracy": 0.17845889329910278,
      "num_tokens": 21389712.0,
      "step": 10555
    },
    {
      "entropy": 5.923545885086059,
      "epoch": 0.9228349209123482,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004921592201911115,
      "loss": 5.5111,
      "mean_token_accuracy": 0.17694517076015473,
      "num_tokens": 21399515.0,
      "step": 10560
    },
    {
      "entropy": 5.836602640151978,
      "epoch": 0.9232718692650529,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004921510635734353,
      "loss": 5.4851,
      "mean_token_accuracy": 0.18607860654592515,
      "num_tokens": 21409704.0,
      "step": 10565
    },
    {
      "entropy": 5.820917844772339,
      "epoch": 0.9237088176177576,
      "grad_norm": 1.03125,
      "learning_rate": 0.000492142902790688,
      "loss": 5.486,
      "mean_token_accuracy": 0.1846115455031395,
      "num_tokens": 21419182.0,
      "step": 10570
    },
    {
      "entropy": 5.815049934387207,
      "epoch": 0.9241457659704623,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004921347378430259,
      "loss": 5.393,
      "mean_token_accuracy": 0.1893997609615326,
      "num_tokens": 21429749.0,
      "step": 10575
    },
    {
      "entropy": 5.945705461502075,
      "epoch": 0.924582714323167,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004921265687306059,
      "loss": 5.6312,
      "mean_token_accuracy": 0.1681041732430458,
      "num_tokens": 21440514.0,
      "step": 10580
    },
    {
      "entropy": 5.9030030250549315,
      "epoch": 0.9250196626758717,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004921183954535844,
      "loss": 5.5145,
      "mean_token_accuracy": 0.18145698606967925,
      "num_tokens": 21450647.0,
      "step": 10585
    },
    {
      "entropy": 5.87932825088501,
      "epoch": 0.9254566110285765,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004921102180121184,
      "loss": 5.5362,
      "mean_token_accuracy": 0.17695201337337493,
      "num_tokens": 21460233.0,
      "step": 10590
    },
    {
      "entropy": 5.874557399749756,
      "epoch": 0.9258935593812812,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004921020364063646,
      "loss": 5.5439,
      "mean_token_accuracy": 0.1866134449839592,
      "num_tokens": 21470207.0,
      "step": 10595
    },
    {
      "entropy": 5.8399687767028805,
      "epoch": 0.9263305077339858,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004920938506364801,
      "loss": 5.577,
      "mean_token_accuracy": 0.17700662463903427,
      "num_tokens": 21480658.0,
      "step": 10600
    },
    {
      "entropy": 5.833864307403564,
      "epoch": 0.9267674560866905,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004920856607026216,
      "loss": 5.4635,
      "mean_token_accuracy": 0.19368511587381362,
      "num_tokens": 21490566.0,
      "step": 10605
    },
    {
      "entropy": 5.889381027221679,
      "epoch": 0.9272044044393952,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004920774666049465,
      "loss": 5.5675,
      "mean_token_accuracy": 0.18101192861795426,
      "num_tokens": 21500752.0,
      "step": 10610
    },
    {
      "entropy": 5.892794561386109,
      "epoch": 0.9276413527921,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004920692683436118,
      "loss": 5.5453,
      "mean_token_accuracy": 0.17699371725320817,
      "num_tokens": 21509802.0,
      "step": 10615
    },
    {
      "entropy": 5.8747509002685545,
      "epoch": 0.9280783011448047,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004920610659187748,
      "loss": 5.6431,
      "mean_token_accuracy": 0.17033015042543412,
      "num_tokens": 21520962.0,
      "step": 10620
    },
    {
      "entropy": 5.935116863250732,
      "epoch": 0.9285152494975094,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004920528593305929,
      "loss": 5.6039,
      "mean_token_accuracy": 0.1758645310997963,
      "num_tokens": 21530580.0,
      "step": 10625
    },
    {
      "entropy": 5.867668485641479,
      "epoch": 0.9289521978502141,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004920446485792233,
      "loss": 5.5349,
      "mean_token_accuracy": 0.17940372675657273,
      "num_tokens": 21541431.0,
      "step": 10630
    },
    {
      "entropy": 5.878044509887696,
      "epoch": 0.9293891462029188,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004920364336648236,
      "loss": 5.5244,
      "mean_token_accuracy": 0.1866188034415245,
      "num_tokens": 21551012.0,
      "step": 10635
    },
    {
      "entropy": 5.818696117401123,
      "epoch": 0.9298260945556235,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004920282145875514,
      "loss": 5.5097,
      "mean_token_accuracy": 0.18398429602384567,
      "num_tokens": 21560676.0,
      "step": 10640
    },
    {
      "entropy": 5.851223182678223,
      "epoch": 0.9302630429083283,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004920199913475644,
      "loss": 5.4642,
      "mean_token_accuracy": 0.193056857585907,
      "num_tokens": 21570848.0,
      "step": 10645
    },
    {
      "entropy": 5.866589784622192,
      "epoch": 0.930699991261033,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004920117639450202,
      "loss": 5.5646,
      "mean_token_accuracy": 0.1768198788166046,
      "num_tokens": 21582358.0,
      "step": 10650
    },
    {
      "entropy": 5.890108299255371,
      "epoch": 0.9311369396137377,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004920035323800767,
      "loss": 5.5007,
      "mean_token_accuracy": 0.1819705605506897,
      "num_tokens": 21591709.0,
      "step": 10655
    },
    {
      "entropy": 5.913449335098266,
      "epoch": 0.9315738879664424,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004919952966528917,
      "loss": 5.4713,
      "mean_token_accuracy": 0.1828962445259094,
      "num_tokens": 21601092.0,
      "step": 10660
    },
    {
      "entropy": 5.815505075454712,
      "epoch": 0.932010836319147,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004919870567636233,
      "loss": 5.4928,
      "mean_token_accuracy": 0.1811884820461273,
      "num_tokens": 21611003.0,
      "step": 10665
    },
    {
      "entropy": 5.882794952392578,
      "epoch": 0.9324477846718517,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004919788127124293,
      "loss": 5.571,
      "mean_token_accuracy": 0.17606115490198135,
      "num_tokens": 21621528.0,
      "step": 10670
    },
    {
      "entropy": 5.847782039642334,
      "epoch": 0.9328847330245565,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000491970564499468,
      "loss": 5.4915,
      "mean_token_accuracy": 0.1919265791773796,
      "num_tokens": 21631431.0,
      "step": 10675
    },
    {
      "entropy": 5.849430227279663,
      "epoch": 0.9333216813772612,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004919623121248976,
      "loss": 5.5222,
      "mean_token_accuracy": 0.1849410966038704,
      "num_tokens": 21640717.0,
      "step": 10680
    },
    {
      "entropy": 5.831820249557495,
      "epoch": 0.9337586297299659,
      "grad_norm": 1.0,
      "learning_rate": 0.0004919540555888765,
      "loss": 5.5021,
      "mean_token_accuracy": 0.186005200445652,
      "num_tokens": 21650493.0,
      "step": 10685
    },
    {
      "entropy": 5.867669248580933,
      "epoch": 0.9341955780826706,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004919457948915628,
      "loss": 5.5057,
      "mean_token_accuracy": 0.18769694864749908,
      "num_tokens": 21660062.0,
      "step": 10690
    },
    {
      "entropy": 5.817519474029541,
      "epoch": 0.9346325264353753,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004919375300331151,
      "loss": 5.4931,
      "mean_token_accuracy": 0.18738001585006714,
      "num_tokens": 21669364.0,
      "step": 10695
    },
    {
      "entropy": 5.865865993499756,
      "epoch": 0.93506947478808,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004919292610136919,
      "loss": 5.5747,
      "mean_token_accuracy": 0.17036822587251663,
      "num_tokens": 21680793.0,
      "step": 10700
    },
    {
      "entropy": 5.91989369392395,
      "epoch": 0.9355064231407848,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004919209878334518,
      "loss": 5.5798,
      "mean_token_accuracy": 0.1781994432210922,
      "num_tokens": 21690862.0,
      "step": 10705
    },
    {
      "entropy": 5.878885078430176,
      "epoch": 0.9359433714934895,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004919127104925535,
      "loss": 5.4915,
      "mean_token_accuracy": 0.18311809450387956,
      "num_tokens": 21700665.0,
      "step": 10710
    },
    {
      "entropy": 5.882126903533935,
      "epoch": 0.9363803198461942,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004919044289911557,
      "loss": 5.5411,
      "mean_token_accuracy": 0.18559230118989944,
      "num_tokens": 21711128.0,
      "step": 10715
    },
    {
      "entropy": 5.9233317375183105,
      "epoch": 0.9368172681988989,
      "grad_norm": 1.0,
      "learning_rate": 0.0004918961433294173,
      "loss": 5.5815,
      "mean_token_accuracy": 0.17413249760866165,
      "num_tokens": 21720233.0,
      "step": 10720
    },
    {
      "entropy": 5.886805295944214,
      "epoch": 0.9372542165516036,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004918878535074973,
      "loss": 5.523,
      "mean_token_accuracy": 0.1882713705301285,
      "num_tokens": 21730980.0,
      "step": 10725
    },
    {
      "entropy": 5.816906356811524,
      "epoch": 0.9376911649043083,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004918795595255545,
      "loss": 5.5451,
      "mean_token_accuracy": 0.18678274750709534,
      "num_tokens": 21742802.0,
      "step": 10730
    },
    {
      "entropy": 5.879956579208374,
      "epoch": 0.9381281132570131,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004918712613837483,
      "loss": 5.5572,
      "mean_token_accuracy": 0.17947341799736022,
      "num_tokens": 21752743.0,
      "step": 10735
    },
    {
      "entropy": 5.917039251327514,
      "epoch": 0.9385650616097178,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004918629590822375,
      "loss": 5.5421,
      "mean_token_accuracy": 0.18346861898899078,
      "num_tokens": 21763198.0,
      "step": 10740
    },
    {
      "entropy": 5.882467222213745,
      "epoch": 0.9390020099624224,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004918546526211816,
      "loss": 5.5329,
      "mean_token_accuracy": 0.18546292781829835,
      "num_tokens": 21773030.0,
      "step": 10745
    },
    {
      "entropy": 5.885100507736206,
      "epoch": 0.9394389583151271,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004918463420007398,
      "loss": 5.4934,
      "mean_token_accuracy": 0.18284184485673904,
      "num_tokens": 21783796.0,
      "step": 10750
    },
    {
      "entropy": 5.8951808452606205,
      "epoch": 0.9398759066678318,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004918380272210715,
      "loss": 5.5264,
      "mean_token_accuracy": 0.18065590113401414,
      "num_tokens": 21792504.0,
      "step": 10755
    },
    {
      "entropy": 5.930423641204834,
      "epoch": 0.9403128550205365,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004918297082823362,
      "loss": 5.5735,
      "mean_token_accuracy": 0.17817942947149276,
      "num_tokens": 21802764.0,
      "step": 10760
    },
    {
      "entropy": 5.95289945602417,
      "epoch": 0.9407498033732413,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004918213851846935,
      "loss": 5.5486,
      "mean_token_accuracy": 0.17517430037260057,
      "num_tokens": 21813051.0,
      "step": 10765
    },
    {
      "entropy": 5.851390695571899,
      "epoch": 0.941186751725946,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004918130579283031,
      "loss": 5.5813,
      "mean_token_accuracy": 0.1796059787273407,
      "num_tokens": 21822941.0,
      "step": 10770
    },
    {
      "entropy": 5.910475635528565,
      "epoch": 0.9416237000786507,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004918047265133246,
      "loss": 5.5147,
      "mean_token_accuracy": 0.17823912501335143,
      "num_tokens": 21833225.0,
      "step": 10775
    },
    {
      "entropy": 5.854309892654419,
      "epoch": 0.9420606484313554,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004917963909399178,
      "loss": 5.5086,
      "mean_token_accuracy": 0.18462529480457307,
      "num_tokens": 21842360.0,
      "step": 10780
    },
    {
      "entropy": 5.869366645812988,
      "epoch": 0.9424975967840601,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004917880512082427,
      "loss": 5.5322,
      "mean_token_accuracy": 0.17773087620735167,
      "num_tokens": 21852515.0,
      "step": 10785
    },
    {
      "entropy": 5.954308319091797,
      "epoch": 0.9429345451367648,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004917797073184591,
      "loss": 5.6153,
      "mean_token_accuracy": 0.1706952303647995,
      "num_tokens": 21863040.0,
      "step": 10790
    },
    {
      "entropy": 5.924374723434449,
      "epoch": 0.9433714934894696,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004917713592707271,
      "loss": 5.56,
      "mean_token_accuracy": 0.1795996055006981,
      "num_tokens": 21872101.0,
      "step": 10795
    },
    {
      "entropy": 5.909116172790528,
      "epoch": 0.9438084418421743,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004917630070652069,
      "loss": 5.53,
      "mean_token_accuracy": 0.184675495326519,
      "num_tokens": 21882148.0,
      "step": 10800
    },
    {
      "entropy": 5.8620603561401365,
      "epoch": 0.944245390194879,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004917546507020587,
      "loss": 5.5185,
      "mean_token_accuracy": 0.18681368827819825,
      "num_tokens": 21891832.0,
      "step": 10805
    },
    {
      "entropy": 5.89509220123291,
      "epoch": 0.9446823385475837,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004917462901814427,
      "loss": 5.5057,
      "mean_token_accuracy": 0.18230771720409394,
      "num_tokens": 21901544.0,
      "step": 10810
    },
    {
      "entropy": 5.909299898147583,
      "epoch": 0.9451192869002883,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004917379255035192,
      "loss": 5.5745,
      "mean_token_accuracy": 0.1808429479598999,
      "num_tokens": 21911690.0,
      "step": 10815
    },
    {
      "entropy": 5.953626441955566,
      "epoch": 0.9455562352529931,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004917295566684487,
      "loss": 5.5765,
      "mean_token_accuracy": 0.1728408619761467,
      "num_tokens": 21922968.0,
      "step": 10820
    },
    {
      "entropy": 5.806056213378906,
      "epoch": 0.9459931836056978,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004917211836763919,
      "loss": 5.4145,
      "mean_token_accuracy": 0.18656312376260759,
      "num_tokens": 21932904.0,
      "step": 10825
    },
    {
      "entropy": 5.8119213581085205,
      "epoch": 0.9464301319584025,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004917128065275091,
      "loss": 5.4376,
      "mean_token_accuracy": 0.19190144687891006,
      "num_tokens": 21943065.0,
      "step": 10830
    },
    {
      "entropy": 5.895235967636109,
      "epoch": 0.9468670803111072,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004917044252219612,
      "loss": 5.5307,
      "mean_token_accuracy": 0.17217247784137726,
      "num_tokens": 21953448.0,
      "step": 10835
    },
    {
      "entropy": 5.92166748046875,
      "epoch": 0.9473040286638119,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004916960397599088,
      "loss": 5.5123,
      "mean_token_accuracy": 0.1827612817287445,
      "num_tokens": 21963374.0,
      "step": 10840
    },
    {
      "entropy": 5.883824443817138,
      "epoch": 0.9477409770165166,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004916876501415129,
      "loss": 5.5584,
      "mean_token_accuracy": 0.17463902980089188,
      "num_tokens": 21973726.0,
      "step": 10845
    },
    {
      "entropy": 5.7936896800994875,
      "epoch": 0.9481779253692214,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004916792563669342,
      "loss": 5.4165,
      "mean_token_accuracy": 0.18814979493618011,
      "num_tokens": 21983203.0,
      "step": 10850
    },
    {
      "entropy": 5.855782127380371,
      "epoch": 0.9486148737219261,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004916708584363339,
      "loss": 5.5489,
      "mean_token_accuracy": 0.18156953603029252,
      "num_tokens": 21993435.0,
      "step": 10855
    },
    {
      "entropy": 5.977449226379394,
      "epoch": 0.9490518220746308,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000491662456349873,
      "loss": 5.5819,
      "mean_token_accuracy": 0.17745113670825957,
      "num_tokens": 22003230.0,
      "step": 10860
    },
    {
      "entropy": 5.911139917373657,
      "epoch": 0.9494887704273355,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004916540501077127,
      "loss": 5.5106,
      "mean_token_accuracy": 0.18332463651895523,
      "num_tokens": 22012910.0,
      "step": 10865
    },
    {
      "entropy": 5.858813333511352,
      "epoch": 0.9499257187800402,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004916456397100141,
      "loss": 5.4773,
      "mean_token_accuracy": 0.18232961595058442,
      "num_tokens": 22022802.0,
      "step": 10870
    },
    {
      "entropy": 5.890202617645263,
      "epoch": 0.9503626671327449,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004916372251569386,
      "loss": 5.5875,
      "mean_token_accuracy": 0.17649697363376618,
      "num_tokens": 22032669.0,
      "step": 10875
    },
    {
      "entropy": 5.866869592666626,
      "epoch": 0.9507996154854497,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004916288064486477,
      "loss": 5.4884,
      "mean_token_accuracy": 0.18898449838161469,
      "num_tokens": 22042477.0,
      "step": 10880
    },
    {
      "entropy": 5.905757045745849,
      "epoch": 0.9512365638381544,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004916203835853027,
      "loss": 5.5427,
      "mean_token_accuracy": 0.18169863671064376,
      "num_tokens": 22052891.0,
      "step": 10885
    },
    {
      "entropy": 5.833992004394531,
      "epoch": 0.951673512190859,
      "grad_norm": 1.0,
      "learning_rate": 0.0004916119565670652,
      "loss": 5.4816,
      "mean_token_accuracy": 0.19188932478427886,
      "num_tokens": 22062413.0,
      "step": 10890
    },
    {
      "entropy": 5.92659215927124,
      "epoch": 0.9521104605435637,
      "grad_norm": 0.96875,
      "learning_rate": 0.000491603525394097,
      "loss": 5.5757,
      "mean_token_accuracy": 0.1789309337735176,
      "num_tokens": 22072565.0,
      "step": 10895
    },
    {
      "entropy": 5.865817594528198,
      "epoch": 0.9525474088962684,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004915950900665596,
      "loss": 5.5394,
      "mean_token_accuracy": 0.17780586034059526,
      "num_tokens": 22082910.0,
      "step": 10900
    },
    {
      "entropy": 5.780698633193969,
      "epoch": 0.9529843572489731,
      "grad_norm": 1.03125,
      "learning_rate": 0.000491586650584615,
      "loss": 5.4477,
      "mean_token_accuracy": 0.19230101108551026,
      "num_tokens": 22092636.0,
      "step": 10905
    },
    {
      "entropy": 5.931867408752441,
      "epoch": 0.9534213056016779,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004915782069484248,
      "loss": 5.5387,
      "mean_token_accuracy": 0.1800084963440895,
      "num_tokens": 22102232.0,
      "step": 10910
    },
    {
      "entropy": 5.991722059249878,
      "epoch": 0.9538582539543826,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004915697591581511,
      "loss": 5.5927,
      "mean_token_accuracy": 0.1769523113965988,
      "num_tokens": 22113592.0,
      "step": 10915
    },
    {
      "entropy": 5.852837944030762,
      "epoch": 0.9542952023070873,
      "grad_norm": 0.90625,
      "learning_rate": 0.000491561307213956,
      "loss": 5.5501,
      "mean_token_accuracy": 0.18109481185674667,
      "num_tokens": 22123731.0,
      "step": 10920
    },
    {
      "entropy": 5.879301500320435,
      "epoch": 0.954732150659792,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004915528511160016,
      "loss": 5.5758,
      "mean_token_accuracy": 0.17398652732372283,
      "num_tokens": 22133801.0,
      "step": 10925
    },
    {
      "entropy": 5.953863191604614,
      "epoch": 0.9551690990124967,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004915443908644501,
      "loss": 5.5504,
      "mean_token_accuracy": 0.18185410648584366,
      "num_tokens": 22143939.0,
      "step": 10930
    },
    {
      "entropy": 5.840577268600464,
      "epoch": 0.9556060473652014,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004915359264594636,
      "loss": 5.5558,
      "mean_token_accuracy": 0.1789732351899147,
      "num_tokens": 22154954.0,
      "step": 10935
    },
    {
      "entropy": 5.894484806060791,
      "epoch": 0.9560429957179062,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004915274579012045,
      "loss": 5.5633,
      "mean_token_accuracy": 0.17984941005706787,
      "num_tokens": 22165225.0,
      "step": 10940
    },
    {
      "entropy": 5.959781360626221,
      "epoch": 0.9564799440706109,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004915189851898355,
      "loss": 5.5714,
      "mean_token_accuracy": 0.17338949888944627,
      "num_tokens": 22174891.0,
      "step": 10945
    },
    {
      "entropy": 5.825815010070801,
      "epoch": 0.9569168924233156,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004915105083255187,
      "loss": 5.5307,
      "mean_token_accuracy": 0.18498843014240265,
      "num_tokens": 22185255.0,
      "step": 10950
    },
    {
      "entropy": 5.8620476722717285,
      "epoch": 0.9573538407760203,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000491502027308417,
      "loss": 5.5943,
      "mean_token_accuracy": 0.17625785171985625,
      "num_tokens": 22196124.0,
      "step": 10955
    },
    {
      "entropy": 5.9080198287963865,
      "epoch": 0.957790789128725,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000491493542138693,
      "loss": 5.5305,
      "mean_token_accuracy": 0.18330998420715333,
      "num_tokens": 22207563.0,
      "step": 10960
    },
    {
      "entropy": 5.88185682296753,
      "epoch": 0.9582277374814296,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004914850528165094,
      "loss": 5.5499,
      "mean_token_accuracy": 0.18151160925626755,
      "num_tokens": 22218527.0,
      "step": 10965
    },
    {
      "entropy": 5.899251890182495,
      "epoch": 0.9586646858341344,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000491476559342029,
      "loss": 5.5285,
      "mean_token_accuracy": 0.17689513713121413,
      "num_tokens": 22229538.0,
      "step": 10970
    },
    {
      "entropy": 5.908556365966797,
      "epoch": 0.9591016341868391,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004914680617154148,
      "loss": 5.46,
      "mean_token_accuracy": 0.18598510771989823,
      "num_tokens": 22240047.0,
      "step": 10975
    },
    {
      "entropy": 5.824999666213989,
      "epoch": 0.9595385825395438,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004914595599368297,
      "loss": 5.4243,
      "mean_token_accuracy": 0.1894690290093422,
      "num_tokens": 22250440.0,
      "step": 10980
    },
    {
      "entropy": 5.868528175354004,
      "epoch": 0.9599755308922485,
      "grad_norm": 1.0,
      "learning_rate": 0.0004914510540064369,
      "loss": 5.4985,
      "mean_token_accuracy": 0.1780909702181816,
      "num_tokens": 22259909.0,
      "step": 10985
    },
    {
      "entropy": 5.91636471748352,
      "epoch": 0.9604124792449532,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004914425439243994,
      "loss": 5.5607,
      "mean_token_accuracy": 0.17821907103061677,
      "num_tokens": 22270358.0,
      "step": 10990
    },
    {
      "entropy": 5.89264030456543,
      "epoch": 0.9608494275976579,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004914340296908805,
      "loss": 5.5858,
      "mean_token_accuracy": 0.1800413504242897,
      "num_tokens": 22280591.0,
      "step": 10995
    },
    {
      "entropy": 5.947883653640747,
      "epoch": 0.9612863759503627,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004914255113060435,
      "loss": 5.5416,
      "mean_token_accuracy": 0.18413980752229692,
      "num_tokens": 22291067.0,
      "step": 11000
    },
    {
      "entropy": 5.850947046279908,
      "epoch": 0.9617233243030674,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004914169887700519,
      "loss": 5.482,
      "mean_token_accuracy": 0.18742386251688004,
      "num_tokens": 22300919.0,
      "step": 11005
    },
    {
      "entropy": 5.903344964981079,
      "epoch": 0.9621602726557721,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004914084620830689,
      "loss": 5.5879,
      "mean_token_accuracy": 0.17408562451601028,
      "num_tokens": 22311562.0,
      "step": 11010
    },
    {
      "entropy": 5.918695974349975,
      "epoch": 0.9625972210084768,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004913999312452583,
      "loss": 5.5318,
      "mean_token_accuracy": 0.17773458510637283,
      "num_tokens": 22322980.0,
      "step": 11015
    },
    {
      "entropy": 5.905825281143189,
      "epoch": 0.9630341693611815,
      "grad_norm": 1.0,
      "learning_rate": 0.0004913913962567836,
      "loss": 5.4475,
      "mean_token_accuracy": 0.18617327213287355,
      "num_tokens": 22333242.0,
      "step": 11020
    },
    {
      "entropy": 5.8967842102050785,
      "epoch": 0.9634711177138863,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004913828571178085,
      "loss": 5.5425,
      "mean_token_accuracy": 0.1808249235153198,
      "num_tokens": 22343618.0,
      "step": 11025
    },
    {
      "entropy": 5.850569343566894,
      "epoch": 0.963908066066591,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004913743138284969,
      "loss": 5.5418,
      "mean_token_accuracy": 0.17831170856952666,
      "num_tokens": 22354249.0,
      "step": 11030
    },
    {
      "entropy": 5.94581618309021,
      "epoch": 0.9643450144192957,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004913657663890124,
      "loss": 5.494,
      "mean_token_accuracy": 0.18199047446250916,
      "num_tokens": 22363824.0,
      "step": 11035
    },
    {
      "entropy": 5.933486461639404,
      "epoch": 0.9647819627720003,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004913572147995193,
      "loss": 5.4852,
      "mean_token_accuracy": 0.18615226149559022,
      "num_tokens": 22373470.0,
      "step": 11040
    },
    {
      "entropy": 5.835460042953491,
      "epoch": 0.965218911124705,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004913486590601814,
      "loss": 5.4579,
      "mean_token_accuracy": 0.17929461300373079,
      "num_tokens": 22382999.0,
      "step": 11045
    },
    {
      "entropy": 5.83815221786499,
      "epoch": 0.9656558594774097,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004913400991711629,
      "loss": 5.518,
      "mean_token_accuracy": 0.18693996965885162,
      "num_tokens": 22393415.0,
      "step": 11050
    },
    {
      "entropy": 5.863249444961548,
      "epoch": 0.9660928078301145,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004913315351326279,
      "loss": 5.4635,
      "mean_token_accuracy": 0.18628067374229432,
      "num_tokens": 22403040.0,
      "step": 11055
    },
    {
      "entropy": 5.870234489440918,
      "epoch": 0.9665297561828192,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004913229669447406,
      "loss": 5.4768,
      "mean_token_accuracy": 0.1815590113401413,
      "num_tokens": 22413434.0,
      "step": 11060
    },
    {
      "entropy": 5.857769918441773,
      "epoch": 0.9669667045355239,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004913143946076656,
      "loss": 5.6197,
      "mean_token_accuracy": 0.17082108706235885,
      "num_tokens": 22424446.0,
      "step": 11065
    },
    {
      "entropy": 5.88768367767334,
      "epoch": 0.9674036528882286,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004913058181215671,
      "loss": 5.5495,
      "mean_token_accuracy": 0.1817133530974388,
      "num_tokens": 22433928.0,
      "step": 11070
    },
    {
      "entropy": 5.939979076385498,
      "epoch": 0.9678406012409333,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004912972374866096,
      "loss": 5.512,
      "mean_token_accuracy": 0.1836621955037117,
      "num_tokens": 22443778.0,
      "step": 11075
    },
    {
      "entropy": 5.894754266738891,
      "epoch": 0.968277549593638,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004912886527029579,
      "loss": 5.5879,
      "mean_token_accuracy": 0.17806703746318817,
      "num_tokens": 22453423.0,
      "step": 11080
    },
    {
      "entropy": 5.914261770248413,
      "epoch": 0.9687144979463428,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004912800637707764,
      "loss": 5.503,
      "mean_token_accuracy": 0.18342599421739578,
      "num_tokens": 22464067.0,
      "step": 11085
    },
    {
      "entropy": 5.858634948730469,
      "epoch": 0.9691514462990475,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00049127147069023,
      "loss": 5.49,
      "mean_token_accuracy": 0.18554236739873886,
      "num_tokens": 22474332.0,
      "step": 11090
    },
    {
      "entropy": 5.876806259155273,
      "epoch": 0.9695883946517522,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004912628734614834,
      "loss": 5.518,
      "mean_token_accuracy": 0.18116229325532912,
      "num_tokens": 22485304.0,
      "step": 11095
    },
    {
      "entropy": 5.860860204696655,
      "epoch": 0.9700253430044569,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004912542720847016,
      "loss": 5.4357,
      "mean_token_accuracy": 0.18594945222139359,
      "num_tokens": 22494447.0,
      "step": 11100
    },
    {
      "entropy": 5.907275629043579,
      "epoch": 0.9704622913571616,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004912456665600495,
      "loss": 5.5285,
      "mean_token_accuracy": 0.18526405692100525,
      "num_tokens": 22503936.0,
      "step": 11105
    },
    {
      "entropy": 5.845846033096313,
      "epoch": 0.9708992397098662,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004912370568876923,
      "loss": 5.426,
      "mean_token_accuracy": 0.1848626971244812,
      "num_tokens": 22513960.0,
      "step": 11110
    },
    {
      "entropy": 5.9067408561706545,
      "epoch": 0.971336188062571,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004912284430677949,
      "loss": 5.5699,
      "mean_token_accuracy": 0.1814932107925415,
      "num_tokens": 22522309.0,
      "step": 11115
    },
    {
      "entropy": 5.883145189285278,
      "epoch": 0.9717731364152757,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004912198251005228,
      "loss": 5.4716,
      "mean_token_accuracy": 0.18539499640464782,
      "num_tokens": 22533510.0,
      "step": 11120
    },
    {
      "entropy": 5.867004776000977,
      "epoch": 0.9722100847679804,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004912112029860411,
      "loss": 5.4817,
      "mean_token_accuracy": 0.18086580634117128,
      "num_tokens": 22543644.0,
      "step": 11125
    },
    {
      "entropy": 5.953012609481812,
      "epoch": 0.9726470331206851,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004912025767245153,
      "loss": 5.6554,
      "mean_token_accuracy": 0.16925821602344512,
      "num_tokens": 22553475.0,
      "step": 11130
    },
    {
      "entropy": 5.877394771575927,
      "epoch": 0.9730839814733898,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004911939463161108,
      "loss": 5.4401,
      "mean_token_accuracy": 0.18776775896549225,
      "num_tokens": 22563190.0,
      "step": 11135
    },
    {
      "entropy": 5.874952936172486,
      "epoch": 0.9735209298260945,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000491185311760993,
      "loss": 5.4656,
      "mean_token_accuracy": 0.18926328122615815,
      "num_tokens": 22572069.0,
      "step": 11140
    },
    {
      "entropy": 5.92037878036499,
      "epoch": 0.9739578781787993,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004911766730593276,
      "loss": 5.6078,
      "mean_token_accuracy": 0.16880012154579163,
      "num_tokens": 22581597.0,
      "step": 11145
    },
    {
      "entropy": 5.944724178314209,
      "epoch": 0.974394826531504,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004911680302112806,
      "loss": 5.4704,
      "mean_token_accuracy": 0.18818040192127228,
      "num_tokens": 22591486.0,
      "step": 11150
    },
    {
      "entropy": 5.952447843551636,
      "epoch": 0.9748317748842087,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004911593832170173,
      "loss": 5.5337,
      "mean_token_accuracy": 0.17976175546646117,
      "num_tokens": 22601615.0,
      "step": 11155
    },
    {
      "entropy": 5.894178152084351,
      "epoch": 0.9752687232369134,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004911507320767038,
      "loss": 5.5157,
      "mean_token_accuracy": 0.17964272499084472,
      "num_tokens": 22611397.0,
      "step": 11160
    },
    {
      "entropy": 5.807279062271118,
      "epoch": 0.9757056715896181,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000491142076790506,
      "loss": 5.4886,
      "mean_token_accuracy": 0.1870104417204857,
      "num_tokens": 22621269.0,
      "step": 11165
    },
    {
      "entropy": 5.9143455028533936,
      "epoch": 0.9761426199423228,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004911334173585899,
      "loss": 5.4751,
      "mean_token_accuracy": 0.1851723909378052,
      "num_tokens": 22631782.0,
      "step": 11170
    },
    {
      "entropy": 5.885642194747925,
      "epoch": 0.9765795682950276,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004911247537811216,
      "loss": 5.5654,
      "mean_token_accuracy": 0.1787995308637619,
      "num_tokens": 22642240.0,
      "step": 11175
    },
    {
      "entropy": 5.914176416397095,
      "epoch": 0.9770165166477323,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004911160860582674,
      "loss": 5.5363,
      "mean_token_accuracy": 0.18191086798906325,
      "num_tokens": 22651977.0,
      "step": 11180
    },
    {
      "entropy": 5.922344064712524,
      "epoch": 0.977453465000437,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004911074141901933,
      "loss": 5.5411,
      "mean_token_accuracy": 0.1847665324807167,
      "num_tokens": 22662871.0,
      "step": 11185
    },
    {
      "entropy": 5.85600004196167,
      "epoch": 0.9778904133531416,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004910987381770657,
      "loss": 5.4744,
      "mean_token_accuracy": 0.19131015241146088,
      "num_tokens": 22672083.0,
      "step": 11190
    },
    {
      "entropy": 5.839209127426147,
      "epoch": 0.9783273617058463,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004910900580190512,
      "loss": 5.4894,
      "mean_token_accuracy": 0.18343789875507355,
      "num_tokens": 22681247.0,
      "step": 11195
    },
    {
      "entropy": 5.851170921325684,
      "epoch": 0.9787643100585511,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004910813737163161,
      "loss": 5.535,
      "mean_token_accuracy": 0.18498896956443786,
      "num_tokens": 22691231.0,
      "step": 11200
    },
    {
      "entropy": 5.889358806610107,
      "epoch": 0.9792012584112558,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000491072685269027,
      "loss": 5.5205,
      "mean_token_accuracy": 0.179593089222908,
      "num_tokens": 22700712.0,
      "step": 11205
    },
    {
      "entropy": 5.980083560943603,
      "epoch": 0.9796382067639605,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004910639926773506,
      "loss": 5.6376,
      "mean_token_accuracy": 0.1765113055706024,
      "num_tokens": 22710961.0,
      "step": 11210
    },
    {
      "entropy": 5.875549364089966,
      "epoch": 0.9800751551166652,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004910552959414538,
      "loss": 5.4845,
      "mean_token_accuracy": 0.1902923658490181,
      "num_tokens": 22720450.0,
      "step": 11215
    },
    {
      "entropy": 5.788787889480591,
      "epoch": 0.9805121034693699,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004910465950615029,
      "loss": 5.4638,
      "mean_token_accuracy": 0.18395162522792816,
      "num_tokens": 22730337.0,
      "step": 11220
    },
    {
      "entropy": 5.880448722839356,
      "epoch": 0.9809490518220746,
      "grad_norm": 1.375,
      "learning_rate": 0.0004910378900376651,
      "loss": 5.5691,
      "mean_token_accuracy": 0.18042166978120805,
      "num_tokens": 22742379.0,
      "step": 11225
    },
    {
      "entropy": 5.989414644241333,
      "epoch": 0.9813860001747794,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004910291808701077,
      "loss": 5.6217,
      "mean_token_accuracy": 0.1700190156698227,
      "num_tokens": 22753176.0,
      "step": 11230
    },
    {
      "entropy": 5.893676280975342,
      "epoch": 0.9818229485274841,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004910204675589972,
      "loss": 5.5253,
      "mean_token_accuracy": 0.18221170455217361,
      "num_tokens": 22763263.0,
      "step": 11235
    },
    {
      "entropy": 5.841779041290283,
      "epoch": 0.9822598968801888,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004910117501045009,
      "loss": 5.559,
      "mean_token_accuracy": 0.1792824611067772,
      "num_tokens": 22774630.0,
      "step": 11240
    },
    {
      "entropy": 5.954283428192139,
      "epoch": 0.9826968452328935,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004910030285067861,
      "loss": 5.5602,
      "mean_token_accuracy": 0.17961601614952089,
      "num_tokens": 22785027.0,
      "step": 11245
    },
    {
      "entropy": 5.935587644577026,
      "epoch": 0.9831337935855982,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004909943027660201,
      "loss": 5.5875,
      "mean_token_accuracy": 0.17173868864774705,
      "num_tokens": 22796399.0,
      "step": 11250
    },
    {
      "entropy": 5.867468023300171,
      "epoch": 0.9835707419383029,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00049098557288237,
      "loss": 5.5244,
      "mean_token_accuracy": 0.17380052357912062,
      "num_tokens": 22806352.0,
      "step": 11255
    },
    {
      "entropy": 5.917151260375976,
      "epoch": 0.9840076902910077,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004909768388560036,
      "loss": 5.5952,
      "mean_token_accuracy": 0.17268310040235518,
      "num_tokens": 22818369.0,
      "step": 11260
    },
    {
      "entropy": 5.911303520202637,
      "epoch": 0.9844446386437123,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004909681006870882,
      "loss": 5.5019,
      "mean_token_accuracy": 0.1829800382256508,
      "num_tokens": 22828063.0,
      "step": 11265
    },
    {
      "entropy": 5.867061185836792,
      "epoch": 0.984881586996417,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004909593583757915,
      "loss": 5.5819,
      "mean_token_accuracy": 0.17511370331048964,
      "num_tokens": 22839265.0,
      "step": 11270
    },
    {
      "entropy": 5.856913280487061,
      "epoch": 0.9853185353491217,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004909506119222812,
      "loss": 5.5053,
      "mean_token_accuracy": 0.1864972546696663,
      "num_tokens": 22848501.0,
      "step": 11275
    },
    {
      "entropy": 5.858864450454712,
      "epoch": 0.9857554837018264,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004909418613267248,
      "loss": 5.4702,
      "mean_token_accuracy": 0.18337075412273407,
      "num_tokens": 22859176.0,
      "step": 11280
    },
    {
      "entropy": 5.8641746044158936,
      "epoch": 0.9861924320545311,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004909331065892905,
      "loss": 5.4372,
      "mean_token_accuracy": 0.19020589143037797,
      "num_tokens": 22869312.0,
      "step": 11285
    },
    {
      "entropy": 5.91535439491272,
      "epoch": 0.9866293804072359,
      "grad_norm": 1.015625,
      "learning_rate": 0.000490924347710146,
      "loss": 5.5524,
      "mean_token_accuracy": 0.17741041332483293,
      "num_tokens": 22878699.0,
      "step": 11290
    },
    {
      "entropy": 5.894830369949341,
      "epoch": 0.9870663287599406,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004909155846894595,
      "loss": 5.5834,
      "mean_token_accuracy": 0.17876483500003815,
      "num_tokens": 22889147.0,
      "step": 11295
    },
    {
      "entropy": 5.936345863342285,
      "epoch": 0.9875032771126453,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004909068175273989,
      "loss": 5.5004,
      "mean_token_accuracy": 0.1854189857840538,
      "num_tokens": 22899085.0,
      "step": 11300
    },
    {
      "entropy": 5.916552448272705,
      "epoch": 0.98794022546535,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004908980462241323,
      "loss": 5.6199,
      "mean_token_accuracy": 0.17722321748733522,
      "num_tokens": 22909924.0,
      "step": 11305
    },
    {
      "entropy": 5.943548488616943,
      "epoch": 0.9883771738180547,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004908892707798281,
      "loss": 5.5814,
      "mean_token_accuracy": 0.18010277003049852,
      "num_tokens": 22920426.0,
      "step": 11310
    },
    {
      "entropy": 5.888430452346801,
      "epoch": 0.9888141221707594,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004908804911946546,
      "loss": 5.5063,
      "mean_token_accuracy": 0.17503743916749953,
      "num_tokens": 22929929.0,
      "step": 11315
    },
    {
      "entropy": 5.875927209854126,
      "epoch": 0.9892510705234642,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004908717074687801,
      "loss": 5.4786,
      "mean_token_accuracy": 0.18969572633504866,
      "num_tokens": 22940102.0,
      "step": 11320
    },
    {
      "entropy": 5.907906770706177,
      "epoch": 0.9896880188761689,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004908629196023732,
      "loss": 5.5256,
      "mean_token_accuracy": 0.18596358001232147,
      "num_tokens": 22949490.0,
      "step": 11325
    },
    {
      "entropy": 5.888055896759033,
      "epoch": 0.9901249672288736,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004908541275956024,
      "loss": 5.4642,
      "mean_token_accuracy": 0.18552766591310502,
      "num_tokens": 22958119.0,
      "step": 11330
    },
    {
      "entropy": 5.90260329246521,
      "epoch": 0.9905619155815782,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004908453314486363,
      "loss": 5.4977,
      "mean_token_accuracy": 0.17876953929662703,
      "num_tokens": 22969447.0,
      "step": 11335
    },
    {
      "entropy": 5.845128774642944,
      "epoch": 0.9909988639342829,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004908365311616437,
      "loss": 5.3792,
      "mean_token_accuracy": 0.19080051481723787,
      "num_tokens": 22978925.0,
      "step": 11340
    },
    {
      "entropy": 5.873134183883667,
      "epoch": 0.9914358122869876,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004908277267347931,
      "loss": 5.5304,
      "mean_token_accuracy": 0.18675396740436553,
      "num_tokens": 22988861.0,
      "step": 11345
    },
    {
      "entropy": 5.839180183410645,
      "epoch": 0.9918727606396924,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004908189181682538,
      "loss": 5.5543,
      "mean_token_accuracy": 0.17834940403699875,
      "num_tokens": 22998700.0,
      "step": 11350
    },
    {
      "entropy": 5.869525289535522,
      "epoch": 0.9923097089923971,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004908101054621946,
      "loss": 5.5028,
      "mean_token_accuracy": 0.18774064928293227,
      "num_tokens": 23009095.0,
      "step": 11355
    },
    {
      "entropy": 5.881543827056885,
      "epoch": 0.9927466573451018,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004908012886167845,
      "loss": 5.5156,
      "mean_token_accuracy": 0.18329519182443618,
      "num_tokens": 23019484.0,
      "step": 11360
    },
    {
      "entropy": 5.852908182144165,
      "epoch": 0.9931836056978065,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004907924676321924,
      "loss": 5.4749,
      "mean_token_accuracy": 0.18290173560380935,
      "num_tokens": 23029867.0,
      "step": 11365
    },
    {
      "entropy": 5.837313938140869,
      "epoch": 0.9936205540505112,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000490783642508588,
      "loss": 5.4603,
      "mean_token_accuracy": 0.18994932174682616,
      "num_tokens": 23040179.0,
      "step": 11370
    },
    {
      "entropy": 5.900873422622681,
      "epoch": 0.9940575024032159,
      "grad_norm": 1.015625,
      "learning_rate": 0.00049077481324614,
      "loss": 5.4288,
      "mean_token_accuracy": 0.1939394161105156,
      "num_tokens": 23050389.0,
      "step": 11375
    },
    {
      "entropy": 5.781686592102051,
      "epoch": 0.9944944507559207,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000490765979845018,
      "loss": 5.4731,
      "mean_token_accuracy": 0.1850649446249008,
      "num_tokens": 23059984.0,
      "step": 11380
    },
    {
      "entropy": 5.895578336715698,
      "epoch": 0.9949313991086254,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004907571423053917,
      "loss": 5.4814,
      "mean_token_accuracy": 0.18117451518774033,
      "num_tokens": 23069828.0,
      "step": 11385
    },
    {
      "entropy": 5.8813835144042965,
      "epoch": 0.9953683474613301,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004907483006274302,
      "loss": 5.5366,
      "mean_token_accuracy": 0.18166570365428925,
      "num_tokens": 23079223.0,
      "step": 11390
    },
    {
      "entropy": 5.868198728561401,
      "epoch": 0.9958052958140348,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004907394548113033,
      "loss": 5.5672,
      "mean_token_accuracy": 0.1767510399222374,
      "num_tokens": 23088540.0,
      "step": 11395
    },
    {
      "entropy": 5.918020963668823,
      "epoch": 0.9962422441667395,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004907306048571806,
      "loss": 5.6177,
      "mean_token_accuracy": 0.17534056305885315,
      "num_tokens": 23098617.0,
      "step": 11400
    },
    {
      "entropy": 5.9210670471191404,
      "epoch": 0.9966791925194443,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004907217507652317,
      "loss": 5.5006,
      "mean_token_accuracy": 0.1844383180141449,
      "num_tokens": 23109702.0,
      "step": 11405
    },
    {
      "entropy": 5.925984621047974,
      "epoch": 0.997116140872149,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004907128925356268,
      "loss": 5.522,
      "mean_token_accuracy": 0.17999950647354127,
      "num_tokens": 23120295.0,
      "step": 11410
    },
    {
      "entropy": 5.872051572799682,
      "epoch": 0.9975530892248536,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004907040301685355,
      "loss": 5.5474,
      "mean_token_accuracy": 0.1788916304707527,
      "num_tokens": 23131999.0,
      "step": 11415
    },
    {
      "entropy": 5.933912563323974,
      "epoch": 0.9979900375775583,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004906951636641279,
      "loss": 5.5594,
      "mean_token_accuracy": 0.1831767365336418,
      "num_tokens": 23141060.0,
      "step": 11420
    },
    {
      "entropy": 5.826948165893555,
      "epoch": 0.998426985930263,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000490686293022574,
      "loss": 5.5012,
      "mean_token_accuracy": 0.18257489800453186,
      "num_tokens": 23151607.0,
      "step": 11425
    },
    {
      "entropy": 5.877261924743652,
      "epoch": 0.9988639342829677,
      "grad_norm": 1.015625,
      "learning_rate": 0.000490677418244044,
      "loss": 5.573,
      "mean_token_accuracy": 0.17539529949426652,
      "num_tokens": 23161892.0,
      "step": 11430
    },
    {
      "entropy": 5.90282793045044,
      "epoch": 0.9993008826356725,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004906685393287081,
      "loss": 5.5537,
      "mean_token_accuracy": 0.18018418103456496,
      "num_tokens": 23172112.0,
      "step": 11435
    },
    {
      "entropy": 5.905782556533813,
      "epoch": 0.9997378309883772,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004906596562767367,
      "loss": 5.4473,
      "mean_token_accuracy": 0.1831355482339859,
      "num_tokens": 23181435.0,
      "step": 11440
    },
    {
      "entropy": 5.837978649139404,
      "epoch": 1.0001747793410818,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004906507690883,
      "loss": 5.4131,
      "mean_token_accuracy": 0.1849936366081238,
      "num_tokens": 23191084.0,
      "step": 11445
    },
    {
      "entropy": 5.8687920570373535,
      "epoch": 1.0006117276937867,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004906418777635685,
      "loss": 5.438,
      "mean_token_accuracy": 0.18681821525096892,
      "num_tokens": 23202372.0,
      "step": 11450
    },
    {
      "entropy": 5.913035202026367,
      "epoch": 1.0010486760464914,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004906329823027129,
      "loss": 5.4966,
      "mean_token_accuracy": 0.17962681651115417,
      "num_tokens": 23214113.0,
      "step": 11455
    },
    {
      "entropy": 5.842164468765259,
      "epoch": 1.001485624399196,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004906240827059036,
      "loss": 5.3958,
      "mean_token_accuracy": 0.19234659820795058,
      "num_tokens": 23223562.0,
      "step": 11460
    },
    {
      "entropy": 5.897656106948853,
      "epoch": 1.0019225727519008,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004906151789733114,
      "loss": 5.4745,
      "mean_token_accuracy": 0.18432479351758957,
      "num_tokens": 23232901.0,
      "step": 11465
    },
    {
      "entropy": 5.927430534362793,
      "epoch": 1.0023595211046055,
      "grad_norm": 1.0,
      "learning_rate": 0.0004906062711051073,
      "loss": 5.4919,
      "mean_token_accuracy": 0.17908660918474198,
      "num_tokens": 23242869.0,
      "step": 11470
    },
    {
      "entropy": 5.868844079971313,
      "epoch": 1.0027964694573102,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004905973591014617,
      "loss": 5.3969,
      "mean_token_accuracy": 0.19357268512248993,
      "num_tokens": 23253653.0,
      "step": 11475
    },
    {
      "entropy": 5.839746522903442,
      "epoch": 1.0032334178100148,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004905884429625459,
      "loss": 5.4198,
      "mean_token_accuracy": 0.1854293704032898,
      "num_tokens": 23263530.0,
      "step": 11480
    },
    {
      "entropy": 5.880763196945191,
      "epoch": 1.0036703661627195,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004905795226885308,
      "loss": 5.4584,
      "mean_token_accuracy": 0.18436275124549867,
      "num_tokens": 23274946.0,
      "step": 11485
    },
    {
      "entropy": 5.948960685729981,
      "epoch": 1.0041073145154242,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004905705982795874,
      "loss": 5.4532,
      "mean_token_accuracy": 0.18723117113113402,
      "num_tokens": 23283811.0,
      "step": 11490
    },
    {
      "entropy": 5.823022031784058,
      "epoch": 1.004544262868129,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000490561669735887,
      "loss": 5.4889,
      "mean_token_accuracy": 0.182684426009655,
      "num_tokens": 23294554.0,
      "step": 11495
    },
    {
      "entropy": 5.8348030090332035,
      "epoch": 1.0049812112208336,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004905527370576009,
      "loss": 5.3471,
      "mean_token_accuracy": 0.19114124178886413,
      "num_tokens": 23304245.0,
      "step": 11500
    },
    {
      "entropy": 5.965893220901489,
      "epoch": 1.0054181595735383,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004905438002449004,
      "loss": 5.4931,
      "mean_token_accuracy": 0.1761218547821045,
      "num_tokens": 23315169.0,
      "step": 11505
    },
    {
      "entropy": 5.8698876857757565,
      "epoch": 1.0058551079262432,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004905348592979567,
      "loss": 5.3897,
      "mean_token_accuracy": 0.1896383836865425,
      "num_tokens": 23325528.0,
      "step": 11510
    },
    {
      "entropy": 5.920818710327149,
      "epoch": 1.006292056278948,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004905259142169415,
      "loss": 5.5236,
      "mean_token_accuracy": 0.18068531900644302,
      "num_tokens": 23336744.0,
      "step": 11515
    },
    {
      "entropy": 5.8515540599823,
      "epoch": 1.0067290046316526,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004905169650020264,
      "loss": 5.3796,
      "mean_token_accuracy": 0.19421585351228715,
      "num_tokens": 23346092.0,
      "step": 11520
    },
    {
      "entropy": 5.808361577987671,
      "epoch": 1.0071659529843573,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004905080116533831,
      "loss": 5.4104,
      "mean_token_accuracy": 0.18218514174222947,
      "num_tokens": 23357326.0,
      "step": 11525
    },
    {
      "entropy": 5.887636470794678,
      "epoch": 1.007602901337062,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000490499054171183,
      "loss": 5.4258,
      "mean_token_accuracy": 0.1892286106944084,
      "num_tokens": 23367789.0,
      "step": 11530
    },
    {
      "entropy": 5.888553094863892,
      "epoch": 1.0080398496897667,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004904900925555982,
      "loss": 5.4281,
      "mean_token_accuracy": 0.18160260766744613,
      "num_tokens": 23377499.0,
      "step": 11535
    },
    {
      "entropy": 5.896770763397217,
      "epoch": 1.0084767980424714,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004904811268068005,
      "loss": 5.4018,
      "mean_token_accuracy": 0.185566645860672,
      "num_tokens": 23387867.0,
      "step": 11540
    },
    {
      "entropy": 5.819874048233032,
      "epoch": 1.008913746395176,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004904721569249618,
      "loss": 5.5174,
      "mean_token_accuracy": 0.1807884693145752,
      "num_tokens": 23398179.0,
      "step": 11545
    },
    {
      "entropy": 5.921413278579712,
      "epoch": 1.0093506947478808,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004904631829102543,
      "loss": 5.5064,
      "mean_token_accuracy": 0.18265709578990935,
      "num_tokens": 23407407.0,
      "step": 11550
    },
    {
      "entropy": 5.903771781921387,
      "epoch": 1.0097876431005854,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004904542047628501,
      "loss": 5.3357,
      "mean_token_accuracy": 0.18950313478708267,
      "num_tokens": 23417324.0,
      "step": 11555
    },
    {
      "entropy": 5.867985820770263,
      "epoch": 1.0102245914532901,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004904452224829212,
      "loss": 5.4009,
      "mean_token_accuracy": 0.1907619282603264,
      "num_tokens": 23427157.0,
      "step": 11560
    },
    {
      "entropy": 5.875456619262695,
      "epoch": 1.0106615398059948,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004904362360706403,
      "loss": 5.5044,
      "mean_token_accuracy": 0.18253983706235885,
      "num_tokens": 23437712.0,
      "step": 11565
    },
    {
      "entropy": 5.880200719833374,
      "epoch": 1.0110984881586997,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004904272455261793,
      "loss": 5.4176,
      "mean_token_accuracy": 0.19042740613222123,
      "num_tokens": 23447949.0,
      "step": 11570
    },
    {
      "entropy": 5.8665567398071286,
      "epoch": 1.0115354365114044,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004904182508497109,
      "loss": 5.3829,
      "mean_token_accuracy": 0.18791229873895646,
      "num_tokens": 23457319.0,
      "step": 11575
    },
    {
      "entropy": 5.897786235809326,
      "epoch": 1.0119723848641091,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004904092520414075,
      "loss": 5.4903,
      "mean_token_accuracy": 0.1840863510966301,
      "num_tokens": 23468245.0,
      "step": 11580
    },
    {
      "entropy": 5.9232865333557125,
      "epoch": 1.0124093332168138,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004904002491014418,
      "loss": 5.457,
      "mean_token_accuracy": 0.18892648071050644,
      "num_tokens": 23477049.0,
      "step": 11585
    },
    {
      "entropy": 5.873874282836914,
      "epoch": 1.0128462815695185,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004903912420299866,
      "loss": 5.5319,
      "mean_token_accuracy": 0.18053671419620515,
      "num_tokens": 23488574.0,
      "step": 11590
    },
    {
      "entropy": 5.909673404693604,
      "epoch": 1.0132832299222232,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004903822308272144,
      "loss": 5.451,
      "mean_token_accuracy": 0.18629641234874725,
      "num_tokens": 23499282.0,
      "step": 11595
    },
    {
      "entropy": 5.887287235260009,
      "epoch": 1.0137201782749279,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004903732154932982,
      "loss": 5.4763,
      "mean_token_accuracy": 0.19171027094125748,
      "num_tokens": 23508777.0,
      "step": 11600
    },
    {
      "entropy": 5.827649164199829,
      "epoch": 1.0141571266276326,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004903641960284108,
      "loss": 5.4346,
      "mean_token_accuracy": 0.1886472746729851,
      "num_tokens": 23518818.0,
      "step": 11605
    },
    {
      "entropy": 5.868232583999633,
      "epoch": 1.0145940749803373,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004903551724327254,
      "loss": 5.4428,
      "mean_token_accuracy": 0.18531371802091598,
      "num_tokens": 23529668.0,
      "step": 11610
    },
    {
      "entropy": 5.943478298187256,
      "epoch": 1.015031023333042,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004903461447064147,
      "loss": 5.4237,
      "mean_token_accuracy": 0.1891187071800232,
      "num_tokens": 23540418.0,
      "step": 11615
    },
    {
      "entropy": 5.887177276611328,
      "epoch": 1.0154679716857467,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004903371128496523,
      "loss": 5.4075,
      "mean_token_accuracy": 0.19094202071428298,
      "num_tokens": 23550035.0,
      "step": 11620
    },
    {
      "entropy": 5.840616846084595,
      "epoch": 1.0159049200384516,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004903280768626113,
      "loss": 5.4668,
      "mean_token_accuracy": 0.18491043150424957,
      "num_tokens": 23559487.0,
      "step": 11625
    },
    {
      "entropy": 5.8646666526794435,
      "epoch": 1.0163418683911563,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004903190367454648,
      "loss": 5.3794,
      "mean_token_accuracy": 0.1918038859963417,
      "num_tokens": 23569146.0,
      "step": 11630
    },
    {
      "entropy": 5.840773105621338,
      "epoch": 1.016778816743861,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004903099924983864,
      "loss": 5.3502,
      "mean_token_accuracy": 0.18685699105262757,
      "num_tokens": 23579458.0,
      "step": 11635
    },
    {
      "entropy": 5.855215358734131,
      "epoch": 1.0172157650965656,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004903009441215495,
      "loss": 5.3908,
      "mean_token_accuracy": 0.18620171695947646,
      "num_tokens": 23589416.0,
      "step": 11640
    },
    {
      "entropy": 5.909987497329712,
      "epoch": 1.0176527134492703,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004902918916151278,
      "loss": 5.4793,
      "mean_token_accuracy": 0.18781950920820237,
      "num_tokens": 23599588.0,
      "step": 11645
    },
    {
      "entropy": 5.922841787338257,
      "epoch": 1.018089661801975,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004902828349792947,
      "loss": 5.471,
      "mean_token_accuracy": 0.18326490372419357,
      "num_tokens": 23610116.0,
      "step": 11650
    },
    {
      "entropy": 5.842015552520752,
      "epoch": 1.0185266101546797,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000490273774214224,
      "loss": 5.3788,
      "mean_token_accuracy": 0.19686615467071533,
      "num_tokens": 23619906.0,
      "step": 11655
    },
    {
      "entropy": 5.895055198669434,
      "epoch": 1.0189635585073844,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004902647093200896,
      "loss": 5.4264,
      "mean_token_accuracy": 0.18627251088619232,
      "num_tokens": 23630351.0,
      "step": 11660
    },
    {
      "entropy": 5.882483100891113,
      "epoch": 1.019400506860089,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004902556402970652,
      "loss": 5.4804,
      "mean_token_accuracy": 0.18572100698947908,
      "num_tokens": 23639696.0,
      "step": 11665
    },
    {
      "entropy": 5.878877973556518,
      "epoch": 1.0198374552127938,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004902465671453248,
      "loss": 5.4645,
      "mean_token_accuracy": 0.18306163549423218,
      "num_tokens": 23649388.0,
      "step": 11670
    },
    {
      "entropy": 5.929340410232544,
      "epoch": 1.0202744035654985,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004902374898650425,
      "loss": 5.4873,
      "mean_token_accuracy": 0.18707587271928788,
      "num_tokens": 23659125.0,
      "step": 11675
    },
    {
      "entropy": 5.974800634384155,
      "epoch": 1.0207113519182032,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004902284084563924,
      "loss": 5.4795,
      "mean_token_accuracy": 0.18587813526391983,
      "num_tokens": 23668817.0,
      "step": 11680
    },
    {
      "entropy": 5.890401315689087,
      "epoch": 1.021148300270908,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004902193229195487,
      "loss": 5.4764,
      "mean_token_accuracy": 0.183877132833004,
      "num_tokens": 23679378.0,
      "step": 11685
    },
    {
      "entropy": 5.87258472442627,
      "epoch": 1.0215852486236128,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004902102332546855,
      "loss": 5.4278,
      "mean_token_accuracy": 0.18260236084461212,
      "num_tokens": 23689400.0,
      "step": 11690
    },
    {
      "entropy": 5.953070402145386,
      "epoch": 1.0220221969763175,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004902011394619772,
      "loss": 5.556,
      "mean_token_accuracy": 0.1748400166630745,
      "num_tokens": 23699503.0,
      "step": 11695
    },
    {
      "entropy": 5.9045271396636965,
      "epoch": 1.0224591453290222,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004901920415415985,
      "loss": 5.4782,
      "mean_token_accuracy": 0.17992380410432815,
      "num_tokens": 23710913.0,
      "step": 11700
    },
    {
      "entropy": 5.856723785400391,
      "epoch": 1.0228960936817268,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004901829394937234,
      "loss": 5.3896,
      "mean_token_accuracy": 0.19051228761672973,
      "num_tokens": 23720730.0,
      "step": 11705
    },
    {
      "entropy": 5.890582132339477,
      "epoch": 1.0233330420344315,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000490173833318527,
      "loss": 5.4717,
      "mean_token_accuracy": 0.18334903568029404,
      "num_tokens": 23730902.0,
      "step": 11710
    },
    {
      "entropy": 5.874481868743897,
      "epoch": 1.0237699903871362,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004901647230161837,
      "loss": 5.5038,
      "mean_token_accuracy": 0.18231870532035827,
      "num_tokens": 23741901.0,
      "step": 11715
    },
    {
      "entropy": 5.92249870300293,
      "epoch": 1.024206938739841,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004901556085868681,
      "loss": 5.3806,
      "mean_token_accuracy": 0.18878458589315414,
      "num_tokens": 23752029.0,
      "step": 11720
    },
    {
      "entropy": 5.8456573486328125,
      "epoch": 1.0246438870925456,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004901464900307553,
      "loss": 5.3883,
      "mean_token_accuracy": 0.190079066157341,
      "num_tokens": 23762125.0,
      "step": 11725
    },
    {
      "entropy": 5.840774059295654,
      "epoch": 1.0250808354452503,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004901373673480201,
      "loss": 5.4457,
      "mean_token_accuracy": 0.18981413990259172,
      "num_tokens": 23772056.0,
      "step": 11730
    },
    {
      "entropy": 5.8627174377441404,
      "epoch": 1.025517783797955,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004901282405388375,
      "loss": 5.3691,
      "mean_token_accuracy": 0.19195899367332458,
      "num_tokens": 23781441.0,
      "step": 11735
    },
    {
      "entropy": 5.87320466041565,
      "epoch": 1.0259547321506597,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004901191096033825,
      "loss": 5.386,
      "mean_token_accuracy": 0.1865521028637886,
      "num_tokens": 23791472.0,
      "step": 11740
    },
    {
      "entropy": 5.895962524414062,
      "epoch": 1.0263916805033646,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004901099745418302,
      "loss": 5.4084,
      "mean_token_accuracy": 0.18438571840524673,
      "num_tokens": 23801940.0,
      "step": 11745
    },
    {
      "entropy": 5.837318181991577,
      "epoch": 1.0268286288560693,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004901008353543559,
      "loss": 5.3329,
      "mean_token_accuracy": 0.19344220757484437,
      "num_tokens": 23811515.0,
      "step": 11750
    },
    {
      "entropy": 5.829492855072021,
      "epoch": 1.027265577208774,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004900916920411349,
      "loss": 5.4373,
      "mean_token_accuracy": 0.17939050793647765,
      "num_tokens": 23821871.0,
      "step": 11755
    },
    {
      "entropy": 5.851590585708618,
      "epoch": 1.0277025255614787,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004900825446023426,
      "loss": 5.3788,
      "mean_token_accuracy": 0.19067484587430955,
      "num_tokens": 23831677.0,
      "step": 11760
    },
    {
      "entropy": 5.844141578674316,
      "epoch": 1.0281394739141834,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004900733930381543,
      "loss": 5.4255,
      "mean_token_accuracy": 0.18944600969552994,
      "num_tokens": 23841415.0,
      "step": 11765
    },
    {
      "entropy": 5.908268308639526,
      "epoch": 1.028576422266888,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004900642373487455,
      "loss": 5.4837,
      "mean_token_accuracy": 0.18451157510280608,
      "num_tokens": 23851809.0,
      "step": 11770
    },
    {
      "entropy": 5.926433372497558,
      "epoch": 1.0290133706195927,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004900550775342921,
      "loss": 5.4262,
      "mean_token_accuracy": 0.1822318211197853,
      "num_tokens": 23861981.0,
      "step": 11775
    },
    {
      "entropy": 5.847351169586181,
      "epoch": 1.0294503189722974,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004900459135949697,
      "loss": 5.4158,
      "mean_token_accuracy": 0.18425562381744384,
      "num_tokens": 23872437.0,
      "step": 11780
    },
    {
      "entropy": 5.863445281982422,
      "epoch": 1.0298872673250021,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004900367455309537,
      "loss": 5.4388,
      "mean_token_accuracy": 0.18581672310829161,
      "num_tokens": 23881650.0,
      "step": 11785
    },
    {
      "entropy": 5.846681833267212,
      "epoch": 1.0303242156777068,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004900275733424205,
      "loss": 5.4386,
      "mean_token_accuracy": 0.18531339466571808,
      "num_tokens": 23891631.0,
      "step": 11790
    },
    {
      "entropy": 5.834491729736328,
      "epoch": 1.0307611640304115,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004900183970295457,
      "loss": 5.4071,
      "mean_token_accuracy": 0.18963743448257447,
      "num_tokens": 23901725.0,
      "step": 11795
    },
    {
      "entropy": 5.897421741485596,
      "epoch": 1.0311981123831164,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004900092165925053,
      "loss": 5.4296,
      "mean_token_accuracy": 0.1841016009449959,
      "num_tokens": 23911850.0,
      "step": 11800
    },
    {
      "entropy": 5.933511972427368,
      "epoch": 1.0316350607358211,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004900000320314757,
      "loss": 5.5181,
      "mean_token_accuracy": 0.18105760216712952,
      "num_tokens": 23922571.0,
      "step": 11805
    },
    {
      "entropy": 5.836334133148194,
      "epoch": 1.0320720090885258,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004899908433466326,
      "loss": 5.4272,
      "mean_token_accuracy": 0.1847069188952446,
      "num_tokens": 23932637.0,
      "step": 11810
    },
    {
      "entropy": 5.925191402435303,
      "epoch": 1.0325089574412305,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004899816505381525,
      "loss": 5.5172,
      "mean_token_accuracy": 0.17859063297510147,
      "num_tokens": 23943288.0,
      "step": 11815
    },
    {
      "entropy": 5.903874158859253,
      "epoch": 1.0329459057939352,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004899724536062117,
      "loss": 5.4049,
      "mean_token_accuracy": 0.18413027077913285,
      "num_tokens": 23952853.0,
      "step": 11820
    },
    {
      "entropy": 5.832293748855591,
      "epoch": 1.0333828541466399,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004899632525509867,
      "loss": 5.4618,
      "mean_token_accuracy": 0.18013489991426468,
      "num_tokens": 23963111.0,
      "step": 11825
    },
    {
      "entropy": 5.836339902877808,
      "epoch": 1.0338198024993446,
      "grad_norm": 1.0,
      "learning_rate": 0.0004899540473726538,
      "loss": 5.3671,
      "mean_token_accuracy": 0.19527381807565689,
      "num_tokens": 23972468.0,
      "step": 11830
    },
    {
      "entropy": 5.921899700164795,
      "epoch": 1.0342567508520493,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004899448380713897,
      "loss": 5.4109,
      "mean_token_accuracy": 0.19330753535032272,
      "num_tokens": 23982567.0,
      "step": 11835
    },
    {
      "entropy": 5.850337791442871,
      "epoch": 1.034693699204754,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004899356246473709,
      "loss": 5.4094,
      "mean_token_accuracy": 0.19046176075935364,
      "num_tokens": 23993283.0,
      "step": 11840
    },
    {
      "entropy": 5.820649909973144,
      "epoch": 1.0351306475574586,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004899264071007741,
      "loss": 5.3927,
      "mean_token_accuracy": 0.19260794073343276,
      "num_tokens": 24004504.0,
      "step": 11845
    },
    {
      "entropy": 5.869478702545166,
      "epoch": 1.0355675959101633,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004899171854317764,
      "loss": 5.4296,
      "mean_token_accuracy": 0.18923589289188386,
      "num_tokens": 24014557.0,
      "step": 11850
    },
    {
      "entropy": 5.847081279754638,
      "epoch": 1.036004544262868,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004899079596405543,
      "loss": 5.4295,
      "mean_token_accuracy": 0.18470271229743956,
      "num_tokens": 24024495.0,
      "step": 11855
    },
    {
      "entropy": 5.803589487075806,
      "epoch": 1.036441492615573,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000489898729727285,
      "loss": 5.4302,
      "mean_token_accuracy": 0.18560323715209961,
      "num_tokens": 24034984.0,
      "step": 11860
    },
    {
      "entropy": 5.872931289672851,
      "epoch": 1.0368784409682776,
      "grad_norm": 1.0,
      "learning_rate": 0.0004898894956921455,
      "loss": 5.3798,
      "mean_token_accuracy": 0.19224686026573182,
      "num_tokens": 24044379.0,
      "step": 11865
    },
    {
      "entropy": 5.9014054298400875,
      "epoch": 1.0373153893209823,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004898802575353129,
      "loss": 5.4453,
      "mean_token_accuracy": 0.18051902055740357,
      "num_tokens": 24054133.0,
      "step": 11870
    },
    {
      "entropy": 5.790961122512817,
      "epoch": 1.037752337673687,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004898710152569642,
      "loss": 5.3722,
      "mean_token_accuracy": 0.18675147294998168,
      "num_tokens": 24064868.0,
      "step": 11875
    },
    {
      "entropy": 5.881299829483032,
      "epoch": 1.0381892860263917,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004898617688572769,
      "loss": 5.4479,
      "mean_token_accuracy": 0.18316660076379776,
      "num_tokens": 24074977.0,
      "step": 11880
    },
    {
      "entropy": 5.878363466262817,
      "epoch": 1.0386262343790964,
      "grad_norm": 0.875,
      "learning_rate": 0.0004898525183364284,
      "loss": 5.5004,
      "mean_token_accuracy": 0.18267498314380645,
      "num_tokens": 24085673.0,
      "step": 11885
    },
    {
      "entropy": 5.882804489135742,
      "epoch": 1.039063182731801,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004898432636945959,
      "loss": 5.4631,
      "mean_token_accuracy": 0.18758496046066284,
      "num_tokens": 24095465.0,
      "step": 11890
    },
    {
      "entropy": 5.856014919281006,
      "epoch": 1.0395001310845058,
      "grad_norm": 0.984375,
      "learning_rate": 0.000489834004931957,
      "loss": 5.4072,
      "mean_token_accuracy": 0.18799237459897994,
      "num_tokens": 24105576.0,
      "step": 11895
    },
    {
      "entropy": 5.896707344055176,
      "epoch": 1.0399370794372105,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004898247420486895,
      "loss": 5.4029,
      "mean_token_accuracy": 0.18475393503904342,
      "num_tokens": 24114998.0,
      "step": 11900
    },
    {
      "entropy": 5.862454175949097,
      "epoch": 1.0403740277899152,
      "grad_norm": 1.0,
      "learning_rate": 0.0004898154750449707,
      "loss": 5.3996,
      "mean_token_accuracy": 0.19288893640041352,
      "num_tokens": 24125059.0,
      "step": 11905
    },
    {
      "entropy": 5.872966480255127,
      "epoch": 1.0408109761426199,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004898062039209786,
      "loss": 5.4218,
      "mean_token_accuracy": 0.18799618035554885,
      "num_tokens": 24134062.0,
      "step": 11910
    },
    {
      "entropy": 5.896388101577759,
      "epoch": 1.0412479244953246,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000489796928676891,
      "loss": 5.4495,
      "mean_token_accuracy": 0.1844935804605484,
      "num_tokens": 24144016.0,
      "step": 11915
    },
    {
      "entropy": 5.895144939422607,
      "epoch": 1.0416848728480295,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004897876493128856,
      "loss": 5.4177,
      "mean_token_accuracy": 0.18701936453580856,
      "num_tokens": 24154337.0,
      "step": 11920
    },
    {
      "entropy": 5.835720682144165,
      "epoch": 1.0421218212007342,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004897783658291406,
      "loss": 5.3914,
      "mean_token_accuracy": 0.19100670665502548,
      "num_tokens": 24164440.0,
      "step": 11925
    },
    {
      "entropy": 5.8663746356964115,
      "epoch": 1.0425587695534388,
      "grad_norm": 1.109375,
      "learning_rate": 0.000489769078225834,
      "loss": 5.5699,
      "mean_token_accuracy": 0.17394503504037856,
      "num_tokens": 24174388.0,
      "step": 11930
    },
    {
      "entropy": 5.848772859573364,
      "epoch": 1.0429957179061435,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004897597865031441,
      "loss": 5.3751,
      "mean_token_accuracy": 0.19149001091718673,
      "num_tokens": 24183543.0,
      "step": 11935
    },
    {
      "entropy": 5.907330656051636,
      "epoch": 1.0434326662588482,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004897504906612487,
      "loss": 5.4231,
      "mean_token_accuracy": 0.192223858833313,
      "num_tokens": 24193847.0,
      "step": 11940
    },
    {
      "entropy": 5.866742944717407,
      "epoch": 1.043869614611553,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004897411907003265,
      "loss": 5.4144,
      "mean_token_accuracy": 0.1850503206253052,
      "num_tokens": 24203072.0,
      "step": 11945
    },
    {
      "entropy": 5.798723030090332,
      "epoch": 1.0443065629642576,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004897318866205558,
      "loss": 5.35,
      "mean_token_accuracy": 0.1925133764743805,
      "num_tokens": 24213610.0,
      "step": 11950
    },
    {
      "entropy": 5.778581619262695,
      "epoch": 1.0447435113169623,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004897225784221149,
      "loss": 5.4126,
      "mean_token_accuracy": 0.18889611065387726,
      "num_tokens": 24223571.0,
      "step": 11955
    },
    {
      "entropy": 5.868343019485474,
      "epoch": 1.045180459669667,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004897132661051826,
      "loss": 5.4263,
      "mean_token_accuracy": 0.18774027824401857,
      "num_tokens": 24233867.0,
      "step": 11960
    },
    {
      "entropy": 5.875905227661133,
      "epoch": 1.0456174080223717,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004897039496699372,
      "loss": 5.4444,
      "mean_token_accuracy": 0.1827916532754898,
      "num_tokens": 24243853.0,
      "step": 11965
    },
    {
      "entropy": 5.831917142868042,
      "epoch": 1.0460543563750764,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004896946291165575,
      "loss": 5.4051,
      "mean_token_accuracy": 0.18603692054748536,
      "num_tokens": 24254049.0,
      "step": 11970
    },
    {
      "entropy": 5.821115016937256,
      "epoch": 1.0464913047277813,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004896853044452224,
      "loss": 5.3837,
      "mean_token_accuracy": 0.19248728454113007,
      "num_tokens": 24265166.0,
      "step": 11975
    },
    {
      "entropy": 5.890525007247925,
      "epoch": 1.046928253080486,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004896759756561106,
      "loss": 5.4241,
      "mean_token_accuracy": 0.18939729183912277,
      "num_tokens": 24274981.0,
      "step": 11980
    },
    {
      "entropy": 5.8979333400726315,
      "epoch": 1.0473652014331907,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004896666427494012,
      "loss": 5.4555,
      "mean_token_accuracy": 0.18553800880908966,
      "num_tokens": 24285009.0,
      "step": 11985
    },
    {
      "entropy": 5.958132648468018,
      "epoch": 1.0478021497858954,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000489657305725273,
      "loss": 5.4966,
      "mean_token_accuracy": 0.1790475904941559,
      "num_tokens": 24294669.0,
      "step": 11990
    },
    {
      "entropy": 5.9100278377532955,
      "epoch": 1.0482390981386,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004896479645839052,
      "loss": 5.4853,
      "mean_token_accuracy": 0.17942507117986678,
      "num_tokens": 24304647.0,
      "step": 11995
    },
    {
      "entropy": 5.801733922958374,
      "epoch": 1.0486760464913047,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004896386193254771,
      "loss": 5.3658,
      "mean_token_accuracy": 0.19148750305175782,
      "num_tokens": 24314266.0,
      "step": 12000
    },
    {
      "epoch": 1.0486760464913047,
      "eval_entropy": 5.5726457786580665,
      "eval_loss": 5.501716613769531,
      "eval_mean_token_accuracy": 0.19160168691712784,
      "eval_num_tokens": 24314266.0,
      "eval_runtime": 21.558,
      "eval_samples_per_second": 1716.95,
      "eval_steps_per_second": 214.63,
      "step": 12000
    }
  ],
  "logging_steps": 5,
  "max_steps": 114420,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 3000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5405040117043200.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}