{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.9325351743423926,
  "eval_steps": 3000,
  "global_step": 45000,
  "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
    },
    {
      "entropy": 5.803380060195923,
      "epoch": 1.0491129948440094,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004896292699501677,
      "loss": 5.406,
      "mean_token_accuracy": 0.1927269294857979,
      "num_tokens": 24323770.0,
      "step": 12005
    },
    {
      "entropy": 5.817874193191528,
      "epoch": 1.0495499431967141,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004896199164581564,
      "loss": 5.3474,
      "mean_token_accuracy": 0.19548232853412628,
      "num_tokens": 24333438.0,
      "step": 12010
    },
    {
      "entropy": 5.866237735748291,
      "epoch": 1.0499868915494188,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004896105588496227,
      "loss": 5.4508,
      "mean_token_accuracy": 0.1850779324769974,
      "num_tokens": 24342929.0,
      "step": 12015
    },
    {
      "entropy": 5.877922773361206,
      "epoch": 1.0504238399021235,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004896011971247461,
      "loss": 5.4592,
      "mean_token_accuracy": 0.17984205931425096,
      "num_tokens": 24352617.0,
      "step": 12020
    },
    {
      "entropy": 5.889573240280152,
      "epoch": 1.0508607882548282,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004895918312837061,
      "loss": 5.3982,
      "mean_token_accuracy": 0.19045104533433915,
      "num_tokens": 24363940.0,
      "step": 12025
    },
    {
      "entropy": 5.891244316101075,
      "epoch": 1.051297736607533,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004895824613266822,
      "loss": 5.4727,
      "mean_token_accuracy": 0.18701689541339875,
      "num_tokens": 24373168.0,
      "step": 12030
    },
    {
      "entropy": 5.883642625808716,
      "epoch": 1.0517346849602376,
      "grad_norm": 1.0,
      "learning_rate": 0.0004895730872538543,
      "loss": 5.4837,
      "mean_token_accuracy": 0.1832486569881439,
      "num_tokens": 24383067.0,
      "step": 12035
    },
    {
      "entropy": 5.889386367797852,
      "epoch": 1.0521716333129425,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004895637090654021,
      "loss": 5.396,
      "mean_token_accuracy": 0.19270980656147002,
      "num_tokens": 24392734.0,
      "step": 12040
    },
    {
      "entropy": 5.847324323654175,
      "epoch": 1.0526085816656472,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004895543267615057,
      "loss": 5.4128,
      "mean_token_accuracy": 0.19035626500844954,
      "num_tokens": 24403970.0,
      "step": 12045
    },
    {
      "entropy": 5.848000717163086,
      "epoch": 1.0530455300183519,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004895449403423447,
      "loss": 5.4355,
      "mean_token_accuracy": 0.18727274537086486,
      "num_tokens": 24413614.0,
      "step": 12050
    },
    {
      "entropy": 5.825310134887696,
      "epoch": 1.0534824783710566,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004895355498080994,
      "loss": 5.3985,
      "mean_token_accuracy": 0.1915897622704506,
      "num_tokens": 24424237.0,
      "step": 12055
    },
    {
      "entropy": 5.849417686462402,
      "epoch": 1.0539194267237613,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004895261551589498,
      "loss": 5.3987,
      "mean_token_accuracy": 0.1924677312374115,
      "num_tokens": 24433220.0,
      "step": 12060
    },
    {
      "entropy": 5.9488897800445555,
      "epoch": 1.054356375076466,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004895167563950763,
      "loss": 5.534,
      "mean_token_accuracy": 0.18047311007976533,
      "num_tokens": 24444224.0,
      "step": 12065
    },
    {
      "entropy": 5.879788541793824,
      "epoch": 1.0547933234291706,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004895073535166588,
      "loss": 5.5318,
      "mean_token_accuracy": 0.17974137961864473,
      "num_tokens": 24454431.0,
      "step": 12070
    },
    {
      "entropy": 5.924104118347168,
      "epoch": 1.0552302717818753,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004894979465238779,
      "loss": 5.4191,
      "mean_token_accuracy": 0.1841446951031685,
      "num_tokens": 24464006.0,
      "step": 12075
    },
    {
      "entropy": 5.880683565139771,
      "epoch": 1.05566722013458,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004894885354169141,
      "loss": 5.3571,
      "mean_token_accuracy": 0.19350225776433944,
      "num_tokens": 24472706.0,
      "step": 12080
    },
    {
      "entropy": 5.84592833518982,
      "epoch": 1.0561041684872847,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004894791201959477,
      "loss": 5.3773,
      "mean_token_accuracy": 0.19352057576179504,
      "num_tokens": 24482785.0,
      "step": 12085
    },
    {
      "entropy": 5.858003807067871,
      "epoch": 1.0565411168399894,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004894697008611594,
      "loss": 5.4705,
      "mean_token_accuracy": 0.1848452180624008,
      "num_tokens": 24493318.0,
      "step": 12090
    },
    {
      "entropy": 5.819749641418457,
      "epoch": 1.0569780651926943,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004894602774127299,
      "loss": 5.467,
      "mean_token_accuracy": 0.1860044300556183,
      "num_tokens": 24503780.0,
      "step": 12095
    },
    {
      "entropy": 5.939620447158814,
      "epoch": 1.057415013545399,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004894508498508399,
      "loss": 5.4801,
      "mean_token_accuracy": 0.17864559441804886,
      "num_tokens": 24513905.0,
      "step": 12100
    },
    {
      "entropy": 5.908574485778809,
      "epoch": 1.0578519618981037,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004894414181756702,
      "loss": 5.4352,
      "mean_token_accuracy": 0.18800886869430541,
      "num_tokens": 24524194.0,
      "step": 12105
    },
    {
      "entropy": 5.900722789764404,
      "epoch": 1.0582889102508084,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004894319823874017,
      "loss": 5.4852,
      "mean_token_accuracy": 0.18400830924510955,
      "num_tokens": 24533456.0,
      "step": 12110
    },
    {
      "entropy": 5.901841068267823,
      "epoch": 1.058725858603513,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004894225424862154,
      "loss": 5.4654,
      "mean_token_accuracy": 0.18447764217853546,
      "num_tokens": 24544191.0,
      "step": 12115
    },
    {
      "entropy": 5.825806522369385,
      "epoch": 1.0591628069562178,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004894130984722926,
      "loss": 5.3756,
      "mean_token_accuracy": 0.19813523441553116,
      "num_tokens": 24554353.0,
      "step": 12120
    },
    {
      "entropy": 5.875379133224487,
      "epoch": 1.0595997553089225,
      "grad_norm": 1.0,
      "learning_rate": 0.000489403650345814,
      "loss": 5.4849,
      "mean_token_accuracy": 0.1834346666932106,
      "num_tokens": 24564313.0,
      "step": 12125
    },
    {
      "entropy": 5.903841924667359,
      "epoch": 1.0600367036616272,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004893941981069612,
      "loss": 5.4945,
      "mean_token_accuracy": 0.18299612700939177,
      "num_tokens": 24573129.0,
      "step": 12130
    },
    {
      "entropy": 5.8414263248443605,
      "epoch": 1.0604736520143319,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004893847417559152,
      "loss": 5.3556,
      "mean_token_accuracy": 0.1929721489548683,
      "num_tokens": 24582324.0,
      "step": 12135
    },
    {
      "entropy": 5.879167795181274,
      "epoch": 1.0609106003670365,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004893752812928576,
      "loss": 5.5091,
      "mean_token_accuracy": 0.18156105726957322,
      "num_tokens": 24593360.0,
      "step": 12140
    },
    {
      "entropy": 5.865224933624267,
      "epoch": 1.0613475487197412,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004893658167179697,
      "loss": 5.415,
      "mean_token_accuracy": 0.18945422619581223,
      "num_tokens": 24603597.0,
      "step": 12145
    },
    {
      "entropy": 5.862979936599731,
      "epoch": 1.061784497072446,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004893563480314331,
      "loss": 5.3661,
      "mean_token_accuracy": 0.18943110704421998,
      "num_tokens": 24612426.0,
      "step": 12150
    },
    {
      "entropy": 5.838386917114258,
      "epoch": 1.0622214454251508,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004893468752334294,
      "loss": 5.392,
      "mean_token_accuracy": 0.18316657990217208,
      "num_tokens": 24623031.0,
      "step": 12155
    },
    {
      "entropy": 5.834109210968018,
      "epoch": 1.0626583937778555,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004893373983241404,
      "loss": 5.4353,
      "mean_token_accuracy": 0.18620124608278274,
      "num_tokens": 24633763.0,
      "step": 12160
    },
    {
      "entropy": 5.854263782501221,
      "epoch": 1.0630953421305602,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004893279173037478,
      "loss": 5.4518,
      "mean_token_accuracy": 0.18241489082574844,
      "num_tokens": 24644340.0,
      "step": 12165
    },
    {
      "entropy": 5.909046173095703,
      "epoch": 1.063532290483265,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004893184321724335,
      "loss": 5.4371,
      "mean_token_accuracy": 0.193858739733696,
      "num_tokens": 24654751.0,
      "step": 12170
    },
    {
      "entropy": 5.835341262817383,
      "epoch": 1.0639692388359696,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004893089429303792,
      "loss": 5.4305,
      "mean_token_accuracy": 0.1900732025504112,
      "num_tokens": 24665134.0,
      "step": 12175
    },
    {
      "entropy": 5.837165403366089,
      "epoch": 1.0644061871886743,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004892994495777672,
      "loss": 5.414,
      "mean_token_accuracy": 0.18636189103126527,
      "num_tokens": 24673796.0,
      "step": 12180
    },
    {
      "entropy": 5.819138288497925,
      "epoch": 1.064843135541379,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004892899521147794,
      "loss": 5.3569,
      "mean_token_accuracy": 0.1958388164639473,
      "num_tokens": 24684104.0,
      "step": 12185
    },
    {
      "entropy": 5.784662866592408,
      "epoch": 1.0652800838940837,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000489280450541598,
      "loss": 5.3793,
      "mean_token_accuracy": 0.19136970043182372,
      "num_tokens": 24693994.0,
      "step": 12190
    },
    {
      "entropy": 5.8527665615081785,
      "epoch": 1.0657170322467884,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004892709448584054,
      "loss": 5.4388,
      "mean_token_accuracy": 0.18928470015525817,
      "num_tokens": 24703815.0,
      "step": 12195
    },
    {
      "entropy": 5.893473434448242,
      "epoch": 1.066153980599493,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004892614350653837,
      "loss": 5.3901,
      "mean_token_accuracy": 0.18365359753370286,
      "num_tokens": 24713631.0,
      "step": 12200
    },
    {
      "entropy": 5.920675420761109,
      "epoch": 1.0665909289521978,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004892519211627156,
      "loss": 5.4027,
      "mean_token_accuracy": 0.1897514745593071,
      "num_tokens": 24723353.0,
      "step": 12205
    },
    {
      "entropy": 5.833736610412598,
      "epoch": 1.0670278773049025,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004892424031505832,
      "loss": 5.3945,
      "mean_token_accuracy": 0.18426562398672103,
      "num_tokens": 24733246.0,
      "step": 12210
    },
    {
      "entropy": 5.814574289321899,
      "epoch": 1.0674648256576074,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004892328810291693,
      "loss": 5.3998,
      "mean_token_accuracy": 0.1890753135085106,
      "num_tokens": 24743097.0,
      "step": 12215
    },
    {
      "entropy": 5.855318164825439,
      "epoch": 1.067901774010312,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004892233547986564,
      "loss": 5.3595,
      "mean_token_accuracy": 0.1867317721247673,
      "num_tokens": 24752618.0,
      "step": 12220
    },
    {
      "entropy": 5.886608457565307,
      "epoch": 1.0683387223630167,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004892138244592274,
      "loss": 5.4527,
      "mean_token_accuracy": 0.18959045857191087,
      "num_tokens": 24763960.0,
      "step": 12225
    },
    {
      "entropy": 5.8215094089508055,
      "epoch": 1.0687756707157214,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004892042900110649,
      "loss": 5.3168,
      "mean_token_accuracy": 0.19419558644294738,
      "num_tokens": 24773040.0,
      "step": 12230
    },
    {
      "entropy": 5.798424625396729,
      "epoch": 1.0692126190684261,
      "grad_norm": 0.984375,
      "learning_rate": 0.000489194751454352,
      "loss": 5.4228,
      "mean_token_accuracy": 0.18658418357372283,
      "num_tokens": 24783049.0,
      "step": 12235
    },
    {
      "entropy": 5.901490306854248,
      "epoch": 1.0696495674211308,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004891852087892714,
      "loss": 5.4363,
      "mean_token_accuracy": 0.18181039541959762,
      "num_tokens": 24794590.0,
      "step": 12240
    },
    {
      "entropy": 5.809011745452881,
      "epoch": 1.0700865157738355,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004891756620160064,
      "loss": 5.3221,
      "mean_token_accuracy": 0.19431810677051545,
      "num_tokens": 24805146.0,
      "step": 12245
    },
    {
      "entropy": 5.869487380981445,
      "epoch": 1.0705234641265402,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004891661111347399,
      "loss": 5.4624,
      "mean_token_accuracy": 0.18381544649600984,
      "num_tokens": 24815573.0,
      "step": 12250
    },
    {
      "entropy": 5.845509338378906,
      "epoch": 1.070960412479245,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004891565561456552,
      "loss": 5.3575,
      "mean_token_accuracy": 0.19724326878786086,
      "num_tokens": 24825417.0,
      "step": 12255
    },
    {
      "entropy": 5.768044519424438,
      "epoch": 1.0713973608319496,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004891469970489355,
      "loss": 5.3845,
      "mean_token_accuracy": 0.1929337352514267,
      "num_tokens": 24835589.0,
      "step": 12260
    },
    {
      "entropy": 5.850324964523315,
      "epoch": 1.0718343091846543,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004891374338447642,
      "loss": 5.4898,
      "mean_token_accuracy": 0.18145294934511186,
      "num_tokens": 24846603.0,
      "step": 12265
    },
    {
      "entropy": 5.94133038520813,
      "epoch": 1.0722712575373592,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004891278665333247,
      "loss": 5.4233,
      "mean_token_accuracy": 0.18630459457635878,
      "num_tokens": 24856368.0,
      "step": 12270
    },
    {
      "entropy": 5.787446784973144,
      "epoch": 1.0727082058900639,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004891182951148005,
      "loss": 5.3235,
      "mean_token_accuracy": 0.190731917321682,
      "num_tokens": 24866322.0,
      "step": 12275
    },
    {
      "entropy": 5.814941453933716,
      "epoch": 1.0731451542427686,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004891087195893753,
      "loss": 5.428,
      "mean_token_accuracy": 0.1863328218460083,
      "num_tokens": 24876686.0,
      "step": 12280
    },
    {
      "entropy": 5.847650051116943,
      "epoch": 1.0735821025954733,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004890991399572325,
      "loss": 5.3709,
      "mean_token_accuracy": 0.18620405793190004,
      "num_tokens": 24886561.0,
      "step": 12285
    },
    {
      "entropy": 5.839345502853393,
      "epoch": 1.074019050948178,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004890895562185562,
      "loss": 5.39,
      "mean_token_accuracy": 0.18674342185258866,
      "num_tokens": 24896010.0,
      "step": 12290
    },
    {
      "entropy": 5.8830467700958256,
      "epoch": 1.0744559993008826,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004890799683735299,
      "loss": 5.4617,
      "mean_token_accuracy": 0.17984665036201478,
      "num_tokens": 24904852.0,
      "step": 12295
    },
    {
      "entropy": 5.915183973312378,
      "epoch": 1.0748929476535873,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004890703764223377,
      "loss": 5.4482,
      "mean_token_accuracy": 0.18362002521753312,
      "num_tokens": 24914828.0,
      "step": 12300
    },
    {
      "entropy": 5.920379781723023,
      "epoch": 1.075329896006292,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004890607803651635,
      "loss": 5.4926,
      "mean_token_accuracy": 0.17208389639854432,
      "num_tokens": 24925602.0,
      "step": 12305
    },
    {
      "entropy": 5.871938610076905,
      "epoch": 1.0757668443589967,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004890511802021914,
      "loss": 5.4474,
      "mean_token_accuracy": 0.1911897838115692,
      "num_tokens": 24935175.0,
      "step": 12310
    },
    {
      "entropy": 5.91336760520935,
      "epoch": 1.0762037927117014,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004890415759336053,
      "loss": 5.4558,
      "mean_token_accuracy": 0.17964351326227188,
      "num_tokens": 24945828.0,
      "step": 12315
    },
    {
      "entropy": 5.870431280136108,
      "epoch": 1.076640741064406,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004890319675595899,
      "loss": 5.3948,
      "mean_token_accuracy": 0.19136930257081985,
      "num_tokens": 24955400.0,
      "step": 12320
    },
    {
      "entropy": 5.829915285110474,
      "epoch": 1.0770776894171108,
      "grad_norm": 0.984375,
      "learning_rate": 0.000489022355080329,
      "loss": 5.4771,
      "mean_token_accuracy": 0.18196005672216414,
      "num_tokens": 24966079.0,
      "step": 12325
    },
    {
      "entropy": 5.872962999343872,
      "epoch": 1.0775146377698157,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004890127384960073,
      "loss": 5.3639,
      "mean_token_accuracy": 0.19530875384807586,
      "num_tokens": 24975090.0,
      "step": 12330
    },
    {
      "entropy": 5.963196182250977,
      "epoch": 1.0779515861225204,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004890031178068091,
      "loss": 5.4828,
      "mean_token_accuracy": 0.17681879848241805,
      "num_tokens": 24986068.0,
      "step": 12335
    },
    {
      "entropy": 5.861756658554077,
      "epoch": 1.078388534475225,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000488993493012919,
      "loss": 5.4913,
      "mean_token_accuracy": 0.1847318544983864,
      "num_tokens": 24995697.0,
      "step": 12340
    },
    {
      "entropy": 5.898888158798218,
      "epoch": 1.0788254828279298,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004889838641145215,
      "loss": 5.4092,
      "mean_token_accuracy": 0.18068873584270478,
      "num_tokens": 25005602.0,
      "step": 12345
    },
    {
      "entropy": 5.89501051902771,
      "epoch": 1.0792624311806345,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004889742311118014,
      "loss": 5.4782,
      "mean_token_accuracy": 0.19017751067876815,
      "num_tokens": 25016407.0,
      "step": 12350
    },
    {
      "entropy": 5.9134965419769285,
      "epoch": 1.0796993795333392,
      "grad_norm": 1.0,
      "learning_rate": 0.0004889645940049434,
      "loss": 5.4764,
      "mean_token_accuracy": 0.1860598936676979,
      "num_tokens": 25025889.0,
      "step": 12355
    },
    {
      "entropy": 5.845358991622925,
      "epoch": 1.0801363278860439,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004889549527941324,
      "loss": 5.3947,
      "mean_token_accuracy": 0.19388455897569656,
      "num_tokens": 25036383.0,
      "step": 12360
    },
    {
      "entropy": 5.9047938823699955,
      "epoch": 1.0805732762387485,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004889453074795534,
      "loss": 5.5387,
      "mean_token_accuracy": 0.17700949758291246,
      "num_tokens": 25045718.0,
      "step": 12365
    },
    {
      "entropy": 5.905872058868408,
      "epoch": 1.0810102245914532,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004889356580613911,
      "loss": 5.5107,
      "mean_token_accuracy": 0.18072729110717772,
      "num_tokens": 25056244.0,
      "step": 12370
    },
    {
      "entropy": 5.879692029953003,
      "epoch": 1.081447172944158,
      "grad_norm": 1.015625,
      "learning_rate": 0.000488926004539831,
      "loss": 5.4477,
      "mean_token_accuracy": 0.18277403116226196,
      "num_tokens": 25065954.0,
      "step": 12375
    },
    {
      "entropy": 5.8667138576507565,
      "epoch": 1.0818841212968626,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004889163469150578,
      "loss": 5.4731,
      "mean_token_accuracy": 0.1834900602698326,
      "num_tokens": 25075851.0,
      "step": 12380
    },
    {
      "entropy": 5.903693675994873,
      "epoch": 1.0823210696495673,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004889066851872571,
      "loss": 5.4349,
      "mean_token_accuracy": 0.18710750192403794,
      "num_tokens": 25085324.0,
      "step": 12385
    },
    {
      "entropy": 5.914773273468017,
      "epoch": 1.0827580180022722,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000488897019356614,
      "loss": 5.465,
      "mean_token_accuracy": 0.18521395623683928,
      "num_tokens": 25096566.0,
      "step": 12390
    },
    {
      "entropy": 5.988368129730224,
      "epoch": 1.083194966354977,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004888873494233141,
      "loss": 5.5384,
      "mean_token_accuracy": 0.17487022578716277,
      "num_tokens": 25107642.0,
      "step": 12395
    },
    {
      "entropy": 5.898044586181641,
      "epoch": 1.0836319147076816,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004888776753875427,
      "loss": 5.4267,
      "mean_token_accuracy": 0.1876099705696106,
      "num_tokens": 25117488.0,
      "step": 12400
    },
    {
      "entropy": 5.876465702056885,
      "epoch": 1.0840688630603863,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004888679972494854,
      "loss": 5.3971,
      "mean_token_accuracy": 0.1878292068839073,
      "num_tokens": 25127107.0,
      "step": 12405
    },
    {
      "entropy": 5.819347286224366,
      "epoch": 1.084505811413091,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000488858315009328,
      "loss": 5.4212,
      "mean_token_accuracy": 0.1898447871208191,
      "num_tokens": 25138272.0,
      "step": 12410
    },
    {
      "entropy": 5.878476572036743,
      "epoch": 1.0849427597657957,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004888486286672559,
      "loss": 5.4548,
      "mean_token_accuracy": 0.18013087511062623,
      "num_tokens": 25147128.0,
      "step": 12415
    },
    {
      "entropy": 5.899073314666748,
      "epoch": 1.0853797081185004,
      "grad_norm": 1.0,
      "learning_rate": 0.0004888389382234551,
      "loss": 5.3675,
      "mean_token_accuracy": 0.19446575194597243,
      "num_tokens": 25157418.0,
      "step": 12420
    },
    {
      "entropy": 5.847907066345215,
      "epoch": 1.085816656471205,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004888292436781114,
      "loss": 5.4135,
      "mean_token_accuracy": 0.18896514177322388,
      "num_tokens": 25168123.0,
      "step": 12425
    },
    {
      "entropy": 5.8020306587219235,
      "epoch": 1.0862536048239098,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004888195450314108,
      "loss": 5.444,
      "mean_token_accuracy": 0.19098411947488786,
      "num_tokens": 25178445.0,
      "step": 12430
    },
    {
      "entropy": 5.886538362503051,
      "epoch": 1.0866905531766144,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004888098422835392,
      "loss": 5.4691,
      "mean_token_accuracy": 0.1794024869799614,
      "num_tokens": 25189246.0,
      "step": 12435
    },
    {
      "entropy": 5.876694059371948,
      "epoch": 1.0871275015293191,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004888001354346828,
      "loss": 5.4498,
      "mean_token_accuracy": 0.18064207285642625,
      "num_tokens": 25201277.0,
      "step": 12440
    },
    {
      "entropy": 5.9135884761810305,
      "epoch": 1.087564449882024,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004887904244850279,
      "loss": 5.4268,
      "mean_token_accuracy": 0.18815436512231826,
      "num_tokens": 25211353.0,
      "step": 12445
    },
    {
      "entropy": 5.876775026321411,
      "epoch": 1.0880013982347287,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004887807094347606,
      "loss": 5.4172,
      "mean_token_accuracy": 0.18467023670673371,
      "num_tokens": 25222236.0,
      "step": 12450
    },
    {
      "entropy": 5.8157988548278805,
      "epoch": 1.0884383465874334,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004887709902840672,
      "loss": 5.3639,
      "mean_token_accuracy": 0.19288412630558013,
      "num_tokens": 25232464.0,
      "step": 12455
    },
    {
      "entropy": 5.871794080734253,
      "epoch": 1.0888752949401381,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004887612670331342,
      "loss": 5.4228,
      "mean_token_accuracy": 0.18400197476148605,
      "num_tokens": 25243518.0,
      "step": 12460
    },
    {
      "entropy": 5.855633401870728,
      "epoch": 1.0893122432928428,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004887515396821481,
      "loss": 5.4296,
      "mean_token_accuracy": 0.18178863376379012,
      "num_tokens": 25253168.0,
      "step": 12465
    },
    {
      "entropy": 5.818670177459717,
      "epoch": 1.0897491916455475,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004887418082312955,
      "loss": 5.3932,
      "mean_token_accuracy": 0.19031433910131454,
      "num_tokens": 25262750.0,
      "step": 12470
    },
    {
      "entropy": 5.876470375061035,
      "epoch": 1.0901861399982522,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000488732072680763,
      "loss": 5.4063,
      "mean_token_accuracy": 0.1890619471669197,
      "num_tokens": 25273520.0,
      "step": 12475
    },
    {
      "entropy": 5.868398427963257,
      "epoch": 1.090623088350957,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004887223330307373,
      "loss": 5.452,
      "mean_token_accuracy": 0.18422561883926392,
      "num_tokens": 25284136.0,
      "step": 12480
    },
    {
      "entropy": 5.89251217842102,
      "epoch": 1.0910600367036616,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004887125892814053,
      "loss": 5.3689,
      "mean_token_accuracy": 0.1854684606194496,
      "num_tokens": 25294523.0,
      "step": 12485
    },
    {
      "entropy": 5.840603828430176,
      "epoch": 1.0914969850563663,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004887028414329538,
      "loss": 5.4164,
      "mean_token_accuracy": 0.18691862225532532,
      "num_tokens": 25305050.0,
      "step": 12490
    },
    {
      "entropy": 5.873907661437988,
      "epoch": 1.091933933409071,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048869308948557,
      "loss": 5.4317,
      "mean_token_accuracy": 0.18690309822559356,
      "num_tokens": 25314445.0,
      "step": 12495
    },
    {
      "entropy": 5.879892778396607,
      "epoch": 1.0923708817617757,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004886833334394406,
      "loss": 5.4287,
      "mean_token_accuracy": 0.18535748422145842,
      "num_tokens": 25324426.0,
      "step": 12500
    },
    {
      "entropy": 5.843155813217163,
      "epoch": 1.0928078301144806,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004886735732947529,
      "loss": 5.4158,
      "mean_token_accuracy": 0.18845745474100112,
      "num_tokens": 25335051.0,
      "step": 12505
    },
    {
      "entropy": 5.843862581253052,
      "epoch": 1.0932447784671853,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004886638090516942,
      "loss": 5.3395,
      "mean_token_accuracy": 0.19064219146966935,
      "num_tokens": 25343955.0,
      "step": 12510
    },
    {
      "entropy": 5.805636405944824,
      "epoch": 1.09368172681989,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004886540407104517,
      "loss": 5.4272,
      "mean_token_accuracy": 0.1845192164182663,
      "num_tokens": 25353817.0,
      "step": 12515
    },
    {
      "entropy": 5.933856105804443,
      "epoch": 1.0941186751725946,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004886442682712126,
      "loss": 5.359,
      "mean_token_accuracy": 0.19068515598773955,
      "num_tokens": 25363882.0,
      "step": 12520
    },
    {
      "entropy": 5.873227071762085,
      "epoch": 1.0945556235252993,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004886344917341646,
      "loss": 5.3844,
      "mean_token_accuracy": 0.18580599278211593,
      "num_tokens": 25373721.0,
      "step": 12525
    },
    {
      "entropy": 5.870410490036011,
      "epoch": 1.094992571878004,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004886247110994952,
      "loss": 5.405,
      "mean_token_accuracy": 0.18613789081573487,
      "num_tokens": 25383520.0,
      "step": 12530
    },
    {
      "entropy": 5.849159812927246,
      "epoch": 1.0954295202307087,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004886149263673916,
      "loss": 5.3989,
      "mean_token_accuracy": 0.18912893235683442,
      "num_tokens": 25394039.0,
      "step": 12535
    },
    {
      "entropy": 5.863416624069214,
      "epoch": 1.0958664685834134,
      "grad_norm": 1.078125,
      "learning_rate": 0.000488605137538042,
      "loss": 5.413,
      "mean_token_accuracy": 0.18931461721658707,
      "num_tokens": 25403404.0,
      "step": 12540
    },
    {
      "entropy": 5.820438861846924,
      "epoch": 1.096303416936118,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004885953446116339,
      "loss": 5.4096,
      "mean_token_accuracy": 0.1867101937532425,
      "num_tokens": 25413800.0,
      "step": 12545
    },
    {
      "entropy": 5.826465272903443,
      "epoch": 1.0967403652888228,
      "grad_norm": 1.0,
      "learning_rate": 0.0004885855475883551,
      "loss": 5.3861,
      "mean_token_accuracy": 0.19400328844785691,
      "num_tokens": 25424252.0,
      "step": 12550
    },
    {
      "entropy": 5.913479042053223,
      "epoch": 1.0971773136415275,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004885757464683937,
      "loss": 5.3944,
      "mean_token_accuracy": 0.1906462088227272,
      "num_tokens": 25433381.0,
      "step": 12555
    },
    {
      "entropy": 5.813988780975341,
      "epoch": 1.0976142619942322,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004885659412519376,
      "loss": 5.3599,
      "mean_token_accuracy": 0.1931947335600853,
      "num_tokens": 25443253.0,
      "step": 12560
    },
    {
      "entropy": 5.808672523498535,
      "epoch": 1.098051210346937,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004885561319391747,
      "loss": 5.4454,
      "mean_token_accuracy": 0.18655913770198823,
      "num_tokens": 25453590.0,
      "step": 12565
    },
    {
      "entropy": 5.852259349822998,
      "epoch": 1.0984881586996418,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004885463185302934,
      "loss": 5.3641,
      "mean_token_accuracy": 0.19692321866750717,
      "num_tokens": 25464065.0,
      "step": 12570
    },
    {
      "entropy": 5.856627321243286,
      "epoch": 1.0989251070523465,
      "grad_norm": 1.125,
      "learning_rate": 0.0004885365010254817,
      "loss": 5.4692,
      "mean_token_accuracy": 0.18393315225839615,
      "num_tokens": 25475236.0,
      "step": 12575
    },
    {
      "entropy": 5.887792301177979,
      "epoch": 1.0993620554050512,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004885266794249281,
      "loss": 5.5045,
      "mean_token_accuracy": 0.183071468770504,
      "num_tokens": 25485200.0,
      "step": 12580
    },
    {
      "entropy": 5.910201740264893,
      "epoch": 1.0997990037577559,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004885168537288209,
      "loss": 5.4175,
      "mean_token_accuracy": 0.19107471853494645,
      "num_tokens": 25494775.0,
      "step": 12585
    },
    {
      "entropy": 5.900674438476562,
      "epoch": 1.1002359521104605,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004885070239373484,
      "loss": 5.4493,
      "mean_token_accuracy": 0.18171190619468688,
      "num_tokens": 25504435.0,
      "step": 12590
    },
    {
      "entropy": 5.817011594772339,
      "epoch": 1.1006729004631652,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004884971900506995,
      "loss": 5.4418,
      "mean_token_accuracy": 0.18118111938238143,
      "num_tokens": 25514085.0,
      "step": 12595
    },
    {
      "entropy": 5.90753698348999,
      "epoch": 1.10110984881587,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004884873520690626,
      "loss": 5.4957,
      "mean_token_accuracy": 0.17907336354255676,
      "num_tokens": 25523988.0,
      "step": 12600
    },
    {
      "entropy": 5.867447566986084,
      "epoch": 1.1015467971685746,
      "grad_norm": 1.0,
      "learning_rate": 0.0004884775099926264,
      "loss": 5.4662,
      "mean_token_accuracy": 0.1843200221657753,
      "num_tokens": 25534469.0,
      "step": 12605
    },
    {
      "entropy": 5.891085767745972,
      "epoch": 1.1019837455212793,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004884676638215797,
      "loss": 5.4988,
      "mean_token_accuracy": 0.17859067767858505,
      "num_tokens": 25544854.0,
      "step": 12610
    },
    {
      "entropy": 5.897203493118286,
      "epoch": 1.102420693873984,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004884578135561113,
      "loss": 5.3935,
      "mean_token_accuracy": 0.18677821308374404,
      "num_tokens": 25555465.0,
      "step": 12615
    },
    {
      "entropy": 5.801769256591797,
      "epoch": 1.102857642226689,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004884479591964103,
      "loss": 5.4641,
      "mean_token_accuracy": 0.1888738289475441,
      "num_tokens": 25565949.0,
      "step": 12620
    },
    {
      "entropy": 5.884456396102905,
      "epoch": 1.1032945905793936,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004884381007426654,
      "loss": 5.4431,
      "mean_token_accuracy": 0.18391295224428178,
      "num_tokens": 25576511.0,
      "step": 12625
    },
    {
      "entropy": 5.817424726486206,
      "epoch": 1.1037315389320983,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004884282381950661,
      "loss": 5.343,
      "mean_token_accuracy": 0.1873527631163597,
      "num_tokens": 25587224.0,
      "step": 12630
    },
    {
      "entropy": 5.883246803283692,
      "epoch": 1.104168487284803,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004884183715538012,
      "loss": 5.4229,
      "mean_token_accuracy": 0.18362582325935364,
      "num_tokens": 25596696.0,
      "step": 12635
    },
    {
      "entropy": 5.7983910083770756,
      "epoch": 1.1046054356375077,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004884085008190602,
      "loss": 5.3826,
      "mean_token_accuracy": 0.19483473896980286,
      "num_tokens": 25606829.0,
      "step": 12640
    },
    {
      "entropy": 5.846244668960571,
      "epoch": 1.1050423839902124,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004883986259910323,
      "loss": 5.4379,
      "mean_token_accuracy": 0.19009916633367538,
      "num_tokens": 25617122.0,
      "step": 12645
    },
    {
      "entropy": 5.873322534561157,
      "epoch": 1.105479332342917,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004883887470699069,
      "loss": 5.3813,
      "mean_token_accuracy": 0.18181691467761993,
      "num_tokens": 25626485.0,
      "step": 12650
    },
    {
      "entropy": 5.879874753952026,
      "epoch": 1.1059162806956218,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004883788640558735,
      "loss": 5.3875,
      "mean_token_accuracy": 0.1882704421877861,
      "num_tokens": 25636695.0,
      "step": 12655
    },
    {
      "entropy": 5.691989612579346,
      "epoch": 1.1063532290483264,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004883689769491218,
      "loss": 5.254,
      "mean_token_accuracy": 0.2022020325064659,
      "num_tokens": 25646169.0,
      "step": 12660
    },
    {
      "entropy": 5.836738252639771,
      "epoch": 1.1067901774010311,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004883590857498413,
      "loss": 5.46,
      "mean_token_accuracy": 0.18475189059972763,
      "num_tokens": 25656728.0,
      "step": 12665
    },
    {
      "entropy": 5.911673355102539,
      "epoch": 1.1072271257537358,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004883491904582217,
      "loss": 5.4041,
      "mean_token_accuracy": 0.19013499319553376,
      "num_tokens": 25666184.0,
      "step": 12670
    },
    {
      "entropy": 5.883521842956543,
      "epoch": 1.1076640741064405,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004883392910744528,
      "loss": 5.4309,
      "mean_token_accuracy": 0.18872207999229432,
      "num_tokens": 25675571.0,
      "step": 12675
    },
    {
      "entropy": 5.905822515487671,
      "epoch": 1.1081010224591452,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004883293875987245,
      "loss": 5.4351,
      "mean_token_accuracy": 0.18367994874715804,
      "num_tokens": 25685485.0,
      "step": 12680
    },
    {
      "entropy": 5.85268268585205,
      "epoch": 1.1085379708118501,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004883194800312268,
      "loss": 5.3367,
      "mean_token_accuracy": 0.1890265315771103,
      "num_tokens": 25695278.0,
      "step": 12685
    },
    {
      "entropy": 5.843277788162231,
      "epoch": 1.1089749191645548,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004883095683721497,
      "loss": 5.4108,
      "mean_token_accuracy": 0.19320338368415832,
      "num_tokens": 25704410.0,
      "step": 12690
    },
    {
      "entropy": 5.870254325866699,
      "epoch": 1.1094118675172595,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004882996526216833,
      "loss": 5.4139,
      "mean_token_accuracy": 0.1879218727350235,
      "num_tokens": 25715027.0,
      "step": 12695
    },
    {
      "entropy": 5.94524827003479,
      "epoch": 1.1098488158699642,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004882897327800178,
      "loss": 5.4387,
      "mean_token_accuracy": 0.18612510561943055,
      "num_tokens": 25723992.0,
      "step": 12700
    },
    {
      "entropy": 5.836612176895142,
      "epoch": 1.110285764222669,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004882798088473435,
      "loss": 5.3939,
      "mean_token_accuracy": 0.184440453350544,
      "num_tokens": 25733791.0,
      "step": 12705
    },
    {
      "entropy": 5.796571540832519,
      "epoch": 1.1107227125753736,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048826988082385065,
      "loss": 5.4216,
      "mean_token_accuracy": 0.17769196033477783,
      "num_tokens": 25743258.0,
      "step": 12710
    },
    {
      "entropy": 5.9105181217193605,
      "epoch": 1.1111596609280783,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048825994870972976,
      "loss": 5.4337,
      "mean_token_accuracy": 0.18345945477485656,
      "num_tokens": 25753183.0,
      "step": 12715
    },
    {
      "entropy": 5.944032526016235,
      "epoch": 1.111596609280783,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048825001250517124,
      "loss": 5.4599,
      "mean_token_accuracy": 0.18412081748247147,
      "num_tokens": 25763308.0,
      "step": 12720
    },
    {
      "entropy": 5.815320062637329,
      "epoch": 1.1120335576334877,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004882400722103658,
      "loss": 5.3632,
      "mean_token_accuracy": 0.19564688056707383,
      "num_tokens": 25774269.0,
      "step": 12725
    },
    {
      "entropy": 5.791265153884888,
      "epoch": 1.1124705059861923,
      "grad_norm": 1.09375,
      "learning_rate": 0.000488230127825504,
      "loss": 5.3274,
      "mean_token_accuracy": 0.1961403891444206,
      "num_tokens": 25785194.0,
      "step": 12730
    },
    {
      "entropy": 5.871878623962402,
      "epoch": 1.112907454338897,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048822017935077673,
      "loss": 5.4101,
      "mean_token_accuracy": 0.19022056162357331,
      "num_tokens": 25793927.0,
      "step": 12735
    },
    {
      "entropy": 5.83724799156189,
      "epoch": 1.113344402691602,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004882102267863746,
      "loss": 5.4676,
      "mean_token_accuracy": 0.18854027092456818,
      "num_tokens": 25805395.0,
      "step": 12740
    },
    {
      "entropy": 5.883472585678101,
      "epoch": 1.1137813510443066,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048820027013248864,
      "loss": 5.4849,
      "mean_token_accuracy": 0.18796162456274032,
      "num_tokens": 25816594.0,
      "step": 12745
    },
    {
      "entropy": 5.903038215637207,
      "epoch": 1.1142182993970113,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048819030938930974,
      "loss": 5.3475,
      "mean_token_accuracy": 0.19175535589456558,
      "num_tokens": 25825943.0,
      "step": 12750
    },
    {
      "entropy": 5.807853698730469,
      "epoch": 1.114655247749716,
      "grad_norm": 1.0,
      "learning_rate": 0.000488180344557029,
      "loss": 5.4003,
      "mean_token_accuracy": 0.18803260177373887,
      "num_tokens": 25835928.0,
      "step": 12755
    },
    {
      "entropy": 5.8199379444122314,
      "epoch": 1.1150921961024207,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004881703756358376,
      "loss": 5.4011,
      "mean_token_accuracy": 0.1851260930299759,
      "num_tokens": 25847363.0,
      "step": 12760
    },
    {
      "entropy": 5.88535623550415,
      "epoch": 1.1155291444551254,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004881604026259266,
      "loss": 5.426,
      "mean_token_accuracy": 0.1896740734577179,
      "num_tokens": 25857246.0,
      "step": 12765
    },
    {
      "entropy": 5.898195266723633,
      "epoch": 1.11596609280783,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048815042552748745,
      "loss": 5.4467,
      "mean_token_accuracy": 0.18312453031539916,
      "num_tokens": 25866615.0,
      "step": 12770
    },
    {
      "entropy": 5.904406785964966,
      "epoch": 1.1164030411605348,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004881404443407115,
      "loss": 5.5205,
      "mean_token_accuracy": 0.17854616343975066,
      "num_tokens": 25876698.0,
      "step": 12775
    },
    {
      "entropy": 5.964991664886474,
      "epoch": 1.1168399895132395,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048813045906579,
      "loss": 5.4691,
      "mean_token_accuracy": 0.18120414167642593,
      "num_tokens": 25888029.0,
      "step": 12780
    },
    {
      "entropy": 5.906395149230957,
      "epoch": 1.1172769378659442,
      "grad_norm": 1.125,
      "learning_rate": 0.00048812046970291473,
      "loss": 5.4908,
      "mean_token_accuracy": 0.18056789189577102,
      "num_tokens": 25897472.0,
      "step": 12785
    },
    {
      "entropy": 5.908102369308471,
      "epoch": 1.1177138862186489,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004881104762522772,
      "loss": 5.4893,
      "mean_token_accuracy": 0.183821901679039,
      "num_tokens": 25908025.0,
      "step": 12790
    },
    {
      "entropy": 5.917673110961914,
      "epoch": 1.1181508345713538,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000488100478714069,
      "loss": 5.4203,
      "mean_token_accuracy": 0.18740236163139343,
      "num_tokens": 25919110.0,
      "step": 12795
    },
    {
      "entropy": 5.8550482273101805,
      "epoch": 1.1185877829240585,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048809047708848194,
      "loss": 5.3533,
      "mean_token_accuracy": 0.19717698991298677,
      "num_tokens": 25928322.0,
      "step": 12800
    },
    {
      "entropy": 5.842849016189575,
      "epoch": 1.1190247312767632,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048808047137570795,
      "loss": 5.3599,
      "mean_token_accuracy": 0.1845258429646492,
      "num_tokens": 25937347.0,
      "step": 12805
    },
    {
      "entropy": 5.794095611572265,
      "epoch": 1.1194616796294679,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004880704615759388,
      "loss": 5.3456,
      "mean_token_accuracy": 0.1885373666882515,
      "num_tokens": 25948408.0,
      "step": 12810
    },
    {
      "entropy": 5.853734064102173,
      "epoch": 1.1198986279821725,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048806044768936655,
      "loss": 5.473,
      "mean_token_accuracy": 0.18463574349880219,
      "num_tokens": 25958249.0,
      "step": 12815
    },
    {
      "entropy": 5.946908855438233,
      "epoch": 1.1203355763348772,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004880504297161833,
      "loss": 5.4765,
      "mean_token_accuracy": 0.18234734386205673,
      "num_tokens": 25968849.0,
      "step": 12820
    },
    {
      "entropy": 5.9202320098876955,
      "epoch": 1.120772524687582,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048804040765658115,
      "loss": 5.458,
      "mean_token_accuracy": 0.1866682916879654,
      "num_tokens": 25978787.0,
      "step": 12825
    },
    {
      "entropy": 5.811966514587402,
      "epoch": 1.1212094730402866,
      "grad_norm": 1.0,
      "learning_rate": 0.00048803038151075227,
      "loss": 5.4235,
      "mean_token_accuracy": 0.1863214284181595,
      "num_tokens": 25988619.0,
      "step": 12830
    },
    {
      "entropy": 5.8596696853637695,
      "epoch": 1.1216464213929913,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004880203512788892,
      "loss": 5.434,
      "mean_token_accuracy": 0.18997293561697007,
      "num_tokens": 25999080.0,
      "step": 12835
    },
    {
      "entropy": 5.874765920639038,
      "epoch": 1.122083369745696,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048801031696118406,
      "loss": 5.3821,
      "mean_token_accuracy": 0.1862630069255829,
      "num_tokens": 26010329.0,
      "step": 12840
    },
    {
      "entropy": 5.853723001480103,
      "epoch": 1.1225203180984007,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004880002785578295,
      "loss": 5.3432,
      "mean_token_accuracy": 0.19074274599552155,
      "num_tokens": 26019147.0,
      "step": 12845
    },
    {
      "entropy": 5.858485746383667,
      "epoch": 1.1229572664511054,
      "grad_norm": 1.046875,
      "learning_rate": 0.000487990236069018,
      "loss": 5.424,
      "mean_token_accuracy": 0.18378909528255463,
      "num_tokens": 26029212.0,
      "step": 12850
    },
    {
      "entropy": 5.844759798049926,
      "epoch": 1.12339421480381,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048798018949494214,
      "loss": 5.2638,
      "mean_token_accuracy": 0.19059445708990097,
      "num_tokens": 26038059.0,
      "step": 12855
    },
    {
      "entropy": 5.896092367172241,
      "epoch": 1.123831163156515,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004879701388357946,
      "loss": 5.4963,
      "mean_token_accuracy": 0.18195103853940964,
      "num_tokens": 26048076.0,
      "step": 12860
    },
    {
      "entropy": 5.876118516921997,
      "epoch": 1.1242681115092197,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004879600840917683,
      "loss": 5.4297,
      "mean_token_accuracy": 0.18582222014665603,
      "num_tokens": 26059006.0,
      "step": 12865
    },
    {
      "entropy": 5.844905376434326,
      "epoch": 1.1247050598619244,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004879500252630559,
      "loss": 5.3565,
      "mean_token_accuracy": 0.1941608741879463,
      "num_tokens": 26069275.0,
      "step": 12870
    },
    {
      "entropy": 5.8491003036499025,
      "epoch": 1.125142008214629,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004879399623498504,
      "loss": 5.3232,
      "mean_token_accuracy": 0.1939675837755203,
      "num_tokens": 26078869.0,
      "step": 12875
    },
    {
      "entropy": 5.834348964691162,
      "epoch": 1.1255789565673338,
      "grad_norm": 0.984375,
      "learning_rate": 0.000487929895352345,
      "loss": 5.3508,
      "mean_token_accuracy": 0.18701401054859162,
      "num_tokens": 26088544.0,
      "step": 12880
    },
    {
      "entropy": 5.773867797851563,
      "epoch": 1.1260159049200384,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004879198242707325,
      "loss": 5.3214,
      "mean_token_accuracy": 0.19591161757707595,
      "num_tokens": 26097503.0,
      "step": 12885
    },
    {
      "entropy": 5.9232910633087155,
      "epoch": 1.1264528532727431,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048790974910520627,
      "loss": 5.463,
      "mean_token_accuracy": 0.18823886960744857,
      "num_tokens": 26107228.0,
      "step": 12890
    },
    {
      "entropy": 5.8033782005310055,
      "epoch": 1.1268898016254478,
      "grad_norm": 1.0,
      "learning_rate": 0.00048789966985595946,
      "loss": 5.3929,
      "mean_token_accuracy": 0.19169084578752518,
      "num_tokens": 26117633.0,
      "step": 12895
    },
    {
      "entropy": 5.851162719726562,
      "epoch": 1.1273267499781525,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004878895865231854,
      "loss": 5.4444,
      "mean_token_accuracy": 0.1904565304517746,
      "num_tokens": 26127781.0,
      "step": 12900
    },
    {
      "entropy": 5.923631000518799,
      "epoch": 1.1277636983308572,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004878794991070776,
      "loss": 5.3749,
      "mean_token_accuracy": 0.1900525152683258,
      "num_tokens": 26137675.0,
      "step": 12905
    },
    {
      "entropy": 5.8813341617584225,
      "epoch": 1.128200646683562,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048786940760782937,
      "loss": 5.4531,
      "mean_token_accuracy": 0.19152350276708602,
      "num_tokens": 26148572.0,
      "step": 12910
    },
    {
      "entropy": 5.7923399925231935,
      "epoch": 1.1286375950362668,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048785931202563443,
      "loss": 5.4761,
      "mean_token_accuracy": 0.19123508185148239,
      "num_tokens": 26159261.0,
      "step": 12915
    },
    {
      "entropy": 5.82788052558899,
      "epoch": 1.1290745433889715,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004878492123606863,
      "loss": 5.4458,
      "mean_token_accuracy": 0.1898757055401802,
      "num_tokens": 26170072.0,
      "step": 12920
    },
    {
      "entropy": 5.876115179061889,
      "epoch": 1.1295114917416762,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048783910861317873,
      "loss": 5.353,
      "mean_token_accuracy": 0.18761317282915116,
      "num_tokens": 26179947.0,
      "step": 12925
    },
    {
      "entropy": 5.929945421218872,
      "epoch": 1.129948440094381,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048782900078330556,
      "loss": 5.4409,
      "mean_token_accuracy": 0.1830088183283806,
      "num_tokens": 26189982.0,
      "step": 12930
    },
    {
      "entropy": 5.9323488712310795,
      "epoch": 1.1303853884470856,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004878188888712606,
      "loss": 5.5262,
      "mean_token_accuracy": 0.17417544275522232,
      "num_tokens": 26200832.0,
      "step": 12935
    },
    {
      "entropy": 5.838306045532226,
      "epoch": 1.1308223367997903,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004878087728772379,
      "loss": 5.4567,
      "mean_token_accuracy": 0.185946723818779,
      "num_tokens": 26211506.0,
      "step": 12940
    },
    {
      "entropy": 5.872031259536743,
      "epoch": 1.131259285152495,
      "grad_norm": 1.0,
      "learning_rate": 0.0004877986528014314,
      "loss": 5.4005,
      "mean_token_accuracy": 0.19040654301643373,
      "num_tokens": 26222045.0,
      "step": 12945
    },
    {
      "entropy": 5.810594940185547,
      "epoch": 1.1316962335051997,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004877885286440353,
      "loss": 5.3219,
      "mean_token_accuracy": 0.19336472153663636,
      "num_tokens": 26231411.0,
      "step": 12950
    },
    {
      "entropy": 5.856339693069458,
      "epoch": 1.1321331818579043,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048777840040524365,
      "loss": 5.4395,
      "mean_token_accuracy": 0.18348445892333984,
      "num_tokens": 26241491.0,
      "step": 12955
    },
    {
      "entropy": 5.8819139957427975,
      "epoch": 1.132570130210609,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048776826808525074,
      "loss": 5.5056,
      "mean_token_accuracy": 0.18401160091161728,
      "num_tokens": 26252037.0,
      "step": 12960
    },
    {
      "entropy": 5.904318761825562,
      "epoch": 1.1330070785633137,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048775813168425104,
      "loss": 5.4904,
      "mean_token_accuracy": 0.18230139017105101,
      "num_tokens": 26262310.0,
      "step": 12965
    },
    {
      "entropy": 5.850106477737427,
      "epoch": 1.1334440269160186,
      "grad_norm": 1.0,
      "learning_rate": 0.00048774799120243883,
      "loss": 5.3815,
      "mean_token_accuracy": 0.1941094920039177,
      "num_tokens": 26273001.0,
      "step": 12970
    },
    {
      "entropy": 5.913328409194946,
      "epoch": 1.1338809752687231,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004877378466400087,
      "loss": 5.4637,
      "mean_token_accuracy": 0.18297885209321976,
      "num_tokens": 26283500.0,
      "step": 12975
    },
    {
      "entropy": 5.846285724639893,
      "epoch": 1.134317923621428,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004877276979971552,
      "loss": 5.3643,
      "mean_token_accuracy": 0.18881782740354539,
      "num_tokens": 26294118.0,
      "step": 12980
    },
    {
      "entropy": 5.876476144790649,
      "epoch": 1.1347548719741327,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000487717545274073,
      "loss": 5.5178,
      "mean_token_accuracy": 0.1791773185133934,
      "num_tokens": 26303833.0,
      "step": 12985
    },
    {
      "entropy": 5.9166953563690186,
      "epoch": 1.1351918203268374,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048770738847095683,
      "loss": 5.4471,
      "mean_token_accuracy": 0.186211059987545,
      "num_tokens": 26313542.0,
      "step": 12990
    },
    {
      "entropy": 5.860346508026123,
      "epoch": 1.135628768679542,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048769722758800144,
      "loss": 5.3881,
      "mean_token_accuracy": 0.19040518254041672,
      "num_tokens": 26322783.0,
      "step": 12995
    },
    {
      "entropy": 5.901995420455933,
      "epoch": 1.1360657170322468,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004876870626254018,
      "loss": 5.4672,
      "mean_token_accuracy": 0.1800348088145256,
      "num_tokens": 26333770.0,
      "step": 13000
    },
    {
      "entropy": 5.892484331130982,
      "epoch": 1.1365026653849515,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048767689358335286,
      "loss": 5.4145,
      "mean_token_accuracy": 0.18897281289100648,
      "num_tokens": 26344435.0,
      "step": 13005
    },
    {
      "entropy": 5.801534175872803,
      "epoch": 1.1369396137376562,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004876667204620497,
      "loss": 5.3769,
      "mean_token_accuracy": 0.19579947888851165,
      "num_tokens": 26355410.0,
      "step": 13010
    },
    {
      "entropy": 5.852757215499878,
      "epoch": 1.1373765620903609,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004876565432616873,
      "loss": 5.4434,
      "mean_token_accuracy": 0.18869644403457642,
      "num_tokens": 26365785.0,
      "step": 13015
    },
    {
      "entropy": 5.863480472564698,
      "epoch": 1.1378135104430656,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048764636198246104,
      "loss": 5.3894,
      "mean_token_accuracy": 0.19584612250328065,
      "num_tokens": 26375489.0,
      "step": 13020
    },
    {
      "entropy": 5.808404159545899,
      "epoch": 1.1382504587957702,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004876361766245661,
      "loss": 5.3716,
      "mean_token_accuracy": 0.19180853366851808,
      "num_tokens": 26385126.0,
      "step": 13025
    },
    {
      "entropy": 5.822219324111939,
      "epoch": 1.138687407148475,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004876259871881979,
      "loss": 5.354,
      "mean_token_accuracy": 0.19167191237211229,
      "num_tokens": 26395314.0,
      "step": 13030
    },
    {
      "entropy": 5.904667568206787,
      "epoch": 1.1391243555011799,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004876157936735519,
      "loss": 5.4453,
      "mean_token_accuracy": 0.1834098681807518,
      "num_tokens": 26404660.0,
      "step": 13035
    },
    {
      "entropy": 5.895364999771118,
      "epoch": 1.1395613038538845,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048760559608082343,
      "loss": 5.4481,
      "mean_token_accuracy": 0.1899169310927391,
      "num_tokens": 26415371.0,
      "step": 13040
    },
    {
      "entropy": 5.871814441680908,
      "epoch": 1.1399982522065892,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004875953944102084,
      "loss": 5.4217,
      "mean_token_accuracy": 0.18714421689510347,
      "num_tokens": 26424896.0,
      "step": 13045
    },
    {
      "entropy": 5.838698530197144,
      "epoch": 1.140435200559294,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048758518866190214,
      "loss": 5.3981,
      "mean_token_accuracy": 0.18518347144126893,
      "num_tokens": 26434847.0,
      "step": 13050
    },
    {
      "entropy": 5.860994625091553,
      "epoch": 1.1408721489119986,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004875749788361006,
      "loss": 5.3863,
      "mean_token_accuracy": 0.18584099262952805,
      "num_tokens": 26444138.0,
      "step": 13055
    },
    {
      "entropy": 5.906236362457276,
      "epoch": 1.1413090972647033,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004875647649329997,
      "loss": 5.394,
      "mean_token_accuracy": 0.18147762417793273,
      "num_tokens": 26453549.0,
      "step": 13060
    },
    {
      "entropy": 5.866051816940308,
      "epoch": 1.141746045617408,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004875545469527952,
      "loss": 5.4217,
      "mean_token_accuracy": 0.18720048666000366,
      "num_tokens": 26464386.0,
      "step": 13065
    },
    {
      "entropy": 5.852841234207153,
      "epoch": 1.1421829939701127,
      "grad_norm": 1.015625,
      "learning_rate": 0.000487544324895683,
      "loss": 5.3729,
      "mean_token_accuracy": 0.18901937156915666,
      "num_tokens": 26474146.0,
      "step": 13070
    },
    {
      "entropy": 5.815174865722656,
      "epoch": 1.1426199423228174,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048753409876185934,
      "loss": 5.3233,
      "mean_token_accuracy": 0.19469380825757981,
      "num_tokens": 26484408.0,
      "step": 13075
    },
    {
      "entropy": 5.856704187393189,
      "epoch": 1.143056890675522,
      "grad_norm": 1.0,
      "learning_rate": 0.0004875238685515203,
      "loss": 5.3539,
      "mean_token_accuracy": 0.18414008468389512,
      "num_tokens": 26494769.0,
      "step": 13080
    },
    {
      "entropy": 5.838315868377686,
      "epoch": 1.1434938390282268,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048751363426486206,
      "loss": 5.4307,
      "mean_token_accuracy": 0.18460324704647063,
      "num_tokens": 26505917.0,
      "step": 13085
    },
    {
      "entropy": 5.897158575057984,
      "epoch": 1.1439307873809317,
      "grad_norm": 0.984375,
      "learning_rate": 0.000487503395902081,
      "loss": 5.4361,
      "mean_token_accuracy": 0.18192838430404662,
      "num_tokens": 26516388.0,
      "step": 13090
    },
    {
      "entropy": 5.992143964767456,
      "epoch": 1.1443677357336364,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004874931534633735,
      "loss": 5.4449,
      "mean_token_accuracy": 0.18099067360162735,
      "num_tokens": 26527180.0,
      "step": 13095
    },
    {
      "entropy": 5.795349264144898,
      "epoch": 1.144804684086341,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004874829069489359,
      "loss": 5.4056,
      "mean_token_accuracy": 0.1844927817583084,
      "num_tokens": 26537705.0,
      "step": 13100
    },
    {
      "entropy": 5.87894287109375,
      "epoch": 1.1452416324390458,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048747265635896487,
      "loss": 5.5822,
      "mean_token_accuracy": 0.17742147594690322,
      "num_tokens": 26548862.0,
      "step": 13105
    },
    {
      "entropy": 5.9421850681304935,
      "epoch": 1.1456785807917504,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048746240169365683,
      "loss": 5.4314,
      "mean_token_accuracy": 0.18827054500579835,
      "num_tokens": 26559344.0,
      "step": 13110
    },
    {
      "entropy": 5.838803625106811,
      "epoch": 1.1461155291444551,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004874521429532087,
      "loss": 5.393,
      "mean_token_accuracy": 0.18893637955188752,
      "num_tokens": 26568439.0,
      "step": 13115
    },
    {
      "entropy": 5.834447431564331,
      "epoch": 1.1465524774971598,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048744188013781714,
      "loss": 5.397,
      "mean_token_accuracy": 0.1893563151359558,
      "num_tokens": 26578588.0,
      "step": 13120
    },
    {
      "entropy": 5.819237947463989,
      "epoch": 1.1469894258498645,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000487431613247679,
      "loss": 5.4822,
      "mean_token_accuracy": 0.18402200043201447,
      "num_tokens": 26588491.0,
      "step": 13125
    },
    {
      "entropy": 5.87409930229187,
      "epoch": 1.1474263742025692,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048742134228299116,
      "loss": 5.4542,
      "mean_token_accuracy": 0.18619582206010818,
      "num_tokens": 26598945.0,
      "step": 13130
    },
    {
      "entropy": 5.837648344039917,
      "epoch": 1.147863322555274,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004874110672439508,
      "loss": 5.3585,
      "mean_token_accuracy": 0.19385939687490464,
      "num_tokens": 26609842.0,
      "step": 13135
    },
    {
      "entropy": 5.934140539169311,
      "epoch": 1.1483002709079786,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004874007881307547,
      "loss": 5.4679,
      "mean_token_accuracy": 0.18283764123916627,
      "num_tokens": 26620836.0,
      "step": 13140
    },
    {
      "entropy": 5.8519447326660154,
      "epoch": 1.1487372192606835,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048739050494360026,
      "loss": 5.4463,
      "mean_token_accuracy": 0.18773891627788544,
      "num_tokens": 26630148.0,
      "step": 13145
    },
    {
      "entropy": 5.90857515335083,
      "epoch": 1.149174167613388,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048738021768268463,
      "loss": 5.4254,
      "mean_token_accuracy": 0.19007526189088822,
      "num_tokens": 26640518.0,
      "step": 13150
    },
    {
      "entropy": 5.837783288955689,
      "epoch": 1.149611115966093,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004873699263482052,
      "loss": 5.3724,
      "mean_token_accuracy": 0.1957351490855217,
      "num_tokens": 26650156.0,
      "step": 13155
    },
    {
      "entropy": 5.871954679489136,
      "epoch": 1.1500480643187976,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048735963094035924,
      "loss": 5.4275,
      "mean_token_accuracy": 0.1880139157176018,
      "num_tokens": 26661251.0,
      "step": 13160
    },
    {
      "entropy": 5.889400911331177,
      "epoch": 1.1504850126715023,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004873493314593443,
      "loss": 5.4084,
      "mean_token_accuracy": 0.1918252468109131,
      "num_tokens": 26671851.0,
      "step": 13165
    },
    {
      "entropy": 5.872822284698486,
      "epoch": 1.150921961024207,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004873390279053579,
      "loss": 5.4292,
      "mean_token_accuracy": 0.18017662763595582,
      "num_tokens": 26682984.0,
      "step": 13170
    },
    {
      "entropy": 5.915852022171021,
      "epoch": 1.1513589093769117,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004873287202785977,
      "loss": 5.3828,
      "mean_token_accuracy": 0.1897522732615471,
      "num_tokens": 26692706.0,
      "step": 13175
    },
    {
      "entropy": 5.881686496734619,
      "epoch": 1.1517958577296163,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004873184085792614,
      "loss": 5.4225,
      "mean_token_accuracy": 0.1850453123450279,
      "num_tokens": 26702152.0,
      "step": 13180
    },
    {
      "entropy": 5.792679119110107,
      "epoch": 1.152232806082321,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048730809280754674,
      "loss": 5.3819,
      "mean_token_accuracy": 0.1906711518764496,
      "num_tokens": 26711404.0,
      "step": 13185
    },
    {
      "entropy": 5.877507877349854,
      "epoch": 1.1526697544350257,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004872977729636516,
      "loss": 5.3796,
      "mean_token_accuracy": 0.1958453729748726,
      "num_tokens": 26720776.0,
      "step": 13190
    },
    {
      "entropy": 5.869913148880005,
      "epoch": 1.1531067027877304,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048728744904777394,
      "loss": 5.4314,
      "mean_token_accuracy": 0.18422608077526093,
      "num_tokens": 26730700.0,
      "step": 13195
    },
    {
      "entropy": 5.88978943824768,
      "epoch": 1.1535436511404351,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048727712106011185,
      "loss": 5.4525,
      "mean_token_accuracy": 0.18718529790639876,
      "num_tokens": 26741453.0,
      "step": 13200
    },
    {
      "entropy": 5.857434034347534,
      "epoch": 1.1539805994931398,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048726678900086327,
      "loss": 5.4202,
      "mean_token_accuracy": 0.18632753640413285,
      "num_tokens": 26750627.0,
      "step": 13205
    },
    {
      "entropy": 5.922332429885865,
      "epoch": 1.1544175478458447,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048725645287022644,
      "loss": 5.4292,
      "mean_token_accuracy": 0.18302399665117264,
      "num_tokens": 26760820.0,
      "step": 13210
    },
    {
      "entropy": 5.893594646453858,
      "epoch": 1.1548544961985494,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004872461126683996,
      "loss": 5.3367,
      "mean_token_accuracy": 0.18910249918699265,
      "num_tokens": 26771535.0,
      "step": 13215
    },
    {
      "entropy": 5.857097911834717,
      "epoch": 1.155291444551254,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004872357683955812,
      "loss": 5.457,
      "mean_token_accuracy": 0.18027589470148087,
      "num_tokens": 26782755.0,
      "step": 13220
    },
    {
      "entropy": 5.923547744750977,
      "epoch": 1.1557283929039588,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048722542005196954,
      "loss": 5.3929,
      "mean_token_accuracy": 0.1922571465373039,
      "num_tokens": 26792834.0,
      "step": 13225
    },
    {
      "entropy": 5.853617334365845,
      "epoch": 1.1561653412566635,
      "grad_norm": 1.0,
      "learning_rate": 0.00048721506763776306,
      "loss": 5.4245,
      "mean_token_accuracy": 0.18615130037069322,
      "num_tokens": 26802865.0,
      "step": 13230
    },
    {
      "entropy": 5.866951274871826,
      "epoch": 1.1566022896093682,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004872047111531604,
      "loss": 5.4099,
      "mean_token_accuracy": 0.1862918272614479,
      "num_tokens": 26813566.0,
      "step": 13235
    },
    {
      "entropy": 5.870985269546509,
      "epoch": 1.1570392379620729,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004871943505983602,
      "loss": 5.3626,
      "mean_token_accuracy": 0.19628733694553374,
      "num_tokens": 26823885.0,
      "step": 13240
    },
    {
      "entropy": 5.790228366851807,
      "epoch": 1.1574761863147776,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004871839859735612,
      "loss": 5.3631,
      "mean_token_accuracy": 0.1970012903213501,
      "num_tokens": 26833236.0,
      "step": 13245
    },
    {
      "entropy": 5.9058574676513675,
      "epoch": 1.1579131346674822,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048717361727896203,
      "loss": 5.4664,
      "mean_token_accuracy": 0.18222344368696214,
      "num_tokens": 26842651.0,
      "step": 13250
    },
    {
      "entropy": 5.8734384059906,
      "epoch": 1.158350083020187,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048716324451476183,
      "loss": 5.3683,
      "mean_token_accuracy": 0.18350328952074052,
      "num_tokens": 26852198.0,
      "step": 13255
    },
    {
      "entropy": 5.881856298446655,
      "epoch": 1.1587870313728916,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004871528676811594,
      "loss": 5.4851,
      "mean_token_accuracy": 0.18032808899879454,
      "num_tokens": 26862807.0,
      "step": 13260
    },
    {
      "entropy": 5.93935112953186,
      "epoch": 1.1592239797255965,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004871424867783538,
      "loss": 5.4623,
      "mean_token_accuracy": 0.18022799789905547,
      "num_tokens": 26871992.0,
      "step": 13265
    },
    {
      "entropy": 5.875900506973267,
      "epoch": 1.1596609280783012,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048713210180654414,
      "loss": 5.4214,
      "mean_token_accuracy": 0.18525197356939316,
      "num_tokens": 26881417.0,
      "step": 13270
    },
    {
      "entropy": 5.818054914474487,
      "epoch": 1.160097876431006,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048712171276592964,
      "loss": 5.3541,
      "mean_token_accuracy": 0.19571597576141359,
      "num_tokens": 26891879.0,
      "step": 13275
    },
    {
      "entropy": 5.835459899902344,
      "epoch": 1.1605348247837106,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004871113196567095,
      "loss": 5.4981,
      "mean_token_accuracy": 0.18209446966648102,
      "num_tokens": 26902729.0,
      "step": 13280
    },
    {
      "entropy": 5.836408996582032,
      "epoch": 1.1609717731364153,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048710092247908316,
      "loss": 5.4128,
      "mean_token_accuracy": 0.1886841744184494,
      "num_tokens": 26914415.0,
      "step": 13285
    },
    {
      "entropy": 5.9029364585876465,
      "epoch": 1.16140872148912,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048709052123324995,
      "loss": 5.4441,
      "mean_token_accuracy": 0.18865645229816436,
      "num_tokens": 26924513.0,
      "step": 13290
    },
    {
      "entropy": 5.804703950881958,
      "epoch": 1.1618456698418247,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048708011591940936,
      "loss": 5.3364,
      "mean_token_accuracy": 0.1993352875113487,
      "num_tokens": 26935039.0,
      "step": 13295
    },
    {
      "entropy": 5.831291913986206,
      "epoch": 1.1622826181945294,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000487069706537761,
      "loss": 5.3449,
      "mean_token_accuracy": 0.1940450042486191,
      "num_tokens": 26945148.0,
      "step": 13300
    },
    {
      "entropy": 5.83007550239563,
      "epoch": 1.162719566547234,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004870592930885046,
      "loss": 5.3657,
      "mean_token_accuracy": 0.18871181607246398,
      "num_tokens": 26955227.0,
      "step": 13305
    },
    {
      "entropy": 5.819167518615723,
      "epoch": 1.1631565148999388,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048704887557183993,
      "loss": 5.3559,
      "mean_token_accuracy": 0.1882757291197777,
      "num_tokens": 26964389.0,
      "step": 13310
    },
    {
      "entropy": 5.876900672912598,
      "epoch": 1.1635934632526435,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048703845398796665,
      "loss": 5.3579,
      "mean_token_accuracy": 0.19533393383026124,
      "num_tokens": 26974695.0,
      "step": 13315
    },
    {
      "entropy": 5.894908761978149,
      "epoch": 1.1640304116053484,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048702802833708466,
      "loss": 5.3477,
      "mean_token_accuracy": 0.19530709236860275,
      "num_tokens": 26984851.0,
      "step": 13320
    },
    {
      "entropy": 5.896000528335572,
      "epoch": 1.1644673599580528,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048701759861939403,
      "loss": 5.4149,
      "mean_token_accuracy": 0.19138826429843903,
      "num_tokens": 26995967.0,
      "step": 13325
    },
    {
      "entropy": 5.807231998443603,
      "epoch": 1.1649043083107578,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048700716483509476,
      "loss": 5.3582,
      "mean_token_accuracy": 0.1951881855726242,
      "num_tokens": 27006469.0,
      "step": 13330
    },
    {
      "entropy": 5.9038379192352295,
      "epoch": 1.1653412566634624,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000486996726984387,
      "loss": 5.4767,
      "mean_token_accuracy": 0.18003494441509246,
      "num_tokens": 27016419.0,
      "step": 13335
    },
    {
      "entropy": 5.910377454757691,
      "epoch": 1.1657782050161671,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004869862850674709,
      "loss": 5.4509,
      "mean_token_accuracy": 0.18395760357379914,
      "num_tokens": 27026322.0,
      "step": 13340
    },
    {
      "entropy": 5.813557147979736,
      "epoch": 1.1662151533688718,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048697583908454675,
      "loss": 5.3922,
      "mean_token_accuracy": 0.1843407556414604,
      "num_tokens": 27036518.0,
      "step": 13345
    },
    {
      "entropy": 5.9143272876739506,
      "epoch": 1.1666521017215765,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000486965389035815,
      "loss": 5.4367,
      "mean_token_accuracy": 0.183232943713665,
      "num_tokens": 27048035.0,
      "step": 13350
    },
    {
      "entropy": 5.873848628997803,
      "epoch": 1.1670890500742812,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048695493492147596,
      "loss": 5.4275,
      "mean_token_accuracy": 0.1816661536693573,
      "num_tokens": 27058507.0,
      "step": 13355
    },
    {
      "entropy": 5.857674646377563,
      "epoch": 1.167525998426986,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048694447674173025,
      "loss": 5.419,
      "mean_token_accuracy": 0.18764633387327195,
      "num_tokens": 27068350.0,
      "step": 13360
    },
    {
      "entropy": 5.790734815597534,
      "epoch": 1.1679629467796906,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004869340144967784,
      "loss": 5.3882,
      "mean_token_accuracy": 0.19035328477621077,
      "num_tokens": 27077840.0,
      "step": 13365
    },
    {
      "entropy": 5.83609619140625,
      "epoch": 1.1683998951323953,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048692354818682113,
      "loss": 5.3646,
      "mean_token_accuracy": 0.18764442950487137,
      "num_tokens": 27087687.0,
      "step": 13370
    },
    {
      "entropy": 5.922061967849731,
      "epoch": 1.1688368434851,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004869130778120591,
      "loss": 5.5159,
      "mean_token_accuracy": 0.18336150497198106,
      "num_tokens": 27099057.0,
      "step": 13375
    },
    {
      "entropy": 5.870805883407593,
      "epoch": 1.1692737918378047,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004869026033726933,
      "loss": 5.4915,
      "mean_token_accuracy": 0.18526766449213028,
      "num_tokens": 27109617.0,
      "step": 13380
    },
    {
      "entropy": 5.897283935546875,
      "epoch": 1.1697107401905096,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004868921248689244,
      "loss": 5.4387,
      "mean_token_accuracy": 0.17936182469129563,
      "num_tokens": 27119456.0,
      "step": 13385
    },
    {
      "entropy": 5.8042439937591555,
      "epoch": 1.1701476885432143,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004868816423009535,
      "loss": 5.3562,
      "mean_token_accuracy": 0.1938677817583084,
      "num_tokens": 27129333.0,
      "step": 13390
    },
    {
      "entropy": 5.8436671733856205,
      "epoch": 1.170584636895919,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004868711556689818,
      "loss": 5.3804,
      "mean_token_accuracy": 0.18369377702474593,
      "num_tokens": 27139205.0,
      "step": 13395
    },
    {
      "entropy": 5.9166782855987545,
      "epoch": 1.1710215852486237,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048686066497321027,
      "loss": 5.432,
      "mean_token_accuracy": 0.1896849051117897,
      "num_tokens": 27148530.0,
      "step": 13400
    },
    {
      "entropy": 5.911931991577148,
      "epoch": 1.1714585336013283,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048685017021384014,
      "loss": 5.4899,
      "mean_token_accuracy": 0.18330288976430892,
      "num_tokens": 27159000.0,
      "step": 13405
    },
    {
      "entropy": 5.915690755844116,
      "epoch": 1.171895481954033,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048683967139107276,
      "loss": 5.4242,
      "mean_token_accuracy": 0.1888327792286873,
      "num_tokens": 27169035.0,
      "step": 13410
    },
    {
      "entropy": 5.8834175109863285,
      "epoch": 1.1723324303067377,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004868291685051095,
      "loss": 5.4549,
      "mean_token_accuracy": 0.18169473707675934,
      "num_tokens": 27179809.0,
      "step": 13415
    },
    {
      "entropy": 5.8587016582489015,
      "epoch": 1.1727693786594424,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004868186615561518,
      "loss": 5.4107,
      "mean_token_accuracy": 0.18802549242973327,
      "num_tokens": 27190445.0,
      "step": 13420
    },
    {
      "entropy": 5.898158979415894,
      "epoch": 1.173206327012147,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048680815054440114,
      "loss": 5.3662,
      "mean_token_accuracy": 0.18900866359472274,
      "num_tokens": 27201097.0,
      "step": 13425
    },
    {
      "entropy": 5.857083988189697,
      "epoch": 1.1736432753648518,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048679763547005925,
      "loss": 5.3584,
      "mean_token_accuracy": 0.19264531284570693,
      "num_tokens": 27210181.0,
      "step": 13430
    },
    {
      "entropy": 5.857580518722534,
      "epoch": 1.1740802237175565,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004867871163333276,
      "loss": 5.3589,
      "mean_token_accuracy": 0.18831344395875932,
      "num_tokens": 27219869.0,
      "step": 13435
    },
    {
      "entropy": 5.89230318069458,
      "epoch": 1.1745171720702614,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004867765931344082,
      "loss": 5.4142,
      "mean_token_accuracy": 0.18696687519550323,
      "num_tokens": 27229730.0,
      "step": 13440
    },
    {
      "entropy": 5.86466007232666,
      "epoch": 1.174954120422966,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004867660658735027,
      "loss": 5.4199,
      "mean_token_accuracy": 0.18815792053937913,
      "num_tokens": 27239534.0,
      "step": 13445
    },
    {
      "entropy": 5.857756090164185,
      "epoch": 1.1753910687756708,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004867555345508131,
      "loss": 5.4853,
      "mean_token_accuracy": 0.17977644950151445,
      "num_tokens": 27249017.0,
      "step": 13450
    },
    {
      "entropy": 5.938849973678589,
      "epoch": 1.1758280171283755,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004867449991665414,
      "loss": 5.5135,
      "mean_token_accuracy": 0.17803620398044587,
      "num_tokens": 27261135.0,
      "step": 13455
    },
    {
      "entropy": 5.913181257247925,
      "epoch": 1.1762649654810802,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004867344597208897,
      "loss": 5.3944,
      "mean_token_accuracy": 0.18075329214334487,
      "num_tokens": 27270763.0,
      "step": 13460
    },
    {
      "entropy": 5.884515619277954,
      "epoch": 1.1767019138337849,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004867239162140601,
      "loss": 5.441,
      "mean_token_accuracy": 0.1851896286010742,
      "num_tokens": 27280115.0,
      "step": 13465
    },
    {
      "entropy": 5.86918830871582,
      "epoch": 1.1771388621864896,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004867133686462548,
      "loss": 5.3905,
      "mean_token_accuracy": 0.186837862432003,
      "num_tokens": 27290162.0,
      "step": 13470
    },
    {
      "entropy": 5.824570989608764,
      "epoch": 1.1775758105391942,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048670281701767616,
      "loss": 5.3616,
      "mean_token_accuracy": 0.19660637974739076,
      "num_tokens": 27300467.0,
      "step": 13475
    },
    {
      "entropy": 5.847992372512818,
      "epoch": 1.178012758891899,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004866922613285266,
      "loss": 5.3979,
      "mean_token_accuracy": 0.1910686105489731,
      "num_tokens": 27310524.0,
      "step": 13480
    },
    {
      "entropy": 5.8818882465362545,
      "epoch": 1.1784497072446036,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004866817015790085,
      "loss": 5.4357,
      "mean_token_accuracy": 0.1889653444290161,
      "num_tokens": 27320861.0,
      "step": 13485
    },
    {
      "entropy": 5.879636287689209,
      "epoch": 1.1788866555973083,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048667113776932453,
      "loss": 5.4421,
      "mean_token_accuracy": 0.18724810779094697,
      "num_tokens": 27331771.0,
      "step": 13490
    },
    {
      "entropy": 5.8989159107208256,
      "epoch": 1.1793236039500132,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004866605698996772,
      "loss": 5.4145,
      "mean_token_accuracy": 0.18520507216453552,
      "num_tokens": 27341449.0,
      "step": 13495
    },
    {
      "entropy": 5.88829231262207,
      "epoch": 1.1797605523027177,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004866499979702691,
      "loss": 5.4954,
      "mean_token_accuracy": 0.1802186980843544,
      "num_tokens": 27353127.0,
      "step": 13500
    },
    {
      "entropy": 5.882900762557983,
      "epoch": 1.1801975006554226,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004866394219813032,
      "loss": 5.3726,
      "mean_token_accuracy": 0.18414562642574311,
      "num_tokens": 27362116.0,
      "step": 13505
    },
    {
      "entropy": 5.866848564147949,
      "epoch": 1.1806344490081273,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048662884193298233,
      "loss": 5.3944,
      "mean_token_accuracy": 0.19529368728399277,
      "num_tokens": 27372204.0,
      "step": 13510
    },
    {
      "entropy": 5.899990224838257,
      "epoch": 1.181071397360832,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004866182578255094,
      "loss": 5.4269,
      "mean_token_accuracy": 0.18688846230506898,
      "num_tokens": 27382963.0,
      "step": 13515
    },
    {
      "entropy": 5.863694334030152,
      "epoch": 1.1815083457135367,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048660766965908736,
      "loss": 5.3935,
      "mean_token_accuracy": 0.1876440167427063,
      "num_tokens": 27392426.0,
      "step": 13520
    },
    {
      "entropy": 5.819086217880249,
      "epoch": 1.1819452940662414,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004865970774339194,
      "loss": 5.3731,
      "mean_token_accuracy": 0.18998550176620482,
      "num_tokens": 27402679.0,
      "step": 13525
    },
    {
      "entropy": 5.904140043258667,
      "epoch": 1.182382242418946,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048658648115020855,
      "loss": 5.4422,
      "mean_token_accuracy": 0.18709316700696946,
      "num_tokens": 27412911.0,
      "step": 13530
    },
    {
      "entropy": 5.915017890930176,
      "epoch": 1.1828191907716508,
      "grad_norm": 1.0,
      "learning_rate": 0.0004865758808081581,
      "loss": 5.3783,
      "mean_token_accuracy": 0.18593098521232604,
      "num_tokens": 27423394.0,
      "step": 13535
    },
    {
      "entropy": 5.854767942428589,
      "epoch": 1.1832561391243555,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048656527640797144,
      "loss": 5.35,
      "mean_token_accuracy": 0.19722918421030045,
      "num_tokens": 27433016.0,
      "step": 13540
    },
    {
      "entropy": 5.896092462539673,
      "epoch": 1.1836930874770601,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048655466794985194,
      "loss": 5.3954,
      "mean_token_accuracy": 0.18738422244787217,
      "num_tokens": 27442414.0,
      "step": 13545
    },
    {
      "entropy": 5.8387950420379635,
      "epoch": 1.1841300358297648,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000486544055434003,
      "loss": 5.447,
      "mean_token_accuracy": 0.18806910365819932,
      "num_tokens": 27452410.0,
      "step": 13550
    },
    {
      "entropy": 5.897030878067016,
      "epoch": 1.1845669841824695,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004865334388606282,
      "loss": 5.3657,
      "mean_token_accuracy": 0.20044133961200714,
      "num_tokens": 27462196.0,
      "step": 13555
    },
    {
      "entropy": 5.865886211395264,
      "epoch": 1.1850039325351744,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048652281822993127,
      "loss": 5.4861,
      "mean_token_accuracy": 0.18228364884853362,
      "num_tokens": 27473934.0,
      "step": 13560
    },
    {
      "entropy": 5.938606834411621,
      "epoch": 1.1854408808878791,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048651219354211585,
      "loss": 5.5006,
      "mean_token_accuracy": 0.1831474244594574,
      "num_tokens": 27484218.0,
      "step": 13565
    },
    {
      "entropy": 5.87084732055664,
      "epoch": 1.1858778292405838,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048650156479738575,
      "loss": 5.4746,
      "mean_token_accuracy": 0.1874558672308922,
      "num_tokens": 27494663.0,
      "step": 13570
    },
    {
      "entropy": 5.878785848617554,
      "epoch": 1.1863147775932885,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048649093199594473,
      "loss": 5.3644,
      "mean_token_accuracy": 0.18964941054582596,
      "num_tokens": 27504233.0,
      "step": 13575
    },
    {
      "entropy": 5.88812484741211,
      "epoch": 1.1867517259459932,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004864802951379969,
      "loss": 5.4001,
      "mean_token_accuracy": 0.19048290103673934,
      "num_tokens": 27514032.0,
      "step": 13580
    },
    {
      "entropy": 5.877618360519409,
      "epoch": 1.187188674298698,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048646965422374615,
      "loss": 5.4725,
      "mean_token_accuracy": 0.1867071956396103,
      "num_tokens": 27524659.0,
      "step": 13585
    },
    {
      "entropy": 5.875240850448608,
      "epoch": 1.1876256226514026,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004864590092533966,
      "loss": 5.3647,
      "mean_token_accuracy": 0.18989177495241166,
      "num_tokens": 27534003.0,
      "step": 13590
    },
    {
      "entropy": 5.863599729537964,
      "epoch": 1.1880625710041073,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004864483602271524,
      "loss": 5.3715,
      "mean_token_accuracy": 0.19008664637804032,
      "num_tokens": 27543586.0,
      "step": 13595
    },
    {
      "entropy": 5.8210700988769535,
      "epoch": 1.188499519356812,
      "grad_norm": 1.03125,
      "learning_rate": 0.000486437707145218,
      "loss": 5.3676,
      "mean_token_accuracy": 0.19515971690416337,
      "num_tokens": 27554210.0,
      "step": 13600
    },
    {
      "entropy": 5.913113212585449,
      "epoch": 1.1889364677095167,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048642705000779754,
      "loss": 5.4782,
      "mean_token_accuracy": 0.18910505026578903,
      "num_tokens": 27565324.0,
      "step": 13605
    },
    {
      "entropy": 5.888992166519165,
      "epoch": 1.1893734160622214,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004864163888150955,
      "loss": 5.3907,
      "mean_token_accuracy": 0.18686602562665938,
      "num_tokens": 27575192.0,
      "step": 13610
    },
    {
      "entropy": 5.853313875198364,
      "epoch": 1.1898103644149263,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048640572356731623,
      "loss": 5.3884,
      "mean_token_accuracy": 0.18516866713762284,
      "num_tokens": 27585294.0,
      "step": 13615
    },
    {
      "entropy": 5.874182796478271,
      "epoch": 1.190247312767631,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048639505426466447,
      "loss": 5.4056,
      "mean_token_accuracy": 0.1917635604739189,
      "num_tokens": 27595667.0,
      "step": 13620
    },
    {
      "entropy": 5.888009357452392,
      "epoch": 1.1906842611203357,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004863843809073448,
      "loss": 5.387,
      "mean_token_accuracy": 0.19407601207494735,
      "num_tokens": 27605816.0,
      "step": 13625
    },
    {
      "entropy": 5.880736017227173,
      "epoch": 1.1911212094730403,
      "grad_norm": 0.96875,
      "learning_rate": 0.000486373703495562,
      "loss": 5.4175,
      "mean_token_accuracy": 0.18775137662887573,
      "num_tokens": 27614874.0,
      "step": 13630
    },
    {
      "entropy": 5.8495323181152346,
      "epoch": 1.191558157825745,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004863630220295207,
      "loss": 5.3669,
      "mean_token_accuracy": 0.18477244526147843,
      "num_tokens": 27625610.0,
      "step": 13635
    },
    {
      "entropy": 5.89583830833435,
      "epoch": 1.1919951061784497,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048635233650942596,
      "loss": 5.4176,
      "mean_token_accuracy": 0.1877331629395485,
      "num_tokens": 27636646.0,
      "step": 13640
    },
    {
      "entropy": 5.921745443344117,
      "epoch": 1.1924320545311544,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048634164693548267,
      "loss": 5.416,
      "mean_token_accuracy": 0.1905466288328171,
      "num_tokens": 27647311.0,
      "step": 13645
    },
    {
      "entropy": 5.8618817806243895,
      "epoch": 1.192869002883859,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004863309533078957,
      "loss": 5.4413,
      "mean_token_accuracy": 0.18529793471097947,
      "num_tokens": 27658347.0,
      "step": 13650
    },
    {
      "entropy": 5.814700269699097,
      "epoch": 1.1933059512365638,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048632025562687047,
      "loss": 5.3873,
      "mean_token_accuracy": 0.19093874394893645,
      "num_tokens": 27669211.0,
      "step": 13655
    },
    {
      "entropy": 5.864947557449341,
      "epoch": 1.1937428995892685,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048630955389261193,
      "loss": 5.3885,
      "mean_token_accuracy": 0.18695630580186845,
      "num_tokens": 27678904.0,
      "step": 13660
    },
    {
      "entropy": 5.847626399993897,
      "epoch": 1.1941798479419732,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048629884810532547,
      "loss": 5.3578,
      "mean_token_accuracy": 0.1938928782939911,
      "num_tokens": 27688928.0,
      "step": 13665
    },
    {
      "entropy": 5.866127967834473,
      "epoch": 1.1946167962946779,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004862881382652164,
      "loss": 5.4486,
      "mean_token_accuracy": 0.18597887456417084,
      "num_tokens": 27699855.0,
      "step": 13670
    },
    {
      "entropy": 5.895108652114868,
      "epoch": 1.1950537446473826,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004862774243724901,
      "loss": 5.4084,
      "mean_token_accuracy": 0.18887837827205659,
      "num_tokens": 27709857.0,
      "step": 13675
    },
    {
      "entropy": 5.91774959564209,
      "epoch": 1.1954906930000875,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004862667064273521,
      "loss": 5.4439,
      "mean_token_accuracy": 0.18849332481622696,
      "num_tokens": 27719612.0,
      "step": 13680
    },
    {
      "entropy": 5.828707027435303,
      "epoch": 1.1959276413527922,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004862559844300079,
      "loss": 5.4132,
      "mean_token_accuracy": 0.1903348073363304,
      "num_tokens": 27730052.0,
      "step": 13685
    },
    {
      "entropy": 5.82106122970581,
      "epoch": 1.1963645897054969,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004862452583806633,
      "loss": 5.3141,
      "mean_token_accuracy": 0.20199884474277496,
      "num_tokens": 27739980.0,
      "step": 13690
    },
    {
      "entropy": 5.845099353790284,
      "epoch": 1.1968015380582016,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048623452827952385,
      "loss": 5.416,
      "mean_token_accuracy": 0.18601446151733397,
      "num_tokens": 27748863.0,
      "step": 13695
    },
    {
      "entropy": 5.807712507247925,
      "epoch": 1.1972384864109062,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004862237941267956,
      "loss": 5.3734,
      "mean_token_accuracy": 0.19702224731445311,
      "num_tokens": 27758672.0,
      "step": 13700
    },
    {
      "entropy": 5.87979564666748,
      "epoch": 1.197675434763611,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004862130559226842,
      "loss": 5.4115,
      "mean_token_accuracy": 0.18855781853199005,
      "num_tokens": 27768195.0,
      "step": 13705
    },
    {
      "entropy": 5.807479190826416,
      "epoch": 1.1981123831163156,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004862023136673958,
      "loss": 5.3502,
      "mean_token_accuracy": 0.195234215259552,
      "num_tokens": 27778135.0,
      "step": 13710
    },
    {
      "entropy": 5.933418607711792,
      "epoch": 1.1985493314690203,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048619156736113634,
      "loss": 5.4383,
      "mean_token_accuracy": 0.17863466292619706,
      "num_tokens": 27788723.0,
      "step": 13715
    },
    {
      "entropy": 5.858183574676514,
      "epoch": 1.198986279821725,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004861808170041119,
      "loss": 5.4391,
      "mean_token_accuracy": 0.18821690380573272,
      "num_tokens": 27799060.0,
      "step": 13720
    },
    {
      "entropy": 5.900593662261963,
      "epoch": 1.1994232281744297,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004861700625965287,
      "loss": 5.4408,
      "mean_token_accuracy": 0.18381839692592622,
      "num_tokens": 27809542.0,
      "step": 13725
    },
    {
      "entropy": 5.862662982940674,
      "epoch": 1.1998601765271344,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048615930413859317,
      "loss": 5.4132,
      "mean_token_accuracy": 0.18872570097446442,
      "num_tokens": 27819738.0,
      "step": 13730
    },
    {
      "entropy": 5.887794399261475,
      "epoch": 1.2002971248798393,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048614854163051143,
      "loss": 5.3823,
      "mean_token_accuracy": 0.18860144913196564,
      "num_tokens": 27829052.0,
      "step": 13735
    },
    {
      "entropy": 5.8577581405639645,
      "epoch": 1.200734073232544,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004861377750724901,
      "loss": 5.3685,
      "mean_token_accuracy": 0.19019396603107452,
      "num_tokens": 27839277.0,
      "step": 13740
    },
    {
      "entropy": 5.907029914855957,
      "epoch": 1.2011710215852487,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004861270044647356,
      "loss": 5.4196,
      "mean_token_accuracy": 0.1884370595216751,
      "num_tokens": 27848550.0,
      "step": 13745
    },
    {
      "entropy": 5.884589290618896,
      "epoch": 1.2016079699379534,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048611622980745446,
      "loss": 5.4647,
      "mean_token_accuracy": 0.1726691484451294,
      "num_tokens": 27858546.0,
      "step": 13750
    },
    {
      "entropy": 5.86145429611206,
      "epoch": 1.202044918290658,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048610545110085347,
      "loss": 5.472,
      "mean_token_accuracy": 0.1789645344018936,
      "num_tokens": 27869192.0,
      "step": 13755
    },
    {
      "entropy": 5.910765266418457,
      "epoch": 1.2024818666433628,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004860946683451393,
      "loss": 5.4564,
      "mean_token_accuracy": 0.18831523656845092,
      "num_tokens": 27878534.0,
      "step": 13760
    },
    {
      "entropy": 5.87614426612854,
      "epoch": 1.2029188149960675,
      "grad_norm": 1.0,
      "learning_rate": 0.00048608388154051875,
      "loss": 5.3538,
      "mean_token_accuracy": 0.1941593199968338,
      "num_tokens": 27888074.0,
      "step": 13765
    },
    {
      "entropy": 5.807686471939087,
      "epoch": 1.2033557633487721,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048607309068719875,
      "loss": 5.4434,
      "mean_token_accuracy": 0.18267462849617006,
      "num_tokens": 27898210.0,
      "step": 13770
    },
    {
      "entropy": 5.835697507858276,
      "epoch": 1.2037927117014768,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004860622957853863,
      "loss": 5.3531,
      "mean_token_accuracy": 0.19521314054727554,
      "num_tokens": 27907472.0,
      "step": 13775
    },
    {
      "entropy": 5.867703437805176,
      "epoch": 1.2042296600541815,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004860514968352885,
      "loss": 5.4327,
      "mean_token_accuracy": 0.1894722580909729,
      "num_tokens": 27917265.0,
      "step": 13780
    },
    {
      "entropy": 5.880539751052856,
      "epoch": 1.2046666084068862,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048604069383711225,
      "loss": 5.3447,
      "mean_token_accuracy": 0.20009371936321257,
      "num_tokens": 27926883.0,
      "step": 13785
    },
    {
      "entropy": 5.9209146976470945,
      "epoch": 1.2051035567595911,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048602988679106497,
      "loss": 5.3797,
      "mean_token_accuracy": 0.1856116011738777,
      "num_tokens": 27937167.0,
      "step": 13790
    },
    {
      "entropy": 5.845889091491699,
      "epoch": 1.2055405051122958,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004860190756973539,
      "loss": 5.3555,
      "mean_token_accuracy": 0.19580881744623185,
      "num_tokens": 27947422.0,
      "step": 13795
    },
    {
      "entropy": 5.855303859710693,
      "epoch": 1.2059774534650005,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048600826055618634,
      "loss": 5.4472,
      "mean_token_accuracy": 0.1834481790661812,
      "num_tokens": 27958407.0,
      "step": 13800
    },
    {
      "entropy": 5.8502098560333256,
      "epoch": 1.2064144018177052,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048599744136776986,
      "loss": 5.2759,
      "mean_token_accuracy": 0.19723487198352813,
      "num_tokens": 27968277.0,
      "step": 13805
    },
    {
      "entropy": 5.876499795913697,
      "epoch": 1.20685135017041,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048598661813231177,
      "loss": 5.3334,
      "mean_token_accuracy": 0.19343792349100114,
      "num_tokens": 27978116.0,
      "step": 13810
    },
    {
      "entropy": 5.84489631652832,
      "epoch": 1.2072882985231146,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048597579085001984,
      "loss": 5.3421,
      "mean_token_accuracy": 0.18769187182188035,
      "num_tokens": 27988383.0,
      "step": 13815
    },
    {
      "entropy": 5.7975836277008055,
      "epoch": 1.2077252468758193,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048596495952110176,
      "loss": 5.3426,
      "mean_token_accuracy": 0.1990964025259018,
      "num_tokens": 27999118.0,
      "step": 13820
    },
    {
      "entropy": 5.851860523223877,
      "epoch": 1.208162195228524,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004859541241457651,
      "loss": 5.3458,
      "mean_token_accuracy": 0.19172056764364243,
      "num_tokens": 28008849.0,
      "step": 13825
    },
    {
      "entropy": 5.868815183639526,
      "epoch": 1.2085991435812287,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004859432847242179,
      "loss": 5.3959,
      "mean_token_accuracy": 0.19181374460458755,
      "num_tokens": 28019063.0,
      "step": 13830
    },
    {
      "entropy": 5.832690668106079,
      "epoch": 1.2090360919339334,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048593244125666793,
      "loss": 5.3086,
      "mean_token_accuracy": 0.19337739050388336,
      "num_tokens": 28028333.0,
      "step": 13835
    },
    {
      "entropy": 5.846954774856568,
      "epoch": 1.209473040286638,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004859215937433232,
      "loss": 5.3861,
      "mean_token_accuracy": 0.19646167606115342,
      "num_tokens": 28037704.0,
      "step": 13840
    },
    {
      "entropy": 5.831296730041504,
      "epoch": 1.2099099886393427,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004859107421843918,
      "loss": 5.3878,
      "mean_token_accuracy": 0.19176595658063889,
      "num_tokens": 28047157.0,
      "step": 13845
    },
    {
      "entropy": 5.910428714752197,
      "epoch": 1.2103469369920474,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048589988658008184,
      "loss": 5.3454,
      "mean_token_accuracy": 0.18757526725530624,
      "num_tokens": 28057582.0,
      "step": 13850
    },
    {
      "entropy": 5.837192821502685,
      "epoch": 1.2107838853447523,
      "grad_norm": 1.0,
      "learning_rate": 0.00048588902693060156,
      "loss": 5.4409,
      "mean_token_accuracy": 0.18069027364253998,
      "num_tokens": 28067458.0,
      "step": 13855
    },
    {
      "entropy": 5.869321298599243,
      "epoch": 1.211220833697457,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004858781632361592,
      "loss": 5.3911,
      "mean_token_accuracy": 0.1876963570713997,
      "num_tokens": 28078967.0,
      "step": 13860
    },
    {
      "entropy": 5.851922225952149,
      "epoch": 1.2116577820501617,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004858672954969632,
      "loss": 5.3595,
      "mean_token_accuracy": 0.1887788236141205,
      "num_tokens": 28088498.0,
      "step": 13865
    },
    {
      "entropy": 5.830109882354736,
      "epoch": 1.2120947304028664,
      "grad_norm": 1.125,
      "learning_rate": 0.00048585642371322196,
      "loss": 5.4489,
      "mean_token_accuracy": 0.18831094354391098,
      "num_tokens": 28098884.0,
      "step": 13870
    },
    {
      "entropy": 5.837340545654297,
      "epoch": 1.212531678755571,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004858455478851441,
      "loss": 5.3643,
      "mean_token_accuracy": 0.19381032437086104,
      "num_tokens": 28109629.0,
      "step": 13875
    },
    {
      "entropy": 5.878021097183227,
      "epoch": 1.2129686271082758,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000485834668012938,
      "loss": 5.3137,
      "mean_token_accuracy": 0.19124649912118913,
      "num_tokens": 28118555.0,
      "step": 13880
    },
    {
      "entropy": 5.912336349487305,
      "epoch": 1.2134055754609805,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004858237840968127,
      "loss": 5.4632,
      "mean_token_accuracy": 0.18103642165660858,
      "num_tokens": 28129028.0,
      "step": 13885
    },
    {
      "entropy": 5.818202686309815,
      "epoch": 1.2138425238136852,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048581289613697654,
      "loss": 5.3893,
      "mean_token_accuracy": 0.1881835028529167,
      "num_tokens": 28139188.0,
      "step": 13890
    },
    {
      "entropy": 5.838582277297974,
      "epoch": 1.2142794721663899,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048580200413363866,
      "loss": 5.3492,
      "mean_token_accuracy": 0.19706420004367828,
      "num_tokens": 28148199.0,
      "step": 13895
    },
    {
      "entropy": 5.846554470062256,
      "epoch": 1.2147164205190946,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004857911080870079,
      "loss": 5.3124,
      "mean_token_accuracy": 0.19376841485500335,
      "num_tokens": 28157811.0,
      "step": 13900
    },
    {
      "entropy": 5.789066457748413,
      "epoch": 1.2151533688717993,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048578020799729327,
      "loss": 5.3394,
      "mean_token_accuracy": 0.20024704486131667,
      "num_tokens": 28168644.0,
      "step": 13905
    },
    {
      "entropy": 5.823336839675903,
      "epoch": 1.2155903172245042,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004857693038647037,
      "loss": 5.4232,
      "mean_token_accuracy": 0.188800211250782,
      "num_tokens": 28179846.0,
      "step": 13910
    },
    {
      "entropy": 5.8480652332305905,
      "epoch": 1.2160272655772089,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004857583956894485,
      "loss": 5.3563,
      "mean_token_accuracy": 0.19831409454345703,
      "num_tokens": 28189400.0,
      "step": 13915
    },
    {
      "entropy": 5.889528560638428,
      "epoch": 1.2164642139299136,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004857474834717368,
      "loss": 5.433,
      "mean_token_accuracy": 0.18608075082302095,
      "num_tokens": 28199695.0,
      "step": 13920
    },
    {
      "entropy": 5.813863325119018,
      "epoch": 1.2169011622826182,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048573656721177805,
      "loss": 5.3452,
      "mean_token_accuracy": 0.1947764828801155,
      "num_tokens": 28209977.0,
      "step": 13925
    },
    {
      "entropy": 5.900140476226807,
      "epoch": 1.217338110635323,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004857256469097814,
      "loss": 5.4191,
      "mean_token_accuracy": 0.18630109280347823,
      "num_tokens": 28220619.0,
      "step": 13930
    },
    {
      "entropy": 5.877826404571533,
      "epoch": 1.2177750589880276,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004857147225659564,
      "loss": 5.381,
      "mean_token_accuracy": 0.1918457642197609,
      "num_tokens": 28230065.0,
      "step": 13935
    },
    {
      "entropy": 5.847424697875977,
      "epoch": 1.2182120073407323,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048570379418051273,
      "loss": 5.3457,
      "mean_token_accuracy": 0.19819722771644593,
      "num_tokens": 28239307.0,
      "step": 13940
    },
    {
      "entropy": 5.918801307678223,
      "epoch": 1.218648955693437,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004856928617536598,
      "loss": 5.3613,
      "mean_token_accuracy": 0.18639079928398133,
      "num_tokens": 28249421.0,
      "step": 13945
    },
    {
      "entropy": 5.815378284454345,
      "epoch": 1.2190859040461417,
      "grad_norm": 1.0,
      "learning_rate": 0.0004856819252856073,
      "loss": 5.3484,
      "mean_token_accuracy": 0.18342800736427306,
      "num_tokens": 28258125.0,
      "step": 13950
    },
    {
      "entropy": 5.814541387557983,
      "epoch": 1.2195228523988464,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048567098477656526,
      "loss": 5.3461,
      "mean_token_accuracy": 0.1894058719277382,
      "num_tokens": 28268897.0,
      "step": 13955
    },
    {
      "entropy": 5.864326667785645,
      "epoch": 1.219959800751551,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004856600402267432,
      "loss": 5.3648,
      "mean_token_accuracy": 0.19623979777097703,
      "num_tokens": 28278506.0,
      "step": 13960
    },
    {
      "entropy": 5.872466087341309,
      "epoch": 1.220396749104256,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004856490916363513,
      "loss": 5.3282,
      "mean_token_accuracy": 0.19490687400102616,
      "num_tokens": 28287905.0,
      "step": 13965
    },
    {
      "entropy": 5.909127902984619,
      "epoch": 1.2208336974569605,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004856381390055994,
      "loss": 5.4535,
      "mean_token_accuracy": 0.18648293167352675,
      "num_tokens": 28297123.0,
      "step": 13970
    },
    {
      "entropy": 5.8877698421478275,
      "epoch": 1.2212706458096654,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048562718233469756,
      "loss": 5.4655,
      "mean_token_accuracy": 0.1792059287428856,
      "num_tokens": 28306621.0,
      "step": 13975
    },
    {
      "entropy": 5.927615833282471,
      "epoch": 1.22170759416237,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048561622162385606,
      "loss": 5.5095,
      "mean_token_accuracy": 0.17759799361228942,
      "num_tokens": 28318369.0,
      "step": 13980
    },
    {
      "entropy": 5.917509698867798,
      "epoch": 1.2221445425150748,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048560525687328497,
      "loss": 5.4355,
      "mean_token_accuracy": 0.18581180572509765,
      "num_tokens": 28328817.0,
      "step": 13985
    },
    {
      "entropy": 5.872539424896241,
      "epoch": 1.2225814908677795,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048559428808319483,
      "loss": 5.4765,
      "mean_token_accuracy": 0.1796051561832428,
      "num_tokens": 28339448.0,
      "step": 13990
    },
    {
      "entropy": 5.87861909866333,
      "epoch": 1.2230184392204841,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004855833152537958,
      "loss": 5.355,
      "mean_token_accuracy": 0.1891946792602539,
      "num_tokens": 28349243.0,
      "step": 13995
    },
    {
      "entropy": 5.87134747505188,
      "epoch": 1.2234553875731888,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004855723383852985,
      "loss": 5.3919,
      "mean_token_accuracy": 0.19148917645215988,
      "num_tokens": 28359392.0,
      "step": 14000
    },
    {
      "entropy": 5.874764728546142,
      "epoch": 1.2238923359258935,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004855613574779134,
      "loss": 5.3596,
      "mean_token_accuracy": 0.1925269439816475,
      "num_tokens": 28369160.0,
      "step": 14005
    },
    {
      "entropy": 5.854053211212158,
      "epoch": 1.2243292842785982,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048555037253185114,
      "loss": 5.434,
      "mean_token_accuracy": 0.1923088788986206,
      "num_tokens": 28379107.0,
      "step": 14010
    },
    {
      "entropy": 5.894358110427857,
      "epoch": 1.224766232631303,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048553938354732243,
      "loss": 5.404,
      "mean_token_accuracy": 0.18984090983867646,
      "num_tokens": 28388956.0,
      "step": 14015
    },
    {
      "entropy": 5.8404230117797855,
      "epoch": 1.2252031809840076,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048552839052453794,
      "loss": 5.38,
      "mean_token_accuracy": 0.18811680525541305,
      "num_tokens": 28398925.0,
      "step": 14020
    },
    {
      "entropy": 5.9403198719024655,
      "epoch": 1.2256401293367123,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048551739346370864,
      "loss": 5.4753,
      "mean_token_accuracy": 0.17910763174295424,
      "num_tokens": 28408714.0,
      "step": 14025
    },
    {
      "entropy": 5.975804471969605,
      "epoch": 1.2260770776894172,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004855063923650455,
      "loss": 5.4567,
      "mean_token_accuracy": 0.1847511723637581,
      "num_tokens": 28419408.0,
      "step": 14030
    },
    {
      "entropy": 5.860580110549927,
      "epoch": 1.226514026042122,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004854953872287593,
      "loss": 5.3868,
      "mean_token_accuracy": 0.18835727125406265,
      "num_tokens": 28430044.0,
      "step": 14035
    },
    {
      "entropy": 5.82306318283081,
      "epoch": 1.2269509743948266,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004854843780550614,
      "loss": 5.3336,
      "mean_token_accuracy": 0.18961650282144546,
      "num_tokens": 28440100.0,
      "step": 14040
    },
    {
      "entropy": 5.8436408042907715,
      "epoch": 1.2273879227475313,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004854733648441628,
      "loss": 5.3389,
      "mean_token_accuracy": 0.1946134716272354,
      "num_tokens": 28449497.0,
      "step": 14045
    },
    {
      "entropy": 5.859148454666138,
      "epoch": 1.227824871100236,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048546234759627477,
      "loss": 5.4003,
      "mean_token_accuracy": 0.19821036756038665,
      "num_tokens": 28459789.0,
      "step": 14050
    },
    {
      "entropy": 5.886418294906616,
      "epoch": 1.2282618194529407,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048545132631160856,
      "loss": 5.4667,
      "mean_token_accuracy": 0.1782274767756462,
      "num_tokens": 28471728.0,
      "step": 14055
    },
    {
      "entropy": 5.869079065322876,
      "epoch": 1.2286987678056454,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048544030099037576,
      "loss": 5.4242,
      "mean_token_accuracy": 0.19285250306129456,
      "num_tokens": 28481876.0,
      "step": 14060
    },
    {
      "entropy": 5.907915925979614,
      "epoch": 1.22913571615835,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048542927163278766,
      "loss": 5.3834,
      "mean_token_accuracy": 0.19241491407155992,
      "num_tokens": 28491800.0,
      "step": 14065
    },
    {
      "entropy": 5.836249971389771,
      "epoch": 1.2295726645110547,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004854182382390559,
      "loss": 5.3473,
      "mean_token_accuracy": 0.19086467921733857,
      "num_tokens": 28502033.0,
      "step": 14070
    },
    {
      "entropy": 5.88521180152893,
      "epoch": 1.2300096128637594,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000485407200809392,
      "loss": 5.4182,
      "mean_token_accuracy": 0.18379060477018355,
      "num_tokens": 28511921.0,
      "step": 14075
    },
    {
      "entropy": 5.9039360046386715,
      "epoch": 1.2304465612164641,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004853961593440078,
      "loss": 5.3523,
      "mean_token_accuracy": 0.19128271341323852,
      "num_tokens": 28521449.0,
      "step": 14080
    },
    {
      "entropy": 5.831187915802002,
      "epoch": 1.230883509569169,
      "grad_norm": 1.015625,
      "learning_rate": 0.000485385113843115,
      "loss": 5.3556,
      "mean_token_accuracy": 0.19393130391836166,
      "num_tokens": 28532088.0,
      "step": 14085
    },
    {
      "entropy": 5.883874130249024,
      "epoch": 1.2313204579218737,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004853740643069255,
      "loss": 5.4324,
      "mean_token_accuracy": 0.19240793883800505,
      "num_tokens": 28541835.0,
      "step": 14090
    },
    {
      "entropy": 5.880634784698486,
      "epoch": 1.2317574062745784,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048536301073565113,
      "loss": 5.4048,
      "mean_token_accuracy": 0.18868185877799987,
      "num_tokens": 28551601.0,
      "step": 14095
    },
    {
      "entropy": 5.837529563903809,
      "epoch": 1.232194354627283,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004853519531295041,
      "loss": 5.3528,
      "mean_token_accuracy": 0.19847090244293214,
      "num_tokens": 28561012.0,
      "step": 14100
    },
    {
      "entropy": 5.879119348526001,
      "epoch": 1.2326313029799878,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004853408914886963,
      "loss": 5.4474,
      "mean_token_accuracy": 0.18829695582389833,
      "num_tokens": 28571431.0,
      "step": 14105
    },
    {
      "entropy": 5.810194969177246,
      "epoch": 1.2330682513326925,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048532982581344004,
      "loss": 5.3552,
      "mean_token_accuracy": 0.20122067779302596,
      "num_tokens": 28581185.0,
      "step": 14110
    },
    {
      "entropy": 5.800151252746582,
      "epoch": 1.2335051996853972,
      "grad_norm": 1.125,
      "learning_rate": 0.0004853187561039475,
      "loss": 5.3769,
      "mean_token_accuracy": 0.19630796760320662,
      "num_tokens": 28591164.0,
      "step": 14115
    },
    {
      "entropy": 5.812721252441406,
      "epoch": 1.2339421480381019,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048530768236043094,
      "loss": 5.3025,
      "mean_token_accuracy": 0.199785515666008,
      "num_tokens": 28600931.0,
      "step": 14120
    },
    {
      "entropy": 5.875689458847046,
      "epoch": 1.2343790963908066,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004852966045831029,
      "loss": 5.4257,
      "mean_token_accuracy": 0.19062697589397432,
      "num_tokens": 28611908.0,
      "step": 14125
    },
    {
      "entropy": 5.89387092590332,
      "epoch": 1.2348160447435113,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004852855227721758,
      "loss": 5.4427,
      "mean_token_accuracy": 0.187044258415699,
      "num_tokens": 28622106.0,
      "step": 14130
    },
    {
      "entropy": 5.932678127288819,
      "epoch": 1.235252993096216,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048527443692786217,
      "loss": 5.4639,
      "mean_token_accuracy": 0.18018227815628052,
      "num_tokens": 28632634.0,
      "step": 14135
    },
    {
      "entropy": 5.894137811660767,
      "epoch": 1.2356899414489209,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004852633470503746,
      "loss": 5.4115,
      "mean_token_accuracy": 0.18734831362962723,
      "num_tokens": 28642936.0,
      "step": 14140
    },
    {
      "entropy": 5.880568313598633,
      "epoch": 1.2361268898016253,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048525225313992594,
      "loss": 5.3751,
      "mean_token_accuracy": 0.18614655435085298,
      "num_tokens": 28653539.0,
      "step": 14145
    },
    {
      "entropy": 5.86704797744751,
      "epoch": 1.2365638381543302,
      "grad_norm": 1.0,
      "learning_rate": 0.0004852411551967289,
      "loss": 5.4298,
      "mean_token_accuracy": 0.18522239476442337,
      "num_tokens": 28664212.0,
      "step": 14150
    },
    {
      "entropy": 5.971092081069946,
      "epoch": 1.237000786507035,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048523005322099626,
      "loss": 5.3868,
      "mean_token_accuracy": 0.19027685523033142,
      "num_tokens": 28676028.0,
      "step": 14155
    },
    {
      "entropy": 5.926460075378418,
      "epoch": 1.2374377348597396,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048521894721294115,
      "loss": 5.3606,
      "mean_token_accuracy": 0.18804315775632857,
      "num_tokens": 28685370.0,
      "step": 14160
    },
    {
      "entropy": 5.805957460403443,
      "epoch": 1.2378746832124443,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048520783717277645,
      "loss": 5.3389,
      "mean_token_accuracy": 0.19799513518810272,
      "num_tokens": 28695300.0,
      "step": 14165
    },
    {
      "entropy": 5.842790508270264,
      "epoch": 1.238311631565149,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048519672310071523,
      "loss": 5.376,
      "mean_token_accuracy": 0.19304944276809693,
      "num_tokens": 28705485.0,
      "step": 14170
    },
    {
      "entropy": 5.889214372634887,
      "epoch": 1.2387485799178537,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048518560499697066,
      "loss": 5.4001,
      "mean_token_accuracy": 0.19088010787963866,
      "num_tokens": 28715407.0,
      "step": 14175
    },
    {
      "entropy": 5.909910058975219,
      "epoch": 1.2391855282705584,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048517448286175617,
      "loss": 5.4285,
      "mean_token_accuracy": 0.187852181494236,
      "num_tokens": 28725283.0,
      "step": 14180
    },
    {
      "entropy": 5.819094610214234,
      "epoch": 1.239622476623263,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048516335669528495,
      "loss": 5.3393,
      "mean_token_accuracy": 0.19574202597141266,
      "num_tokens": 28736131.0,
      "step": 14185
    },
    {
      "entropy": 5.860550165176392,
      "epoch": 1.2400594249759678,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004851522264977704,
      "loss": 5.3509,
      "mean_token_accuracy": 0.19361283332109452,
      "num_tokens": 28745446.0,
      "step": 14190
    },
    {
      "entropy": 5.845378637313843,
      "epoch": 1.2404963733286725,
      "grad_norm": 1.046875,
      "learning_rate": 0.000485141092269426,
      "loss": 5.315,
      "mean_token_accuracy": 0.19165366739034653,
      "num_tokens": 28754639.0,
      "step": 14195
    },
    {
      "entropy": 5.859380912780762,
      "epoch": 1.2409333216813772,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004851299540104654,
      "loss": 5.383,
      "mean_token_accuracy": 0.1913836658000946,
      "num_tokens": 28764397.0,
      "step": 14200
    },
    {
      "entropy": 5.844194221496582,
      "epoch": 1.241370270034082,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048511881172110216,
      "loss": 5.3653,
      "mean_token_accuracy": 0.19461780041456223,
      "num_tokens": 28773680.0,
      "step": 14205
    },
    {
      "entropy": 5.901930665969848,
      "epoch": 1.2418072183867868,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048510766540154997,
      "loss": 5.4401,
      "mean_token_accuracy": 0.18724251240491868,
      "num_tokens": 28784034.0,
      "step": 14210
    },
    {
      "entropy": 5.872540378570557,
      "epoch": 1.2422441667394915,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048509651505202275,
      "loss": 5.4387,
      "mean_token_accuracy": 0.18674774169921876,
      "num_tokens": 28794487.0,
      "step": 14215
    },
    {
      "entropy": 5.930041837692261,
      "epoch": 1.2426811150921961,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004850853606727342,
      "loss": 5.3327,
      "mean_token_accuracy": 0.19405905455350875,
      "num_tokens": 28803942.0,
      "step": 14220
    },
    {
      "entropy": 5.88749098777771,
      "epoch": 1.2431180634449008,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004850742022638984,
      "loss": 5.3638,
      "mean_token_accuracy": 0.19473775774240493,
      "num_tokens": 28814498.0,
      "step": 14225
    },
    {
      "entropy": 5.790334320068359,
      "epoch": 1.2435550117976055,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048506303982572924,
      "loss": 5.3472,
      "mean_token_accuracy": 0.18966357558965682,
      "num_tokens": 28824513.0,
      "step": 14230
    },
    {
      "entropy": 5.89182562828064,
      "epoch": 1.2439919601503102,
      "grad_norm": 1.0,
      "learning_rate": 0.000485051873358441,
      "loss": 5.484,
      "mean_token_accuracy": 0.17860383838415145,
      "num_tokens": 28835643.0,
      "step": 14235
    },
    {
      "entropy": 5.985145616531372,
      "epoch": 1.244428908503015,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004850407028622477,
      "loss": 5.4967,
      "mean_token_accuracy": 0.1846096158027649,
      "num_tokens": 28845615.0,
      "step": 14240
    },
    {
      "entropy": 5.845870780944824,
      "epoch": 1.2448658568557196,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048502952833736365,
      "loss": 5.3722,
      "mean_token_accuracy": 0.19929196387529374,
      "num_tokens": 28856152.0,
      "step": 14245
    },
    {
      "entropy": 5.799424266815185,
      "epoch": 1.2453028052084243,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048501834978400323,
      "loss": 5.3169,
      "mean_token_accuracy": 0.19631199389696122,
      "num_tokens": 28866926.0,
      "step": 14250
    },
    {
      "entropy": 5.875818586349487,
      "epoch": 1.245739753561129,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004850071672023809,
      "loss": 5.4002,
      "mean_token_accuracy": 0.1848309561610222,
      "num_tokens": 28877188.0,
      "step": 14255
    },
    {
      "entropy": 5.886665868759155,
      "epoch": 1.246176701913834,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048499598059271087,
      "loss": 5.3851,
      "mean_token_accuracy": 0.1906994342803955,
      "num_tokens": 28887710.0,
      "step": 14260
    },
    {
      "entropy": 5.898268842697144,
      "epoch": 1.2466136502665386,
      "grad_norm": 1.015625,
      "learning_rate": 0.000484984789955208,
      "loss": 5.4198,
      "mean_token_accuracy": 0.18735795319080353,
      "num_tokens": 28897064.0,
      "step": 14265
    },
    {
      "entropy": 5.868411111831665,
      "epoch": 1.2470505986192433,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048497359529008684,
      "loss": 5.4121,
      "mean_token_accuracy": 0.1859988585114479,
      "num_tokens": 28907103.0,
      "step": 14270
    },
    {
      "entropy": 5.879853248596191,
      "epoch": 1.247487546971948,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004849623965975621,
      "loss": 5.3966,
      "mean_token_accuracy": 0.18619431406259537,
      "num_tokens": 28916851.0,
      "step": 14275
    },
    {
      "entropy": 5.897002077102661,
      "epoch": 1.2479244953246527,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004849511938778485,
      "loss": 5.4237,
      "mean_token_accuracy": 0.1887039750814438,
      "num_tokens": 28926638.0,
      "step": 14280
    },
    {
      "entropy": 5.9065605163574215,
      "epoch": 1.2483614436773574,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048493998713116103,
      "loss": 5.382,
      "mean_token_accuracy": 0.19154519885778426,
      "num_tokens": 28936764.0,
      "step": 14285
    },
    {
      "entropy": 5.8299318790435795,
      "epoch": 1.248798392030062,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048492877635771464,
      "loss": 5.394,
      "mean_token_accuracy": 0.18984332233667373,
      "num_tokens": 28946494.0,
      "step": 14290
    },
    {
      "entropy": 5.830911493301391,
      "epoch": 1.2492353403827667,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048491756155772433,
      "loss": 5.3621,
      "mean_token_accuracy": 0.19841751903295518,
      "num_tokens": 28955626.0,
      "step": 14295
    },
    {
      "entropy": 5.909454584121704,
      "epoch": 1.2496722887354714,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048490634273140517,
      "loss": 5.3744,
      "mean_token_accuracy": 0.18856683522462844,
      "num_tokens": 28966317.0,
      "step": 14300
    },
    {
      "entropy": 5.863029336929321,
      "epoch": 1.2501092370881761,
      "grad_norm": 1.125,
      "learning_rate": 0.00048489511987897236,
      "loss": 5.396,
      "mean_token_accuracy": 0.19374693632125856,
      "num_tokens": 28976063.0,
      "step": 14305
    },
    {
      "entropy": 5.884814977645874,
      "epoch": 1.2505461854408808,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004848838930006411,
      "loss": 5.377,
      "mean_token_accuracy": 0.19407933205366135,
      "num_tokens": 28985789.0,
      "step": 14310
    },
    {
      "entropy": 5.884112310409546,
      "epoch": 1.2509831337935857,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004848726620966269,
      "loss": 5.4677,
      "mean_token_accuracy": 0.188780015707016,
      "num_tokens": 28996458.0,
      "step": 14315
    },
    {
      "entropy": 5.877016973495484,
      "epoch": 1.2514200821462902,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000484861427167145,
      "loss": 5.3849,
      "mean_token_accuracy": 0.19079308360815048,
      "num_tokens": 29007641.0,
      "step": 14320
    },
    {
      "entropy": 5.931864070892334,
      "epoch": 1.251857030498995,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048485018821241104,
      "loss": 5.3841,
      "mean_token_accuracy": 0.18249916583299636,
      "num_tokens": 29017351.0,
      "step": 14325
    },
    {
      "entropy": 5.899076128005982,
      "epoch": 1.2522939788516998,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048483894523264045,
      "loss": 5.4298,
      "mean_token_accuracy": 0.18295432180166243,
      "num_tokens": 29026951.0,
      "step": 14330
    },
    {
      "entropy": 5.814954996109009,
      "epoch": 1.2527309272044045,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004848276982280489,
      "loss": 5.3893,
      "mean_token_accuracy": 0.19187174886465072,
      "num_tokens": 29037596.0,
      "step": 14335
    },
    {
      "entropy": 5.861359977722168,
      "epoch": 1.2531678755571092,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004848164471988522,
      "loss": 5.3863,
      "mean_token_accuracy": 0.18971292972564696,
      "num_tokens": 29049010.0,
      "step": 14340
    },
    {
      "entropy": 5.884432554244995,
      "epoch": 1.2536048239098139,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048480519214526616,
      "loss": 5.4184,
      "mean_token_accuracy": 0.18844424188137054,
      "num_tokens": 29059028.0,
      "step": 14345
    },
    {
      "entropy": 5.8452061176300045,
      "epoch": 1.2540417722625186,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048479393306750647,
      "loss": 5.3845,
      "mean_token_accuracy": 0.18955647349357604,
      "num_tokens": 29069661.0,
      "step": 14350
    },
    {
      "entropy": 5.874566650390625,
      "epoch": 1.2544787206152233,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004847826699657893,
      "loss": 5.3672,
      "mean_token_accuracy": 0.19522956311702727,
      "num_tokens": 29079971.0,
      "step": 14355
    },
    {
      "entropy": 5.87875657081604,
      "epoch": 1.254915668967928,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048477140284033054,
      "loss": 5.4273,
      "mean_token_accuracy": 0.1888917237520218,
      "num_tokens": 29091378.0,
      "step": 14360
    },
    {
      "entropy": 5.8304352283477785,
      "epoch": 1.2553526173206326,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004847601316913464,
      "loss": 5.367,
      "mean_token_accuracy": 0.1911866009235382,
      "num_tokens": 29101516.0,
      "step": 14365
    },
    {
      "entropy": 5.865463638305664,
      "epoch": 1.2557895656733375,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000484748856519053,
      "loss": 5.4451,
      "mean_token_accuracy": 0.1837229013442993,
      "num_tokens": 29112147.0,
      "step": 14370
    },
    {
      "entropy": 5.854351139068603,
      "epoch": 1.256226514026042,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004847375773236667,
      "loss": 5.3583,
      "mean_token_accuracy": 0.19456009268760682,
      "num_tokens": 29122084.0,
      "step": 14375
    },
    {
      "entropy": 5.949103832244873,
      "epoch": 1.256663462378747,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004847262941054037,
      "loss": 5.3944,
      "mean_token_accuracy": 0.19309551417827606,
      "num_tokens": 29131988.0,
      "step": 14380
    },
    {
      "entropy": 5.935528039932251,
      "epoch": 1.2571004107314516,
      "grad_norm": 1.125,
      "learning_rate": 0.0004847150068644805,
      "loss": 5.4609,
      "mean_token_accuracy": 0.1857457920908928,
      "num_tokens": 29141731.0,
      "step": 14385
    },
    {
      "entropy": 5.874402189254761,
      "epoch": 1.2575373590841563,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048470371560111353,
      "loss": 5.4733,
      "mean_token_accuracy": 0.1815667062997818,
      "num_tokens": 29152185.0,
      "step": 14390
    },
    {
      "entropy": 5.888930797576904,
      "epoch": 1.257974307436861,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004846924203155194,
      "loss": 5.2927,
      "mean_token_accuracy": 0.19628028869628905,
      "num_tokens": 29162337.0,
      "step": 14395
    },
    {
      "entropy": 5.886438465118408,
      "epoch": 1.2584112557895657,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048468112100791483,
      "loss": 5.4493,
      "mean_token_accuracy": 0.1877085968852043,
      "num_tokens": 29173246.0,
      "step": 14400
    },
    {
      "entropy": 5.871572208404541,
      "epoch": 1.2588482041422704,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004846698176785165,
      "loss": 5.3871,
      "mean_token_accuracy": 0.18904392719268798,
      "num_tokens": 29183700.0,
      "step": 14405
    },
    {
      "entropy": 5.885785865783691,
      "epoch": 1.259285152494975,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004846585103275412,
      "loss": 5.4047,
      "mean_token_accuracy": 0.1845958113670349,
      "num_tokens": 29193686.0,
      "step": 14410
    },
    {
      "entropy": 5.832409620285034,
      "epoch": 1.2597221008476798,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048464719895520577,
      "loss": 5.4292,
      "mean_token_accuracy": 0.18742207884788514,
      "num_tokens": 29203575.0,
      "step": 14415
    },
    {
      "entropy": 5.856665849685669,
      "epoch": 1.2601590492003845,
      "grad_norm": 1.0,
      "learning_rate": 0.0004846358835617273,
      "loss": 5.3859,
      "mean_token_accuracy": 0.18939711898565292,
      "num_tokens": 29213591.0,
      "step": 14420
    },
    {
      "entropy": 5.8546350479125975,
      "epoch": 1.2605959975530892,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048462456414732257,
      "loss": 5.3451,
      "mean_token_accuracy": 0.19572096019983293,
      "num_tokens": 29222799.0,
      "step": 14425
    },
    {
      "entropy": 5.871230888366699,
      "epoch": 1.2610329459057938,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048461324071220896,
      "loss": 5.3918,
      "mean_token_accuracy": 0.18986170142889022,
      "num_tokens": 29232477.0,
      "step": 14430
    },
    {
      "entropy": 5.868650913238525,
      "epoch": 1.2614698942584988,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048460191325660356,
      "loss": 5.3757,
      "mean_token_accuracy": 0.19189951717853546,
      "num_tokens": 29243477.0,
      "step": 14435
    },
    {
      "entropy": 5.874276733398437,
      "epoch": 1.2619068426112032,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048459058178072366,
      "loss": 5.4531,
      "mean_token_accuracy": 0.18086506724357604,
      "num_tokens": 29253426.0,
      "step": 14440
    },
    {
      "entropy": 5.873958110809326,
      "epoch": 1.2623437909639081,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004845792462847866,
      "loss": 5.4051,
      "mean_token_accuracy": 0.18930511623620988,
      "num_tokens": 29263639.0,
      "step": 14445
    },
    {
      "entropy": 5.881258296966553,
      "epoch": 1.2627807393166128,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048456790676900963,
      "loss": 5.4267,
      "mean_token_accuracy": 0.18349214643239975,
      "num_tokens": 29273980.0,
      "step": 14450
    },
    {
      "entropy": 5.830571460723877,
      "epoch": 1.2632176876693175,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048455656323361055,
      "loss": 5.3386,
      "mean_token_accuracy": 0.20076336562633515,
      "num_tokens": 29283983.0,
      "step": 14455
    },
    {
      "entropy": 5.942102241516113,
      "epoch": 1.2636546360220222,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004845452156788067,
      "loss": 5.4419,
      "mean_token_accuracy": 0.1865229144692421,
      "num_tokens": 29294460.0,
      "step": 14460
    },
    {
      "entropy": 5.847047328948975,
      "epoch": 1.264091584374727,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004845338641048158,
      "loss": 5.4457,
      "mean_token_accuracy": 0.18649567663669586,
      "num_tokens": 29305788.0,
      "step": 14465
    },
    {
      "entropy": 5.925645542144776,
      "epoch": 1.2645285327274316,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048452250851185565,
      "loss": 5.4178,
      "mean_token_accuracy": 0.18901634961366653,
      "num_tokens": 29315457.0,
      "step": 14470
    },
    {
      "entropy": 5.888629531860351,
      "epoch": 1.2649654810801363,
      "grad_norm": 1.0,
      "learning_rate": 0.000484511148900144,
      "loss": 5.3575,
      "mean_token_accuracy": 0.19004132598638535,
      "num_tokens": 29325713.0,
      "step": 14475
    },
    {
      "entropy": 5.826466464996338,
      "epoch": 1.265402429432841,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004844997852698987,
      "loss": 5.3836,
      "mean_token_accuracy": 0.18858492076396943,
      "num_tokens": 29336489.0,
      "step": 14480
    },
    {
      "entropy": 5.8472682476043705,
      "epoch": 1.2658393777855457,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004844884176213377,
      "loss": 5.3751,
      "mean_token_accuracy": 0.19219130128622056,
      "num_tokens": 29345567.0,
      "step": 14485
    },
    {
      "entropy": 5.89044942855835,
      "epoch": 1.2662763261382506,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048447704595467917,
      "loss": 5.3823,
      "mean_token_accuracy": 0.18720629811286926,
      "num_tokens": 29355890.0,
      "step": 14490
    },
    {
      "entropy": 5.886935091018676,
      "epoch": 1.266713274490955,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048446567027014104,
      "loss": 5.4112,
      "mean_token_accuracy": 0.19072587341070174,
      "num_tokens": 29365948.0,
      "step": 14495
    },
    {
      "entropy": 5.817038345336914,
      "epoch": 1.26715022284366,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004844542905679417,
      "loss": 5.354,
      "mean_token_accuracy": 0.1922011584043503,
      "num_tokens": 29375673.0,
      "step": 14500
    },
    {
      "entropy": 5.875889587402344,
      "epoch": 1.2675871711963647,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004844429068482993,
      "loss": 5.3654,
      "mean_token_accuracy": 0.18682362735271454,
      "num_tokens": 29384567.0,
      "step": 14505
    },
    {
      "entropy": 5.86551947593689,
      "epoch": 1.2680241195490694,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048443151911143213,
      "loss": 5.289,
      "mean_token_accuracy": 0.19846013337373733,
      "num_tokens": 29393747.0,
      "step": 14510
    },
    {
      "entropy": 5.8559716701507565,
      "epoch": 1.268461067901774,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004844201273575587,
      "loss": 5.3813,
      "mean_token_accuracy": 0.1904388889670372,
      "num_tokens": 29404095.0,
      "step": 14515
    },
    {
      "entropy": 5.814774084091186,
      "epoch": 1.2688980162544787,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048440873158689754,
      "loss": 5.3785,
      "mean_token_accuracy": 0.19310490190982818,
      "num_tokens": 29414450.0,
      "step": 14520
    },
    {
      "entropy": 5.8578719139099125,
      "epoch": 1.2693349646071834,
      "grad_norm": 1.125,
      "learning_rate": 0.0004843973317996672,
      "loss": 5.3495,
      "mean_token_accuracy": 0.18873665481805801,
      "num_tokens": 29424054.0,
      "step": 14525
    },
    {
      "entropy": 5.898781967163086,
      "epoch": 1.2697719129598881,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004843859279960862,
      "loss": 5.4373,
      "mean_token_accuracy": 0.18835653960704804,
      "num_tokens": 29434585.0,
      "step": 14530
    },
    {
      "entropy": 5.869612121582032,
      "epoch": 1.2702088613125928,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004843745201763735,
      "loss": 5.3894,
      "mean_token_accuracy": 0.1968138411641121,
      "num_tokens": 29443863.0,
      "step": 14535
    },
    {
      "entropy": 5.849461603164673,
      "epoch": 1.2706458096652975,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048436310834074764,
      "loss": 5.3672,
      "mean_token_accuracy": 0.1936037629842758,
      "num_tokens": 29453851.0,
      "step": 14540
    },
    {
      "entropy": 5.82773494720459,
      "epoch": 1.2710827580180022,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004843516924894278,
      "loss": 5.3062,
      "mean_token_accuracy": 0.19270657747983932,
      "num_tokens": 29465152.0,
      "step": 14545
    },
    {
      "entropy": 5.8580645561218265,
      "epoch": 1.2715197063707069,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048434027262263274,
      "loss": 5.315,
      "mean_token_accuracy": 0.20127885937690734,
      "num_tokens": 29474749.0,
      "step": 14550
    },
    {
      "entropy": 5.866223001480103,
      "epoch": 1.2719566547234118,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004843288487405816,
      "loss": 5.3878,
      "mean_token_accuracy": 0.1859459847211838,
      "num_tokens": 29483554.0,
      "step": 14555
    },
    {
      "entropy": 5.846228075027466,
      "epoch": 1.2723936030761165,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048431742084349347,
      "loss": 5.3499,
      "mean_token_accuracy": 0.1972474694252014,
      "num_tokens": 29493718.0,
      "step": 14560
    },
    {
      "entropy": 5.862989282608032,
      "epoch": 1.2728305514288212,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004843059889315874,
      "loss": 5.4088,
      "mean_token_accuracy": 0.18288561552762986,
      "num_tokens": 29504407.0,
      "step": 14565
    },
    {
      "entropy": 5.9054155349731445,
      "epoch": 1.2732674997815259,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004842945530050829,
      "loss": 5.421,
      "mean_token_accuracy": 0.18296885639429092,
      "num_tokens": 29514992.0,
      "step": 14570
    },
    {
      "entropy": 5.923240852355957,
      "epoch": 1.2737044481342306,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048428311306419913,
      "loss": 5.4462,
      "mean_token_accuracy": 0.18700589388608932,
      "num_tokens": 29525109.0,
      "step": 14575
    },
    {
      "entropy": 5.866377973556519,
      "epoch": 1.2741413964869353,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048427166910915565,
      "loss": 5.4436,
      "mean_token_accuracy": 0.18513092398643494,
      "num_tokens": 29536131.0,
      "step": 14580
    },
    {
      "entropy": 5.789508819580078,
      "epoch": 1.27457834483964,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004842602211401719,
      "loss": 5.3489,
      "mean_token_accuracy": 0.19735613316297532,
      "num_tokens": 29547614.0,
      "step": 14585
    },
    {
      "entropy": 5.864696359634399,
      "epoch": 1.2750152931923446,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004842487691574675,
      "loss": 5.4009,
      "mean_token_accuracy": 0.19181082546710967,
      "num_tokens": 29557499.0,
      "step": 14590
    },
    {
      "entropy": 5.933199262619018,
      "epoch": 1.2754522415450493,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048423731316126194,
      "loss": 5.4549,
      "mean_token_accuracy": 0.18273298293352128,
      "num_tokens": 29569909.0,
      "step": 14595
    },
    {
      "entropy": 5.964929246902466,
      "epoch": 1.275889189897754,
      "grad_norm": 1.0,
      "learning_rate": 0.0004842258531517752,
      "loss": 5.5372,
      "mean_token_accuracy": 0.1812294140458107,
      "num_tokens": 29581261.0,
      "step": 14600
    },
    {
      "entropy": 5.895100879669189,
      "epoch": 1.2763261382504587,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004842143891292269,
      "loss": 5.3458,
      "mean_token_accuracy": 0.19377636164426804,
      "num_tokens": 29590604.0,
      "step": 14605
    },
    {
      "entropy": 5.825971841812134,
      "epoch": 1.2767630866031636,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048420292109383704,
      "loss": 5.4295,
      "mean_token_accuracy": 0.18738411962985993,
      "num_tokens": 29600129.0,
      "step": 14610
    },
    {
      "entropy": 5.819851732254028,
      "epoch": 1.277200034955868,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048419144904582545,
      "loss": 5.2995,
      "mean_token_accuracy": 0.2023376077413559,
      "num_tokens": 29609608.0,
      "step": 14615
    },
    {
      "entropy": 5.851500797271728,
      "epoch": 1.277636983308573,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048417997298541235,
      "loss": 5.308,
      "mean_token_accuracy": 0.20094409584999084,
      "num_tokens": 29618764.0,
      "step": 14620
    },
    {
      "entropy": 5.887333583831787,
      "epoch": 1.2780739316612777,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048416849291281765,
      "loss": 5.4793,
      "mean_token_accuracy": 0.18316165059804917,
      "num_tokens": 29628703.0,
      "step": 14625
    },
    {
      "entropy": 5.8482444286346436,
      "epoch": 1.2785108800139824,
      "grad_norm": 1.0,
      "learning_rate": 0.00048415700882826167,
      "loss": 5.3727,
      "mean_token_accuracy": 0.19112706184387207,
      "num_tokens": 29639291.0,
      "step": 14630
    },
    {
      "entropy": 5.8699850082397464,
      "epoch": 1.278947828366687,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048414552073196477,
      "loss": 5.3645,
      "mean_token_accuracy": 0.1857313483953476,
      "num_tokens": 29649058.0,
      "step": 14635
    },
    {
      "entropy": 5.907702779769897,
      "epoch": 1.2793847767193918,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004841340286241471,
      "loss": 5.4089,
      "mean_token_accuracy": 0.19056427776813506,
      "num_tokens": 29658629.0,
      "step": 14640
    },
    {
      "entropy": 5.908832311630249,
      "epoch": 1.2798217250720965,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004841225325050292,
      "loss": 5.4694,
      "mean_token_accuracy": 0.18414859622716903,
      "num_tokens": 29668875.0,
      "step": 14645
    },
    {
      "entropy": 5.862551975250244,
      "epoch": 1.2802586734248012,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048411103237483156,
      "loss": 5.4449,
      "mean_token_accuracy": 0.18495285362005234,
      "num_tokens": 29678427.0,
      "step": 14650
    },
    {
      "entropy": 5.831882524490356,
      "epoch": 1.2806956217775058,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048409952823377474,
      "loss": 5.3419,
      "mean_token_accuracy": 0.1913759544491768,
      "num_tokens": 29688975.0,
      "step": 14655
    },
    {
      "entropy": 5.888248920440674,
      "epoch": 1.2811325701302105,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048408802008207937,
      "loss": 5.4245,
      "mean_token_accuracy": 0.1888510376214981,
      "num_tokens": 29699550.0,
      "step": 14660
    },
    {
      "entropy": 5.866792011260986,
      "epoch": 1.2815695184829154,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048407650791996624,
      "loss": 5.3698,
      "mean_token_accuracy": 0.1919221043586731,
      "num_tokens": 29709597.0,
      "step": 14665
    },
    {
      "entropy": 5.902280521392822,
      "epoch": 1.28200646683562,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004840649917476561,
      "loss": 5.4188,
      "mean_token_accuracy": 0.18913152813911438,
      "num_tokens": 29719250.0,
      "step": 14670
    },
    {
      "entropy": 5.917704629898071,
      "epoch": 1.2824434151883248,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048405347156536995,
      "loss": 5.4396,
      "mean_token_accuracy": 0.18626571744680404,
      "num_tokens": 29728809.0,
      "step": 14675
    },
    {
      "entropy": 5.880138874053955,
      "epoch": 1.2828803635410295,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004840419473733287,
      "loss": 5.3431,
      "mean_token_accuracy": 0.19813480228185654,
      "num_tokens": 29739240.0,
      "step": 14680
    },
    {
      "entropy": 5.866731643676758,
      "epoch": 1.2833173118937342,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048403041917175323,
      "loss": 5.3992,
      "mean_token_accuracy": 0.19080274254083635,
      "num_tokens": 29749284.0,
      "step": 14685
    },
    {
      "entropy": 5.907725286483765,
      "epoch": 1.283754260246439,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048401888696086485,
      "loss": 5.4428,
      "mean_token_accuracy": 0.18542453795671462,
      "num_tokens": 29758770.0,
      "step": 14690
    },
    {
      "entropy": 5.866218900680542,
      "epoch": 1.2841912085991436,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004840073507408847,
      "loss": 5.389,
      "mean_token_accuracy": 0.19117940962314606,
      "num_tokens": 29770111.0,
      "step": 14695
    },
    {
      "entropy": 5.865109062194824,
      "epoch": 1.2846281569518483,
      "grad_norm": 1.0,
      "learning_rate": 0.000483995810512034,
      "loss": 5.3534,
      "mean_token_accuracy": 0.19342575669288636,
      "num_tokens": 29780210.0,
      "step": 14700
    },
    {
      "entropy": 5.894086313247681,
      "epoch": 1.285065105304553,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048398426627453423,
      "loss": 5.3668,
      "mean_token_accuracy": 0.18945474326610565,
      "num_tokens": 29789903.0,
      "step": 14705
    },
    {
      "entropy": 5.939513254165649,
      "epoch": 1.2855020536572577,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048397271802860667,
      "loss": 5.4582,
      "mean_token_accuracy": 0.1851410150527954,
      "num_tokens": 29799377.0,
      "step": 14710
    },
    {
      "entropy": 5.853586626052857,
      "epoch": 1.2859390020099624,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004839611657744729,
      "loss": 5.3577,
      "mean_token_accuracy": 0.19180207550525666,
      "num_tokens": 29809173.0,
      "step": 14715
    },
    {
      "entropy": 5.838394403457642,
      "epoch": 1.286375950362667,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048394960951235444,
      "loss": 5.4715,
      "mean_token_accuracy": 0.1832541272044182,
      "num_tokens": 29819475.0,
      "step": 14720
    },
    {
      "entropy": 5.8546875,
      "epoch": 1.2868128987153717,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000483938049242473,
      "loss": 5.3374,
      "mean_token_accuracy": 0.18925970047712326,
      "num_tokens": 29829109.0,
      "step": 14725
    },
    {
      "entropy": 5.92034502029419,
      "epoch": 1.2872498470680767,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048392648496505024,
      "loss": 5.3657,
      "mean_token_accuracy": 0.1877871885895729,
      "num_tokens": 29839735.0,
      "step": 14730
    },
    {
      "entropy": 5.823561000823974,
      "epoch": 1.2876867954207811,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004839149166803081,
      "loss": 5.3444,
      "mean_token_accuracy": 0.1887977123260498,
      "num_tokens": 29850039.0,
      "step": 14735
    },
    {
      "entropy": 5.8804511547088625,
      "epoch": 1.288123743773486,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004839033443884684,
      "loss": 5.3734,
      "mean_token_accuracy": 0.1900712102651596,
      "num_tokens": 29859411.0,
      "step": 14740
    },
    {
      "entropy": 5.887357568740844,
      "epoch": 1.2885606921261907,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004838917680897531,
      "loss": 5.4281,
      "mean_token_accuracy": 0.1882951632142067,
      "num_tokens": 29870426.0,
      "step": 14745
    },
    {
      "entropy": 5.894597721099854,
      "epoch": 1.2889976404788954,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048388018778438413,
      "loss": 5.3733,
      "mean_token_accuracy": 0.1881062611937523,
      "num_tokens": 29881992.0,
      "step": 14750
    },
    {
      "entropy": 5.875158500671387,
      "epoch": 1.2894345888316001,
      "grad_norm": 1.0,
      "learning_rate": 0.0004838686034725838,
      "loss": 5.3892,
      "mean_token_accuracy": 0.19313714057207107,
      "num_tokens": 29891496.0,
      "step": 14755
    },
    {
      "entropy": 5.920532894134522,
      "epoch": 1.2898715371843048,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048385701515457413,
      "loss": 5.3709,
      "mean_token_accuracy": 0.18592728823423385,
      "num_tokens": 29901529.0,
      "step": 14760
    },
    {
      "entropy": 5.847044897079468,
      "epoch": 1.2903084855370095,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004838454228305775,
      "loss": 5.4357,
      "mean_token_accuracy": 0.19429722875356675,
      "num_tokens": 29910773.0,
      "step": 14765
    },
    {
      "entropy": 5.853058099746704,
      "epoch": 1.2907454338897142,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048383382650081633,
      "loss": 5.4161,
      "mean_token_accuracy": 0.1876201555132866,
      "num_tokens": 29919381.0,
      "step": 14770
    },
    {
      "entropy": 5.916363477706909,
      "epoch": 1.2911823822424189,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004838222261655129,
      "loss": 5.4293,
      "mean_token_accuracy": 0.18731344640254974,
      "num_tokens": 29929179.0,
      "step": 14775
    },
    {
      "entropy": 5.888109922409058,
      "epoch": 1.2916193305951236,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048381062182488964,
      "loss": 5.3475,
      "mean_token_accuracy": 0.1975086197257042,
      "num_tokens": 29938252.0,
      "step": 14780
    },
    {
      "entropy": 5.839063835144043,
      "epoch": 1.2920562789478285,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048379901347916926,
      "loss": 5.411,
      "mean_token_accuracy": 0.19016746580600738,
      "num_tokens": 29947489.0,
      "step": 14785
    },
    {
      "entropy": 5.851382255554199,
      "epoch": 1.292493227300533,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048378740112857443,
      "loss": 5.341,
      "mean_token_accuracy": 0.20083664655685424,
      "num_tokens": 29956788.0,
      "step": 14790
    },
    {
      "entropy": 5.905893707275391,
      "epoch": 1.2929301756532379,
      "grad_norm": 1.125,
      "learning_rate": 0.00048377578477332785,
      "loss": 5.4384,
      "mean_token_accuracy": 0.18436674922704696,
      "num_tokens": 29967161.0,
      "step": 14795
    },
    {
      "entropy": 5.831804990768433,
      "epoch": 1.2933671240059426,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004837641644136523,
      "loss": 5.2607,
      "mean_token_accuracy": 0.20683836042881013,
      "num_tokens": 29976739.0,
      "step": 14800
    },
    {
      "entropy": 5.8330306053161625,
      "epoch": 1.2938040723586473,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048375254004977074,
      "loss": 5.3734,
      "mean_token_accuracy": 0.19115053564310075,
      "num_tokens": 29986739.0,
      "step": 14805
    },
    {
      "entropy": 5.82058310508728,
      "epoch": 1.294241020711352,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004837409116819061,
      "loss": 5.3566,
      "mean_token_accuracy": 0.1932534545660019,
      "num_tokens": 29996687.0,
      "step": 14810
    },
    {
      "entropy": 5.836975860595703,
      "epoch": 1.2946779690640566,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004837292793102813,
      "loss": 5.3959,
      "mean_token_accuracy": 0.19238838106393813,
      "num_tokens": 30007757.0,
      "step": 14815
    },
    {
      "entropy": 5.795372390747071,
      "epoch": 1.2951149174167613,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048371764293511963,
      "loss": 5.3456,
      "mean_token_accuracy": 0.20221833884716034,
      "num_tokens": 30018129.0,
      "step": 14820
    },
    {
      "entropy": 5.901572179794312,
      "epoch": 1.295551865769466,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004837060025566441,
      "loss": 5.3559,
      "mean_token_accuracy": 0.18933685272932052,
      "num_tokens": 30028194.0,
      "step": 14825
    },
    {
      "entropy": 5.87561264038086,
      "epoch": 1.2959888141221707,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004836943581750782,
      "loss": 5.5102,
      "mean_token_accuracy": 0.17577943205833435,
      "num_tokens": 30038182.0,
      "step": 14830
    },
    {
      "entropy": 5.915106153488159,
      "epoch": 1.2964257624748754,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004836827097906451,
      "loss": 5.4001,
      "mean_token_accuracy": 0.18971140384674073,
      "num_tokens": 30047848.0,
      "step": 14835
    },
    {
      "entropy": 5.896192169189453,
      "epoch": 1.2968627108275803,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048367105740356825,
      "loss": 5.4649,
      "mean_token_accuracy": 0.18906660377979279,
      "num_tokens": 30058697.0,
      "step": 14840
    },
    {
      "entropy": 5.871225786209107,
      "epoch": 1.2972996591802848,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048365940101407123,
      "loss": 5.3007,
      "mean_token_accuracy": 0.19518664330244065,
      "num_tokens": 30068611.0,
      "step": 14845
    },
    {
      "entropy": 5.843736219406128,
      "epoch": 1.2977366075329897,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004836477406223775,
      "loss": 5.36,
      "mean_token_accuracy": 0.19373082369565964,
      "num_tokens": 30077992.0,
      "step": 14850
    },
    {
      "entropy": 5.973509407043457,
      "epoch": 1.2981735558856944,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048363607622871086,
      "loss": 5.5321,
      "mean_token_accuracy": 0.18172764927148818,
      "num_tokens": 30088940.0,
      "step": 14855
    },
    {
      "entropy": 5.9663347721099855,
      "epoch": 1.298610504238399,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048362440783329485,
      "loss": 5.4065,
      "mean_token_accuracy": 0.1838187724351883,
      "num_tokens": 30099249.0,
      "step": 14860
    },
    {
      "entropy": 5.840696048736572,
      "epoch": 1.2990474525911038,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004836127354363535,
      "loss": 5.4253,
      "mean_token_accuracy": 0.1932043895125389,
      "num_tokens": 30110548.0,
      "step": 14865
    },
    {
      "entropy": 5.847960138320923,
      "epoch": 1.2994844009438085,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048360105903811045,
      "loss": 5.3799,
      "mean_token_accuracy": 0.2013795182108879,
      "num_tokens": 30120842.0,
      "step": 14870
    },
    {
      "entropy": 5.9378594875335695,
      "epoch": 1.2999213492965132,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048358937863878984,
      "loss": 5.43,
      "mean_token_accuracy": 0.18580985218286514,
      "num_tokens": 30130439.0,
      "step": 14875
    },
    {
      "entropy": 5.925860595703125,
      "epoch": 1.3003582976492178,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004835776942386157,
      "loss": 5.4473,
      "mean_token_accuracy": 0.18574035465717315,
      "num_tokens": 30140976.0,
      "step": 14880
    },
    {
      "entropy": 5.863036108016968,
      "epoch": 1.3007952460019225,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000483566005837812,
      "loss": 5.3964,
      "mean_token_accuracy": 0.188516204059124,
      "num_tokens": 30150989.0,
      "step": 14885
    },
    {
      "entropy": 5.851482725143432,
      "epoch": 1.3012321943546272,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048355431343660303,
      "loss": 5.3508,
      "mean_token_accuracy": 0.19175131469964982,
      "num_tokens": 30161605.0,
      "step": 14890
    },
    {
      "entropy": 5.855576705932617,
      "epoch": 1.301669142707332,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004835426170352131,
      "loss": 5.3677,
      "mean_token_accuracy": 0.19269084334373474,
      "num_tokens": 30171361.0,
      "step": 14895
    },
    {
      "entropy": 5.886048221588135,
      "epoch": 1.3021060910600366,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048353091663386655,
      "loss": 5.3133,
      "mean_token_accuracy": 0.19601616114377976,
      "num_tokens": 30180352.0,
      "step": 14900
    },
    {
      "entropy": 5.844298553466797,
      "epoch": 1.3025430394127415,
      "grad_norm": 1.0,
      "learning_rate": 0.0004835192122327877,
      "loss": 5.3344,
      "mean_token_accuracy": 0.1906298041343689,
      "num_tokens": 30190624.0,
      "step": 14905
    },
    {
      "entropy": 5.84540548324585,
      "epoch": 1.302979987765446,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048350750383220105,
      "loss": 5.3724,
      "mean_token_accuracy": 0.19287998229265213,
      "num_tokens": 30200203.0,
      "step": 14910
    },
    {
      "entropy": 5.849333715438843,
      "epoch": 1.303416936118151,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048349579143233126,
      "loss": 5.3334,
      "mean_token_accuracy": 0.20082523971796035,
      "num_tokens": 30210546.0,
      "step": 14915
    },
    {
      "entropy": 5.937950229644775,
      "epoch": 1.3038538844708556,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004834840750334029,
      "loss": 5.5013,
      "mean_token_accuracy": 0.18268941938877106,
      "num_tokens": 30221316.0,
      "step": 14920
    },
    {
      "entropy": 5.930574798583985,
      "epoch": 1.3042908328235603,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004834723546356408,
      "loss": 5.421,
      "mean_token_accuracy": 0.1882130518555641,
      "num_tokens": 30232580.0,
      "step": 14925
    },
    {
      "entropy": 5.894370365142822,
      "epoch": 1.304727781176265,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048346063023926964,
      "loss": 5.3672,
      "mean_token_accuracy": 0.19104282408952714,
      "num_tokens": 30241976.0,
      "step": 14930
    },
    {
      "entropy": 5.853033971786499,
      "epoch": 1.3051647295289697,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048344890184451436,
      "loss": 5.3529,
      "mean_token_accuracy": 0.19086937755346298,
      "num_tokens": 30252723.0,
      "step": 14935
    },
    {
      "entropy": 5.911515712738037,
      "epoch": 1.3056016778816744,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048343716945159997,
      "loss": 5.4091,
      "mean_token_accuracy": 0.1909930869936943,
      "num_tokens": 30263125.0,
      "step": 14940
    },
    {
      "entropy": 5.8324501514434814,
      "epoch": 1.306038626234379,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004834254330607514,
      "loss": 5.3236,
      "mean_token_accuracy": 0.2007976695895195,
      "num_tokens": 30273113.0,
      "step": 14945
    },
    {
      "entropy": 5.826126384735107,
      "epoch": 1.3064755745870837,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004834136926721939,
      "loss": 5.3025,
      "mean_token_accuracy": 0.20009291172027588,
      "num_tokens": 30282550.0,
      "step": 14950
    },
    {
      "entropy": 5.816371631622315,
      "epoch": 1.3069125229397884,
      "grad_norm": 1.0,
      "learning_rate": 0.00048340194828615246,
      "loss": 5.3397,
      "mean_token_accuracy": 0.18655403703451157,
      "num_tokens": 30293377.0,
      "step": 14955
    },
    {
      "entropy": 5.888375854492187,
      "epoch": 1.3073494712924933,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004833901999028525,
      "loss": 5.3593,
      "mean_token_accuracy": 0.19305720776319504,
      "num_tokens": 30303147.0,
      "step": 14960
    },
    {
      "entropy": 5.890161943435669,
      "epoch": 1.3077864196451978,
      "grad_norm": 1.0,
      "learning_rate": 0.0004833784475225192,
      "loss": 5.497,
      "mean_token_accuracy": 0.1822294518351555,
      "num_tokens": 30313426.0,
      "step": 14965
    },
    {
      "entropy": 5.882660818099976,
      "epoch": 1.3082233679979027,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004833666911453783,
      "loss": 5.4225,
      "mean_token_accuracy": 0.18194729834794998,
      "num_tokens": 30323896.0,
      "step": 14970
    },
    {
      "entropy": 5.868225479125977,
      "epoch": 1.3086603163506074,
      "grad_norm": 0.953125,
      "learning_rate": 0.000483354930771655,
      "loss": 5.4034,
      "mean_token_accuracy": 0.18461208343505858,
      "num_tokens": 30333523.0,
      "step": 14975
    },
    {
      "entropy": 5.855851459503174,
      "epoch": 1.3090972647033121,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048334316640157493,
      "loss": 5.3659,
      "mean_token_accuracy": 0.1924608677625656,
      "num_tokens": 30344035.0,
      "step": 14980
    },
    {
      "entropy": 5.876605701446533,
      "epoch": 1.3095342130560168,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004833313980353637,
      "loss": 5.4479,
      "mean_token_accuracy": 0.19019196629524232,
      "num_tokens": 30354385.0,
      "step": 14985
    },
    {
      "entropy": 5.900607872009277,
      "epoch": 1.3099711614087215,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048331962567324717,
      "loss": 5.4912,
      "mean_token_accuracy": 0.18084898144006728,
      "num_tokens": 30364941.0,
      "step": 14990
    },
    {
      "entropy": 5.8725484848022464,
      "epoch": 1.3104081097614262,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000483307849315451,
      "loss": 5.3431,
      "mean_token_accuracy": 0.1920125037431717,
      "num_tokens": 30376160.0,
      "step": 14995
    },
    {
      "entropy": 5.9344690322875975,
      "epoch": 1.3108450581141309,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048329606896220116,
      "loss": 5.4489,
      "mean_token_accuracy": 0.1863630786538124,
      "num_tokens": 30385018.0,
      "step": 15000
    },
    {
      "epoch": 1.3108450581141309,
      "eval_entropy": 5.64429235880901,
      "eval_loss": 5.436652660369873,
      "eval_mean_token_accuracy": 0.19698683400232822,
      "eval_num_tokens": 30385018.0,
      "eval_runtime": 21.4942,
      "eval_samples_per_second": 1722.047,
      "eval_steps_per_second": 215.268,
      "step": 15000
    },
    {
      "entropy": 5.858981561660767,
      "epoch": 1.3112820064668356,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004832842846137236,
      "loss": 5.3806,
      "mean_token_accuracy": 0.1915394902229309,
      "num_tokens": 30394618.0,
      "step": 15005
    },
    {
      "entropy": 5.891623640060425,
      "epoch": 1.3117189548195403,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048327249627024433,
      "loss": 5.3869,
      "mean_token_accuracy": 0.19138570725917817,
      "num_tokens": 30404081.0,
      "step": 15010
    },
    {
      "entropy": 5.88922815322876,
      "epoch": 1.3121559031722452,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004832607039319894,
      "loss": 5.3538,
      "mean_token_accuracy": 0.18897589296102524,
      "num_tokens": 30413892.0,
      "step": 15015
    },
    {
      "entropy": 5.933477306365967,
      "epoch": 1.3125928515249496,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048324890759918513,
      "loss": 5.4351,
      "mean_token_accuracy": 0.183906252682209,
      "num_tokens": 30424523.0,
      "step": 15020
    },
    {
      "entropy": 5.857788753509522,
      "epoch": 1.3130297998776546,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004832371072720576,
      "loss": 5.3719,
      "mean_token_accuracy": 0.19573804140090942,
      "num_tokens": 30435092.0,
      "step": 15025
    },
    {
      "entropy": 5.752363586425782,
      "epoch": 1.3134667482303592,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004832253029508332,
      "loss": 5.3326,
      "mean_token_accuracy": 0.19133069813251496,
      "num_tokens": 30445628.0,
      "step": 15030
    },
    {
      "entropy": 5.8328770160675045,
      "epoch": 1.313903696583064,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004832134946357384,
      "loss": 5.401,
      "mean_token_accuracy": 0.19171629697084427,
      "num_tokens": 30457433.0,
      "step": 15035
    },
    {
      "entropy": 5.902551746368408,
      "epoch": 1.3143406449357686,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004832016823269998,
      "loss": 5.381,
      "mean_token_accuracy": 0.18880934864282609,
      "num_tokens": 30467987.0,
      "step": 15040
    },
    {
      "entropy": 5.948355579376221,
      "epoch": 1.3147775932884733,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004831898660248437,
      "loss": 5.3883,
      "mean_token_accuracy": 0.18448175489902496,
      "num_tokens": 30476577.0,
      "step": 15045
    },
    {
      "entropy": 5.832091808319092,
      "epoch": 1.315214541641178,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048317804572949697,
      "loss": 5.3414,
      "mean_token_accuracy": 0.18954382985830306,
      "num_tokens": 30487410.0,
      "step": 15050
    },
    {
      "entropy": 5.91442174911499,
      "epoch": 1.3156514899938827,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004831662214411862,
      "loss": 5.4606,
      "mean_token_accuracy": 0.18375538289546967,
      "num_tokens": 30497778.0,
      "step": 15055
    },
    {
      "entropy": 5.866795825958252,
      "epoch": 1.3160884383465874,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004831543931601382,
      "loss": 5.3618,
      "mean_token_accuracy": 0.19221826046705245,
      "num_tokens": 30507369.0,
      "step": 15060
    },
    {
      "entropy": 5.888374137878418,
      "epoch": 1.316525386699292,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048314256088657985,
      "loss": 5.4107,
      "mean_token_accuracy": 0.1877640813589096,
      "num_tokens": 30518126.0,
      "step": 15065
    },
    {
      "entropy": 5.875938796997071,
      "epoch": 1.3169623350519968,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004831307246207382,
      "loss": 5.3852,
      "mean_token_accuracy": 0.1980038732290268,
      "num_tokens": 30528167.0,
      "step": 15070
    },
    {
      "entropy": 5.84874701499939,
      "epoch": 1.3173992834047015,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048311888436284017,
      "loss": 5.31,
      "mean_token_accuracy": 0.2073795661330223,
      "num_tokens": 30538111.0,
      "step": 15075
    },
    {
      "entropy": 5.880398511886597,
      "epoch": 1.3178362317574064,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004831070401131129,
      "loss": 5.4388,
      "mean_token_accuracy": 0.1962588131427765,
      "num_tokens": 30547773.0,
      "step": 15080
    },
    {
      "entropy": 5.9066469192504885,
      "epoch": 1.3182731801101109,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004830951918717834,
      "loss": 5.3623,
      "mean_token_accuracy": 0.19016316831111907,
      "num_tokens": 30558058.0,
      "step": 15085
    },
    {
      "entropy": 5.860610055923462,
      "epoch": 1.3187101284628158,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048308333963907927,
      "loss": 5.3137,
      "mean_token_accuracy": 0.19653818011283875,
      "num_tokens": 30567297.0,
      "step": 15090
    },
    {
      "entropy": 5.872169733047485,
      "epoch": 1.3191470768155205,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048307148341522756,
      "loss": 5.3848,
      "mean_token_accuracy": 0.1947799265384674,
      "num_tokens": 30578159.0,
      "step": 15095
    },
    {
      "entropy": 5.872853565216064,
      "epoch": 1.3195840251682251,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048305962320045574,
      "loss": 5.3947,
      "mean_token_accuracy": 0.19242664128541948,
      "num_tokens": 30588734.0,
      "step": 15100
    },
    {
      "entropy": 5.853034973144531,
      "epoch": 1.3200209735209298,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048304775899499143,
      "loss": 5.2948,
      "mean_token_accuracy": 0.1954769089818001,
      "num_tokens": 30598496.0,
      "step": 15105
    },
    {
      "entropy": 5.868622779846191,
      "epoch": 1.3204579218736345,
      "grad_norm": 0.953125,
      "learning_rate": 0.000483035890799062,
      "loss": 5.3744,
      "mean_token_accuracy": 0.1930827796459198,
      "num_tokens": 30608330.0,
      "step": 15110
    },
    {
      "entropy": 5.873728704452515,
      "epoch": 1.3208948702263392,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004830240186128952,
      "loss": 5.3685,
      "mean_token_accuracy": 0.19524016380310058,
      "num_tokens": 30618887.0,
      "step": 15115
    },
    {
      "entropy": 5.8988203525543215,
      "epoch": 1.321331818579044,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048301214243671884,
      "loss": 5.32,
      "mean_token_accuracy": 0.19525677859783172,
      "num_tokens": 30629479.0,
      "step": 15120
    },
    {
      "entropy": 5.8420392036437985,
      "epoch": 1.3217687669317486,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004830002622707604,
      "loss": 5.3162,
      "mean_token_accuracy": 0.19492077380418776,
      "num_tokens": 30639270.0,
      "step": 15125
    },
    {
      "entropy": 5.889102602005005,
      "epoch": 1.3222057152844533,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004829883781152481,
      "loss": 5.4424,
      "mean_token_accuracy": 0.18795035183429717,
      "num_tokens": 30649156.0,
      "step": 15130
    },
    {
      "entropy": 5.8656412124633786,
      "epoch": 1.3226426636371582,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048297648997040966,
      "loss": 5.4597,
      "mean_token_accuracy": 0.1846207246184349,
      "num_tokens": 30659919.0,
      "step": 15135
    },
    {
      "entropy": 5.8821053981781,
      "epoch": 1.3230796119898627,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048296459783647304,
      "loss": 5.4281,
      "mean_token_accuracy": 0.18857254087924957,
      "num_tokens": 30670523.0,
      "step": 15140
    },
    {
      "entropy": 5.937315654754639,
      "epoch": 1.3235165603425676,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048295270171366665,
      "loss": 5.43,
      "mean_token_accuracy": 0.19149835407733917,
      "num_tokens": 30680803.0,
      "step": 15145
    },
    {
      "entropy": 5.891228294372558,
      "epoch": 1.3239535086952723,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048294080160221837,
      "loss": 5.2818,
      "mean_token_accuracy": 0.1966610699892044,
      "num_tokens": 30690401.0,
      "step": 15150
    },
    {
      "entropy": 5.8729747295379635,
      "epoch": 1.324390457047977,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048292889750235665,
      "loss": 5.4342,
      "mean_token_accuracy": 0.18680298924446107,
      "num_tokens": 30700779.0,
      "step": 15155
    },
    {
      "entropy": 5.910748100280761,
      "epoch": 1.3248274054006817,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048291698941430955,
      "loss": 5.4064,
      "mean_token_accuracy": 0.18460820764303207,
      "num_tokens": 30711556.0,
      "step": 15160
    },
    {
      "entropy": 5.795103979110718,
      "epoch": 1.3252643537533864,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004829050773383058,
      "loss": 5.2964,
      "mean_token_accuracy": 0.19949681609869002,
      "num_tokens": 30721929.0,
      "step": 15165
    },
    {
      "entropy": 5.891295337677002,
      "epoch": 1.325701302106091,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004828931612745737,
      "loss": 5.411,
      "mean_token_accuracy": 0.18677400499582292,
      "num_tokens": 30732848.0,
      "step": 15170
    },
    {
      "entropy": 5.89900484085083,
      "epoch": 1.3261382504587957,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048288124122334176,
      "loss": 5.3689,
      "mean_token_accuracy": 0.18739913851022721,
      "num_tokens": 30742628.0,
      "step": 15175
    },
    {
      "entropy": 5.883143281936645,
      "epoch": 1.3265751988115004,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004828693171848388,
      "loss": 5.4167,
      "mean_token_accuracy": 0.18938528299331664,
      "num_tokens": 30752866.0,
      "step": 15180
    },
    {
      "entropy": 5.83332257270813,
      "epoch": 1.3270121471642051,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004828573891592933,
      "loss": 5.3477,
      "mean_token_accuracy": 0.19434332996606826,
      "num_tokens": 30762861.0,
      "step": 15185
    },
    {
      "entropy": 5.924528026580811,
      "epoch": 1.32744909551691,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004828454571469342,
      "loss": 5.4119,
      "mean_token_accuracy": 0.19105525463819503,
      "num_tokens": 30772555.0,
      "step": 15190
    },
    {
      "entropy": 5.926086473464966,
      "epoch": 1.3278860438696145,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004828335211479903,
      "loss": 5.3938,
      "mean_token_accuracy": 0.1925220564007759,
      "num_tokens": 30782739.0,
      "step": 15195
    },
    {
      "entropy": 5.883752155303955,
      "epoch": 1.3283229922223194,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048282158116269065,
      "loss": 5.3696,
      "mean_token_accuracy": 0.19216094464063643,
      "num_tokens": 30792600.0,
      "step": 15200
    },
    {
      "entropy": 5.903535270690918,
      "epoch": 1.3287599405750241,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004828096371912641,
      "loss": 5.3677,
      "mean_token_accuracy": 0.1884703144431114,
      "num_tokens": 30801433.0,
      "step": 15205
    },
    {
      "entropy": 5.898988819122314,
      "epoch": 1.3291968889277288,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048279768923393987,
      "loss": 5.4615,
      "mean_token_accuracy": 0.18595934063196182,
      "num_tokens": 30811277.0,
      "step": 15210
    },
    {
      "entropy": 5.882837963104248,
      "epoch": 1.3296338372804335,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048278573729094705,
      "loss": 5.434,
      "mean_token_accuracy": 0.18404652029275895,
      "num_tokens": 30821120.0,
      "step": 15215
    },
    {
      "entropy": 5.88394775390625,
      "epoch": 1.3300707856331382,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048277378136251487,
      "loss": 5.3394,
      "mean_token_accuracy": 0.19765067398548125,
      "num_tokens": 30831660.0,
      "step": 15220
    },
    {
      "entropy": 5.946486616134644,
      "epoch": 1.3305077339858429,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048276182144887274,
      "loss": 5.4195,
      "mean_token_accuracy": 0.1817472904920578,
      "num_tokens": 30842530.0,
      "step": 15225
    },
    {
      "entropy": 5.8820055484771725,
      "epoch": 1.3309446823385476,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048274985755025003,
      "loss": 5.385,
      "mean_token_accuracy": 0.1887974977493286,
      "num_tokens": 30852954.0,
      "step": 15230
    },
    {
      "entropy": 5.829825687408447,
      "epoch": 1.3313816306912523,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048273788966687615,
      "loss": 5.3027,
      "mean_token_accuracy": 0.19733135402202606,
      "num_tokens": 30862864.0,
      "step": 15235
    },
    {
      "entropy": 5.875374221801758,
      "epoch": 1.331818579043957,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004827259177989807,
      "loss": 5.3738,
      "mean_token_accuracy": 0.19092947989702225,
      "num_tokens": 30873533.0,
      "step": 15240
    },
    {
      "entropy": 5.895572662353516,
      "epoch": 1.3322555273966616,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048271394194679336,
      "loss": 5.3191,
      "mean_token_accuracy": 0.1931588813662529,
      "num_tokens": 30883564.0,
      "step": 15245
    },
    {
      "entropy": 5.853583860397339,
      "epoch": 1.3326924757493663,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004827019621105436,
      "loss": 5.3383,
      "mean_token_accuracy": 0.19729857444763182,
      "num_tokens": 30893516.0,
      "step": 15250
    },
    {
      "entropy": 5.822067546844482,
      "epoch": 1.3331294241020712,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048268997829046156,
      "loss": 5.3589,
      "mean_token_accuracy": 0.19124917685985565,
      "num_tokens": 30903104.0,
      "step": 15255
    },
    {
      "entropy": 5.859000492095947,
      "epoch": 1.3335663724547757,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048267799048677684,
      "loss": 5.3512,
      "mean_token_accuracy": 0.19187511056661605,
      "num_tokens": 30912759.0,
      "step": 15260
    },
    {
      "entropy": 5.883078145980835,
      "epoch": 1.3340033208074806,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004826659986997194,
      "loss": 5.3706,
      "mean_token_accuracy": 0.19076020121574402,
      "num_tokens": 30921361.0,
      "step": 15265
    },
    {
      "entropy": 5.881567859649659,
      "epoch": 1.3344402691601853,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004826540029295193,
      "loss": 5.3629,
      "mean_token_accuracy": 0.18945441544055938,
      "num_tokens": 30931006.0,
      "step": 15270
    },
    {
      "entropy": 5.860344076156617,
      "epoch": 1.33487721751289,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048264200317640665,
      "loss": 5.3523,
      "mean_token_accuracy": 0.1937045633792877,
      "num_tokens": 30940647.0,
      "step": 15275
    },
    {
      "entropy": 5.90737624168396,
      "epoch": 1.3353141658655947,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004826299994406116,
      "loss": 5.4189,
      "mean_token_accuracy": 0.18991085588932038,
      "num_tokens": 30951028.0,
      "step": 15280
    },
    {
      "entropy": 5.85226731300354,
      "epoch": 1.3357511142182994,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048261799172236435,
      "loss": 5.3764,
      "mean_token_accuracy": 0.1878795638680458,
      "num_tokens": 30960726.0,
      "step": 15285
    },
    {
      "entropy": 5.8754651069641115,
      "epoch": 1.336188062571004,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004826059800218952,
      "loss": 5.4449,
      "mean_token_accuracy": 0.18681329637765884,
      "num_tokens": 30971061.0,
      "step": 15290
    },
    {
      "entropy": 5.831655502319336,
      "epoch": 1.3366250109237088,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004825939643394346,
      "loss": 5.2466,
      "mean_token_accuracy": 0.20437230169773102,
      "num_tokens": 30980518.0,
      "step": 15295
    },
    {
      "entropy": 5.8913592338562015,
      "epoch": 1.3370619592764135,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048258194467521294,
      "loss": 5.4651,
      "mean_token_accuracy": 0.18600432723760604,
      "num_tokens": 30990759.0,
      "step": 15300
    },
    {
      "entropy": 5.957705450057984,
      "epoch": 1.3374989076291182,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048256992102946084,
      "loss": 5.4232,
      "mean_token_accuracy": 0.18189277201890947,
      "num_tokens": 31002218.0,
      "step": 15305
    },
    {
      "entropy": 5.823714542388916,
      "epoch": 1.337935855981823,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004825578934024089,
      "loss": 5.3634,
      "mean_token_accuracy": 0.19580103904008866,
      "num_tokens": 31012434.0,
      "step": 15310
    },
    {
      "entropy": 5.925708913803101,
      "epoch": 1.3383728043345275,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048254586179428776,
      "loss": 5.4834,
      "mean_token_accuracy": 0.1821560338139534,
      "num_tokens": 31022805.0,
      "step": 15315
    },
    {
      "entropy": 5.897369337081909,
      "epoch": 1.3388097526872325,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004825338262053283,
      "loss": 5.3554,
      "mean_token_accuracy": 0.19571781903505325,
      "num_tokens": 31032563.0,
      "step": 15320
    },
    {
      "entropy": 5.8488304138183596,
      "epoch": 1.3392467010399371,
      "grad_norm": 1.125,
      "learning_rate": 0.0004825217866357613,
      "loss": 5.351,
      "mean_token_accuracy": 0.1950315237045288,
      "num_tokens": 31041445.0,
      "step": 15325
    },
    {
      "entropy": 5.841315126419067,
      "epoch": 1.3396836493926418,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048250974308581765,
      "loss": 5.3635,
      "mean_token_accuracy": 0.19335675984621048,
      "num_tokens": 31051413.0,
      "step": 15330
    },
    {
      "entropy": 5.866778945922851,
      "epoch": 1.3401205977453465,
      "grad_norm": 1.0,
      "learning_rate": 0.00048249769555572844,
      "loss": 5.3883,
      "mean_token_accuracy": 0.18972503542900085,
      "num_tokens": 31061822.0,
      "step": 15335
    },
    {
      "entropy": 5.844515180587768,
      "epoch": 1.3405575460980512,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048248564404572463,
      "loss": 5.3513,
      "mean_token_accuracy": 0.18915417045354843,
      "num_tokens": 31071889.0,
      "step": 15340
    },
    {
      "entropy": 5.822140455245972,
      "epoch": 1.340994494450756,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048247358855603754,
      "loss": 5.3131,
      "mean_token_accuracy": 0.19309778660535812,
      "num_tokens": 31080933.0,
      "step": 15345
    },
    {
      "entropy": 5.870017099380493,
      "epoch": 1.3414314428034606,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004824615290868982,
      "loss": 5.4145,
      "mean_token_accuracy": 0.18805603086948394,
      "num_tokens": 31091349.0,
      "step": 15350
    },
    {
      "entropy": 5.993978023529053,
      "epoch": 1.3418683911561653,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004824494656385381,
      "loss": 5.4223,
      "mean_token_accuracy": 0.18563077449798585,
      "num_tokens": 31101485.0,
      "step": 15355
    },
    {
      "entropy": 5.879104948043823,
      "epoch": 1.34230533950887,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004824373982111886,
      "loss": 5.355,
      "mean_token_accuracy": 0.18818260431289674,
      "num_tokens": 31111965.0,
      "step": 15360
    },
    {
      "entropy": 5.838479518890381,
      "epoch": 1.3427422878615747,
      "grad_norm": 1.046875,
      "learning_rate": 0.000482425326805081,
      "loss": 5.3649,
      "mean_token_accuracy": 0.1926906704902649,
      "num_tokens": 31122284.0,
      "step": 15365
    },
    {
      "entropy": 5.898171520233154,
      "epoch": 1.3431792362142794,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048241325142044697,
      "loss": 5.3776,
      "mean_token_accuracy": 0.18991642445325851,
      "num_tokens": 31132156.0,
      "step": 15370
    },
    {
      "entropy": 5.898426294326782,
      "epoch": 1.3436161845669843,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048240117205751817,
      "loss": 5.3025,
      "mean_token_accuracy": 0.1992608442902565,
      "num_tokens": 31141609.0,
      "step": 15375
    },
    {
      "entropy": 5.790284490585327,
      "epoch": 1.344053132919689,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004823890887165261,
      "loss": 5.3659,
      "mean_token_accuracy": 0.19094816148281096,
      "num_tokens": 31153159.0,
      "step": 15380
    },
    {
      "entropy": 5.799456167221069,
      "epoch": 1.3444900812723937,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048237700139770273,
      "loss": 5.422,
      "mean_token_accuracy": 0.18992496132850648,
      "num_tokens": 31163330.0,
      "step": 15385
    },
    {
      "entropy": 5.975083208084106,
      "epoch": 1.3449270296250984,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004823649101012797,
      "loss": 5.4417,
      "mean_token_accuracy": 0.1825023517012596,
      "num_tokens": 31173874.0,
      "step": 15390
    },
    {
      "entropy": 5.834002876281739,
      "epoch": 1.345363977977803,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004823528148274891,
      "loss": 5.3166,
      "mean_token_accuracy": 0.19434939175844193,
      "num_tokens": 31183933.0,
      "step": 15395
    },
    {
      "entropy": 5.820701646804809,
      "epoch": 1.3458009263305077,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048234071557656295,
      "loss": 5.3119,
      "mean_token_accuracy": 0.2023688018321991,
      "num_tokens": 31194308.0,
      "step": 15400
    },
    {
      "entropy": 5.895405530929565,
      "epoch": 1.3462378746832124,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004823286123487331,
      "loss": 5.3647,
      "mean_token_accuracy": 0.19260916560888292,
      "num_tokens": 31203757.0,
      "step": 15405
    },
    {
      "entropy": 5.912086820602417,
      "epoch": 1.3466748230359171,
      "grad_norm": 1.0,
      "learning_rate": 0.00048231650514423194,
      "loss": 5.4345,
      "mean_token_accuracy": 0.18519600629806518,
      "num_tokens": 31213980.0,
      "step": 15410
    },
    {
      "entropy": 5.940784311294555,
      "epoch": 1.3471117713886218,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048230439396329143,
      "loss": 5.385,
      "mean_token_accuracy": 0.19287507086992264,
      "num_tokens": 31225034.0,
      "step": 15415
    },
    {
      "entropy": 5.883170509338379,
      "epoch": 1.3475487197413265,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048229227880614413,
      "loss": 5.3784,
      "mean_token_accuracy": 0.18636671900749208,
      "num_tokens": 31236393.0,
      "step": 15420
    },
    {
      "entropy": 5.86506872177124,
      "epoch": 1.3479856680940312,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004822801596730222,
      "loss": 5.3827,
      "mean_token_accuracy": 0.18923078775405883,
      "num_tokens": 31246577.0,
      "step": 15425
    },
    {
      "entropy": 5.849088668823242,
      "epoch": 1.348422616446736,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048226803656415833,
      "loss": 5.3558,
      "mean_token_accuracy": 0.19144546538591384,
      "num_tokens": 31256004.0,
      "step": 15430
    },
    {
      "entropy": 5.9331458568572994,
      "epoch": 1.3488595647994406,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004822559094797849,
      "loss": 5.4193,
      "mean_token_accuracy": 0.18820457756519318,
      "num_tokens": 31267294.0,
      "step": 15435
    },
    {
      "entropy": 5.871834421157837,
      "epoch": 1.3492965131521455,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048224377842013443,
      "loss": 5.3546,
      "mean_token_accuracy": 0.19601739048957825,
      "num_tokens": 31276896.0,
      "step": 15440
    },
    {
      "entropy": 5.849543046951294,
      "epoch": 1.3497334615048502,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004822316433854397,
      "loss": 5.4084,
      "mean_token_accuracy": 0.1904663234949112,
      "num_tokens": 31287248.0,
      "step": 15445
    },
    {
      "entropy": 5.872874546051025,
      "epoch": 1.3501704098575549,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048221950437593353,
      "loss": 5.3611,
      "mean_token_accuracy": 0.19439342319965364,
      "num_tokens": 31297422.0,
      "step": 15450
    },
    {
      "entropy": 5.8913572311401365,
      "epoch": 1.3506073582102596,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048220736139184864,
      "loss": 5.4323,
      "mean_token_accuracy": 0.19087142944335939,
      "num_tokens": 31307182.0,
      "step": 15455
    },
    {
      "entropy": 5.896520328521729,
      "epoch": 1.3510443065629643,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000482195214433418,
      "loss": 5.4831,
      "mean_token_accuracy": 0.18876738995313644,
      "num_tokens": 31317234.0,
      "step": 15460
    },
    {
      "entropy": 5.909190320968628,
      "epoch": 1.351481254915669,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004821830635008746,
      "loss": 5.4638,
      "mean_token_accuracy": 0.18490086495876312,
      "num_tokens": 31328294.0,
      "step": 15465
    },
    {
      "entropy": 5.9185772895812985,
      "epoch": 1.3519182032683736,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004821709085944515,
      "loss": 5.3291,
      "mean_token_accuracy": 0.1916299045085907,
      "num_tokens": 31337250.0,
      "step": 15470
    },
    {
      "entropy": 5.8689305782318115,
      "epoch": 1.3523551516210783,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004821587497143817,
      "loss": 5.4306,
      "mean_token_accuracy": 0.183685040473938,
      "num_tokens": 31346900.0,
      "step": 15475
    },
    {
      "entropy": 5.87187442779541,
      "epoch": 1.352792099973783,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004821465868608986,
      "loss": 5.4081,
      "mean_token_accuracy": 0.18394791185855866,
      "num_tokens": 31357233.0,
      "step": 15480
    },
    {
      "entropy": 5.919680881500244,
      "epoch": 1.353229048326488,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004821344200342355,
      "loss": 5.3222,
      "mean_token_accuracy": 0.19170920103788375,
      "num_tokens": 31367043.0,
      "step": 15485
    },
    {
      "entropy": 5.873647832870484,
      "epoch": 1.3536659966791924,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004821222492346256,
      "loss": 5.3485,
      "mean_token_accuracy": 0.1876885026693344,
      "num_tokens": 31376770.0,
      "step": 15490
    },
    {
      "entropy": 5.866624879837036,
      "epoch": 1.3541029450318973,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048211007446230243,
      "loss": 5.3905,
      "mean_token_accuracy": 0.19387502521276473,
      "num_tokens": 31386056.0,
      "step": 15495
    },
    {
      "entropy": 5.910285091400146,
      "epoch": 1.354539893384602,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048209789571749954,
      "loss": 5.4356,
      "mean_token_accuracy": 0.18545076698064805,
      "num_tokens": 31395129.0,
      "step": 15500
    },
    {
      "entropy": 5.967115163803101,
      "epoch": 1.3549768417373067,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048208571300045045,
      "loss": 5.4505,
      "mean_token_accuracy": 0.18466798067092896,
      "num_tokens": 31404720.0,
      "step": 15505
    },
    {
      "entropy": 5.898995113372803,
      "epoch": 1.3554137900900114,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048207352631138897,
      "loss": 5.4145,
      "mean_token_accuracy": 0.18946489095687866,
      "num_tokens": 31415588.0,
      "step": 15510
    },
    {
      "entropy": 5.906092500686645,
      "epoch": 1.355850738442716,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048206133565054863,
      "loss": 5.4809,
      "mean_token_accuracy": 0.17970745712518693,
      "num_tokens": 31425904.0,
      "step": 15515
    },
    {
      "entropy": 5.917495965957642,
      "epoch": 1.3562876867954208,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004820491410181634,
      "loss": 5.457,
      "mean_token_accuracy": 0.18233900368213654,
      "num_tokens": 31435282.0,
      "step": 15520
    },
    {
      "entropy": 5.964746618270874,
      "epoch": 1.3567246351481255,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048203694241446724,
      "loss": 5.4996,
      "mean_token_accuracy": 0.18483246266841888,
      "num_tokens": 31445681.0,
      "step": 15525
    },
    {
      "entropy": 5.824182605743408,
      "epoch": 1.3571615835008302,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048202473983969396,
      "loss": 5.2148,
      "mean_token_accuracy": 0.19929396361112595,
      "num_tokens": 31456934.0,
      "step": 15530
    },
    {
      "entropy": 5.847763347625732,
      "epoch": 1.3575985318535349,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048201253329407766,
      "loss": 5.3605,
      "mean_token_accuracy": 0.19461302161216737,
      "num_tokens": 31466692.0,
      "step": 15535
    },
    {
      "entropy": 5.862225961685181,
      "epoch": 1.3580354802062395,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004820003227778526,
      "loss": 5.3744,
      "mean_token_accuracy": 0.19523693174123763,
      "num_tokens": 31477166.0,
      "step": 15540
    },
    {
      "entropy": 5.969606590270996,
      "epoch": 1.3584724285589442,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004819881082912528,
      "loss": 5.3867,
      "mean_token_accuracy": 0.19014601707458495,
      "num_tokens": 31486998.0,
      "step": 15545
    },
    {
      "entropy": 5.918473815917968,
      "epoch": 1.3589093769116491,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048197588983451265,
      "loss": 5.4337,
      "mean_token_accuracy": 0.18486632108688356,
      "num_tokens": 31496674.0,
      "step": 15550
    },
    {
      "entropy": 5.875294399261475,
      "epoch": 1.3593463252643536,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048196366740786646,
      "loss": 5.3991,
      "mean_token_accuracy": 0.18855599015951158,
      "num_tokens": 31507412.0,
      "step": 15555
    },
    {
      "entropy": 5.904902458190918,
      "epoch": 1.3597832736170585,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048195144101154867,
      "loss": 5.3766,
      "mean_token_accuracy": 0.19173528701066972,
      "num_tokens": 31517048.0,
      "step": 15560
    },
    {
      "entropy": 5.913545417785644,
      "epoch": 1.3602202219697632,
      "grad_norm": 1.125,
      "learning_rate": 0.00048193921064579385,
      "loss": 5.4498,
      "mean_token_accuracy": 0.18410247266292573,
      "num_tokens": 31526353.0,
      "step": 15565
    },
    {
      "entropy": 5.895209074020386,
      "epoch": 1.360657170322468,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004819269763108365,
      "loss": 5.4488,
      "mean_token_accuracy": 0.186563640832901,
      "num_tokens": 31537194.0,
      "step": 15570
    },
    {
      "entropy": 5.951923799514771,
      "epoch": 1.3610941186751726,
      "grad_norm": 1.0,
      "learning_rate": 0.00048191473800691145,
      "loss": 5.3791,
      "mean_token_accuracy": 0.18535400480031966,
      "num_tokens": 31547896.0,
      "step": 15575
    },
    {
      "entropy": 5.86262583732605,
      "epoch": 1.3615310670278773,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004819024957342531,
      "loss": 5.3954,
      "mean_token_accuracy": 0.18460704535245895,
      "num_tokens": 31558750.0,
      "step": 15580
    },
    {
      "entropy": 5.861621141433716,
      "epoch": 1.361968015380582,
      "grad_norm": 1.0,
      "learning_rate": 0.00048189024949309655,
      "loss": 5.3537,
      "mean_token_accuracy": 0.19129656702280046,
      "num_tokens": 31568721.0,
      "step": 15585
    },
    {
      "entropy": 5.777478647232056,
      "epoch": 1.3624049637332867,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004818779992836766,
      "loss": 5.3432,
      "mean_token_accuracy": 0.1920260965824127,
      "num_tokens": 31579907.0,
      "step": 15590
    },
    {
      "entropy": 5.905076265335083,
      "epoch": 1.3628419120859914,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048186574510622823,
      "loss": 5.4361,
      "mean_token_accuracy": 0.18732618242502214,
      "num_tokens": 31590137.0,
      "step": 15595
    },
    {
      "entropy": 5.9349912166595455,
      "epoch": 1.363278860438696,
      "grad_norm": 1.0,
      "learning_rate": 0.00048185348696098653,
      "loss": 5.4454,
      "mean_token_accuracy": 0.18598635643720626,
      "num_tokens": 31599969.0,
      "step": 15600
    },
    {
      "entropy": 5.905179882049561,
      "epoch": 1.363715808791401,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004818412248481865,
      "loss": 5.4082,
      "mean_token_accuracy": 0.18852966874837876,
      "num_tokens": 31609951.0,
      "step": 15605
    },
    {
      "entropy": 5.927187871932984,
      "epoch": 1.3641527571441054,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048182895876806343,
      "loss": 5.334,
      "mean_token_accuracy": 0.19803985357284545,
      "num_tokens": 31619214.0,
      "step": 15610
    },
    {
      "entropy": 5.879168510437012,
      "epoch": 1.3645897054968104,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004818166887208524,
      "loss": 5.3673,
      "mean_token_accuracy": 0.19381660372018814,
      "num_tokens": 31629562.0,
      "step": 15615
    },
    {
      "entropy": 5.845631790161133,
      "epoch": 1.365026653849515,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004818044147067891,
      "loss": 5.3283,
      "mean_token_accuracy": 0.19436752647161484,
      "num_tokens": 31639801.0,
      "step": 15620
    },
    {
      "entropy": 5.957575416564941,
      "epoch": 1.3654636022022197,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048179213672610866,
      "loss": 5.4286,
      "mean_token_accuracy": 0.1835531622171402,
      "num_tokens": 31650593.0,
      "step": 15625
    },
    {
      "entropy": 5.887137699127197,
      "epoch": 1.3659005505549244,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048177985477904675,
      "loss": 5.378,
      "mean_token_accuracy": 0.18776555955410004,
      "num_tokens": 31660942.0,
      "step": 15630
    },
    {
      "entropy": 5.8918938636779785,
      "epoch": 1.3663374989076291,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048176756886583884,
      "loss": 5.4269,
      "mean_token_accuracy": 0.17932044565677643,
      "num_tokens": 31671105.0,
      "step": 15635
    },
    {
      "entropy": 5.911219835281372,
      "epoch": 1.3667744472603338,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004817552789867206,
      "loss": 5.3744,
      "mean_token_accuracy": 0.19333153069019318,
      "num_tokens": 31680696.0,
      "step": 15640
    },
    {
      "entropy": 5.808177947998047,
      "epoch": 1.3672113956130385,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048174298514192784,
      "loss": 5.3207,
      "mean_token_accuracy": 0.1962072566151619,
      "num_tokens": 31689755.0,
      "step": 15645
    },
    {
      "entropy": 5.80774416923523,
      "epoch": 1.3676483439657432,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048173068733169615,
      "loss": 5.341,
      "mean_token_accuracy": 0.1954355373978615,
      "num_tokens": 31699614.0,
      "step": 15650
    },
    {
      "entropy": 5.927312660217285,
      "epoch": 1.3680852923184479,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004817183855562617,
      "loss": 5.3377,
      "mean_token_accuracy": 0.19454119652509688,
      "num_tokens": 31709106.0,
      "step": 15655
    },
    {
      "entropy": 5.957083749771118,
      "epoch": 1.3685222406711528,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048170607981586023,
      "loss": 5.4849,
      "mean_token_accuracy": 0.18326641023159027,
      "num_tokens": 31719542.0,
      "step": 15660
    },
    {
      "entropy": 5.891589403152466,
      "epoch": 1.3689591890238573,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048169377011072784,
      "loss": 5.3682,
      "mean_token_accuracy": 0.19153203666210175,
      "num_tokens": 31730477.0,
      "step": 15665
    },
    {
      "entropy": 5.855697154998779,
      "epoch": 1.3693961373765622,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004816814564411006,
      "loss": 5.3547,
      "mean_token_accuracy": 0.19592152237892152,
      "num_tokens": 31740269.0,
      "step": 15670
    },
    {
      "entropy": 5.8853006839752195,
      "epoch": 1.3698330857292669,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004816691388072148,
      "loss": 5.3501,
      "mean_token_accuracy": 0.19129445105791093,
      "num_tokens": 31750153.0,
      "step": 15675
    },
    {
      "entropy": 5.963100242614746,
      "epoch": 1.3702700340819716,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048165681720930653,
      "loss": 5.4242,
      "mean_token_accuracy": 0.1881497710943222,
      "num_tokens": 31760171.0,
      "step": 15680
    },
    {
      "entropy": 5.865595531463623,
      "epoch": 1.3707069824346763,
      "grad_norm": 1.0,
      "learning_rate": 0.0004816444916476123,
      "loss": 5.4133,
      "mean_token_accuracy": 0.18289484828710556,
      "num_tokens": 31771601.0,
      "step": 15685
    },
    {
      "entropy": 5.893989706039429,
      "epoch": 1.371143930787381,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048163216212236844,
      "loss": 5.3206,
      "mean_token_accuracy": 0.19743732661008834,
      "num_tokens": 31781423.0,
      "step": 15690
    },
    {
      "entropy": 5.884299659729004,
      "epoch": 1.3715808791400856,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004816198286338115,
      "loss": 5.2662,
      "mean_token_accuracy": 0.2019323080778122,
      "num_tokens": 31791025.0,
      "step": 15695
    },
    {
      "entropy": 5.8284162998199465,
      "epoch": 1.3720178274927903,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048160749118217783,
      "loss": 5.4019,
      "mean_token_accuracy": 0.1864023268222809,
      "num_tokens": 31801125.0,
      "step": 15700
    },
    {
      "entropy": 5.87870683670044,
      "epoch": 1.372454775845495,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004815951497677043,
      "loss": 5.3952,
      "mean_token_accuracy": 0.1840527430176735,
      "num_tokens": 31811100.0,
      "step": 15705
    },
    {
      "entropy": 5.919846296310425,
      "epoch": 1.3728917241981997,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004815828043906275,
      "loss": 5.4065,
      "mean_token_accuracy": 0.17994038313627242,
      "num_tokens": 31821667.0,
      "step": 15710
    },
    {
      "entropy": 5.900341796875,
      "epoch": 1.3733286725509044,
      "grad_norm": 1.0,
      "learning_rate": 0.0004815704550511843,
      "loss": 5.3485,
      "mean_token_accuracy": 0.19399411976337433,
      "num_tokens": 31831859.0,
      "step": 15715
    },
    {
      "entropy": 5.903557395935058,
      "epoch": 1.373765620903609,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004815581017496115,
      "loss": 5.3381,
      "mean_token_accuracy": 0.18635916262865065,
      "num_tokens": 31842952.0,
      "step": 15720
    },
    {
      "entropy": 5.818217611312866,
      "epoch": 1.374202569256314,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048154574448614607,
      "loss": 5.4493,
      "mean_token_accuracy": 0.1915601298213005,
      "num_tokens": 31852869.0,
      "step": 15725
    },
    {
      "entropy": 5.900901508331299,
      "epoch": 1.3746395176090185,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004815333832610251,
      "loss": 5.3557,
      "mean_token_accuracy": 0.19757130593061448,
      "num_tokens": 31862590.0,
      "step": 15730
    },
    {
      "entropy": 5.888856267929077,
      "epoch": 1.3750764659617234,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048152101807448554,
      "loss": 5.3951,
      "mean_token_accuracy": 0.19623784124851226,
      "num_tokens": 31873071.0,
      "step": 15735
    },
    {
      "entropy": 5.870918560028076,
      "epoch": 1.375513414314428,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004815086489267647,
      "loss": 5.3965,
      "mean_token_accuracy": 0.19107567965984346,
      "num_tokens": 31883325.0,
      "step": 15740
    },
    {
      "entropy": 5.879379796981811,
      "epoch": 1.3759503626671328,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048149627581809974,
      "loss": 5.4044,
      "mean_token_accuracy": 0.1909478172659874,
      "num_tokens": 31893071.0,
      "step": 15745
    },
    {
      "entropy": 5.8621784210205075,
      "epoch": 1.3763873110198375,
      "grad_norm": 1.015625,
      "learning_rate": 0.000481483898748728,
      "loss": 5.3599,
      "mean_token_accuracy": 0.19353826344013214,
      "num_tokens": 31902948.0,
      "step": 15750
    },
    {
      "entropy": 5.90810923576355,
      "epoch": 1.3768242593725422,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004814715177188869,
      "loss": 5.3895,
      "mean_token_accuracy": 0.18683548420667648,
      "num_tokens": 31913425.0,
      "step": 15755
    },
    {
      "entropy": 5.8360920429229735,
      "epoch": 1.3772612077252468,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004814591327288139,
      "loss": 5.3865,
      "mean_token_accuracy": 0.19077949076890946,
      "num_tokens": 31923608.0,
      "step": 15760
    },
    {
      "entropy": 5.843848323822021,
      "epoch": 1.3776981560779515,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048144674377874656,
      "loss": 5.3218,
      "mean_token_accuracy": 0.1976539224386215,
      "num_tokens": 31932863.0,
      "step": 15765
    },
    {
      "entropy": 5.877580356597901,
      "epoch": 1.3781351044306562,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048143435086892244,
      "loss": 5.4,
      "mean_token_accuracy": 0.1887669548392296,
      "num_tokens": 31942425.0,
      "step": 15770
    },
    {
      "entropy": 5.912509393692017,
      "epoch": 1.378572052783361,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004814219539995793,
      "loss": 5.3881,
      "mean_token_accuracy": 0.1908213511109352,
      "num_tokens": 31952152.0,
      "step": 15775
    },
    {
      "entropy": 5.932145738601685,
      "epoch": 1.3790090011360658,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048140955317095503,
      "loss": 5.4218,
      "mean_token_accuracy": 0.1870438665151596,
      "num_tokens": 31962224.0,
      "step": 15780
    },
    {
      "entropy": 5.871249389648438,
      "epoch": 1.3794459494887703,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004813971483832873,
      "loss": 5.3524,
      "mean_token_accuracy": 0.19505152404308318,
      "num_tokens": 31971517.0,
      "step": 15785
    },
    {
      "entropy": 5.894184923171997,
      "epoch": 1.3798828978414752,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004813847396368141,
      "loss": 5.4132,
      "mean_token_accuracy": 0.19147757142782212,
      "num_tokens": 31982461.0,
      "step": 15790
    },
    {
      "entropy": 5.87084813117981,
      "epoch": 1.38031984619418,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004813723269317734,
      "loss": 5.3004,
      "mean_token_accuracy": 0.20205197930336,
      "num_tokens": 31992202.0,
      "step": 15795
    },
    {
      "entropy": 5.899134922027588,
      "epoch": 1.3807567945468846,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004813599102684034,
      "loss": 5.3834,
      "mean_token_accuracy": 0.19240363240242003,
      "num_tokens": 32002283.0,
      "step": 15800
    },
    {
      "entropy": 5.87912335395813,
      "epoch": 1.3811937428995893,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048134748964694227,
      "loss": 5.3613,
      "mean_token_accuracy": 0.18643795400857927,
      "num_tokens": 32013135.0,
      "step": 15805
    },
    {
      "entropy": 5.893652391433716,
      "epoch": 1.381630691252294,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004813350650676281,
      "loss": 5.3927,
      "mean_token_accuracy": 0.1844712406396866,
      "num_tokens": 32022631.0,
      "step": 15810
    },
    {
      "entropy": 5.935748863220215,
      "epoch": 1.3820676396049987,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004813226365306992,
      "loss": 5.4001,
      "mean_token_accuracy": 0.18510207682847976,
      "num_tokens": 32032216.0,
      "step": 15815
    },
    {
      "entropy": 5.872505950927734,
      "epoch": 1.3825045879577034,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048131020403639405,
      "loss": 5.4016,
      "mean_token_accuracy": 0.1958578422665596,
      "num_tokens": 32042382.0,
      "step": 15820
    },
    {
      "entropy": 5.868051767349243,
      "epoch": 1.382941536310408,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048129776758495116,
      "loss": 5.3638,
      "mean_token_accuracy": 0.19156569391489028,
      "num_tokens": 32052592.0,
      "step": 15825
    },
    {
      "entropy": 5.865209484100342,
      "epoch": 1.3833784846631128,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004812853271766089,
      "loss": 5.2997,
      "mean_token_accuracy": 0.19482193142175674,
      "num_tokens": 32062260.0,
      "step": 15830
    },
    {
      "entropy": 5.931509494781494,
      "epoch": 1.3838154330158177,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000481272882811606,
      "loss": 5.3852,
      "mean_token_accuracy": 0.19716639071702957,
      "num_tokens": 32072079.0,
      "step": 15835
    },
    {
      "entropy": 5.859059810638428,
      "epoch": 1.3842523813685221,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048126043449018115,
      "loss": 5.3936,
      "mean_token_accuracy": 0.1935328558087349,
      "num_tokens": 32082151.0,
      "step": 15840
    },
    {
      "entropy": 5.947337913513183,
      "epoch": 1.384689329721227,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000481247982212573,
      "loss": 5.377,
      "mean_token_accuracy": 0.1966334581375122,
      "num_tokens": 32091893.0,
      "step": 15845
    },
    {
      "entropy": 5.899599218368531,
      "epoch": 1.3851262780739317,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048123552597902053,
      "loss": 5.3514,
      "mean_token_accuracy": 0.1939542844891548,
      "num_tokens": 32101550.0,
      "step": 15850
    },
    {
      "entropy": 5.763808536529541,
      "epoch": 1.3855632264266364,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048122306578976266,
      "loss": 5.2588,
      "mean_token_accuracy": 0.20597356408834458,
      "num_tokens": 32112278.0,
      "step": 15855
    },
    {
      "entropy": 5.924475240707397,
      "epoch": 1.3860001747793411,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048121060164503835,
      "loss": 5.4344,
      "mean_token_accuracy": 0.18299015909433364,
      "num_tokens": 32122284.0,
      "step": 15860
    },
    {
      "entropy": 5.9628157138824465,
      "epoch": 1.3864371231320458,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004811981335450865,
      "loss": 5.4033,
      "mean_token_accuracy": 0.18859420269727706,
      "num_tokens": 32131552.0,
      "step": 15865
    },
    {
      "entropy": 5.9957630157470705,
      "epoch": 1.3868740714847505,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048118566149014653,
      "loss": 5.4736,
      "mean_token_accuracy": 0.18229355663061142,
      "num_tokens": 32141758.0,
      "step": 15870
    },
    {
      "entropy": 5.880095195770264,
      "epoch": 1.3873110198374552,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004811731854804575,
      "loss": 5.3612,
      "mean_token_accuracy": 0.18894474804401398,
      "num_tokens": 32150679.0,
      "step": 15875
    },
    {
      "entropy": 5.795073556900024,
      "epoch": 1.3877479681901599,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004811607055162587,
      "loss": 5.326,
      "mean_token_accuracy": 0.19227624535560608,
      "num_tokens": 32160967.0,
      "step": 15880
    },
    {
      "entropy": 5.959837341308594,
      "epoch": 1.3881849165428646,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004811482215977895,
      "loss": 5.4237,
      "mean_token_accuracy": 0.1832110807299614,
      "num_tokens": 32170516.0,
      "step": 15885
    },
    {
      "entropy": 5.912219285964966,
      "epoch": 1.3886218648955693,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048113573372528956,
      "loss": 5.4141,
      "mean_token_accuracy": 0.18904127329587936,
      "num_tokens": 32180741.0,
      "step": 15890
    },
    {
      "entropy": 5.919985628128051,
      "epoch": 1.389058813248274,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004811232418989981,
      "loss": 5.4237,
      "mean_token_accuracy": 0.19147088378667831,
      "num_tokens": 32190646.0,
      "step": 15895
    },
    {
      "entropy": 5.839835691452026,
      "epoch": 1.3894957616009789,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048111074611915483,
      "loss": 5.3717,
      "mean_token_accuracy": 0.1885887548327446,
      "num_tokens": 32201361.0,
      "step": 15900
    },
    {
      "entropy": 5.8916247367858885,
      "epoch": 1.3899327099536833,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004810982463859996,
      "loss": 5.4779,
      "mean_token_accuracy": 0.1750300019979477,
      "num_tokens": 32211759.0,
      "step": 15905
    },
    {
      "entropy": 5.937699890136718,
      "epoch": 1.3903696583063883,
      "grad_norm": 1.0,
      "learning_rate": 0.0004810857426997719,
      "loss": 5.441,
      "mean_token_accuracy": 0.18625663071870804,
      "num_tokens": 32222582.0,
      "step": 15910
    },
    {
      "entropy": 5.913283014297486,
      "epoch": 1.390806606659093,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048107323506071167,
      "loss": 5.4024,
      "mean_token_accuracy": 0.19173415303230285,
      "num_tokens": 32232936.0,
      "step": 15915
    },
    {
      "entropy": 5.883814764022827,
      "epoch": 1.3912435550117976,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048106072346905875,
      "loss": 5.4033,
      "mean_token_accuracy": 0.18612465560436248,
      "num_tokens": 32242392.0,
      "step": 15920
    },
    {
      "entropy": 5.931699657440186,
      "epoch": 1.3916805033645023,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048104820792505335,
      "loss": 5.4762,
      "mean_token_accuracy": 0.1815352663397789,
      "num_tokens": 32252385.0,
      "step": 15925
    },
    {
      "entropy": 5.916854047775269,
      "epoch": 1.392117451717207,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048103568842893527,
      "loss": 5.4197,
      "mean_token_accuracy": 0.1823353260755539,
      "num_tokens": 32263537.0,
      "step": 15930
    },
    {
      "entropy": 5.8919822692871096,
      "epoch": 1.3925544000699117,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048102316498094467,
      "loss": 5.4072,
      "mean_token_accuracy": 0.18695046305656432,
      "num_tokens": 32273450.0,
      "step": 15935
    },
    {
      "entropy": 5.925897455215454,
      "epoch": 1.3929913484226164,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048101063758132186,
      "loss": 5.3942,
      "mean_token_accuracy": 0.18871243447065353,
      "num_tokens": 32283243.0,
      "step": 15940
    },
    {
      "entropy": 5.842831993103028,
      "epoch": 1.393428296775321,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004809981062303071,
      "loss": 5.3329,
      "mean_token_accuracy": 0.19150988906621932,
      "num_tokens": 32292458.0,
      "step": 15945
    },
    {
      "entropy": 5.882598876953125,
      "epoch": 1.3938652451280258,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048098557092814066,
      "loss": 5.3501,
      "mean_token_accuracy": 0.19431083500385285,
      "num_tokens": 32302574.0,
      "step": 15950
    },
    {
      "entropy": 5.848750352859497,
      "epoch": 1.3943021934807307,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000480973031675063,
      "loss": 5.3582,
      "mean_token_accuracy": 0.19579221457242965,
      "num_tokens": 32312658.0,
      "step": 15955
    },
    {
      "entropy": 5.917879056930542,
      "epoch": 1.3947391418334352,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004809604884713148,
      "loss": 5.4425,
      "mean_token_accuracy": 0.18741825670003892,
      "num_tokens": 32322717.0,
      "step": 15960
    },
    {
      "entropy": 5.857727813720703,
      "epoch": 1.39517609018614,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048094794131713645,
      "loss": 5.3474,
      "mean_token_accuracy": 0.1959444046020508,
      "num_tokens": 32333316.0,
      "step": 15965
    },
    {
      "entropy": 5.895614814758301,
      "epoch": 1.3956130385388448,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004809353902127686,
      "loss": 5.3675,
      "mean_token_accuracy": 0.1902415081858635,
      "num_tokens": 32343857.0,
      "step": 15970
    },
    {
      "entropy": 5.869587469100952,
      "epoch": 1.3960499868915495,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048092283515845216,
      "loss": 5.2825,
      "mean_token_accuracy": 0.1976398006081581,
      "num_tokens": 32353883.0,
      "step": 15975
    },
    {
      "entropy": 5.812234926223755,
      "epoch": 1.3964869352442542,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004809102761544278,
      "loss": 5.3038,
      "mean_token_accuracy": 0.19045919328927993,
      "num_tokens": 32363355.0,
      "step": 15980
    },
    {
      "entropy": 5.88984751701355,
      "epoch": 1.3969238835969588,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004808977132009364,
      "loss": 5.3788,
      "mean_token_accuracy": 0.1875956490635872,
      "num_tokens": 32373066.0,
      "step": 15985
    },
    {
      "entropy": 5.918769693374633,
      "epoch": 1.3973608319496635,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000480885146298219,
      "loss": 5.3617,
      "mean_token_accuracy": 0.19586149156093596,
      "num_tokens": 32383720.0,
      "step": 15990
    },
    {
      "entropy": 5.920205450057983,
      "epoch": 1.3977977803023682,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004808725754465167,
      "loss": 5.367,
      "mean_token_accuracy": 0.18658673018217087,
      "num_tokens": 32393823.0,
      "step": 15995
    },
    {
      "entropy": 5.771985960006714,
      "epoch": 1.398234728655073,
      "grad_norm": 1.125,
      "learning_rate": 0.00048086000064607044,
      "loss": 5.3252,
      "mean_token_accuracy": 0.1949838399887085,
      "num_tokens": 32404745.0,
      "step": 16000
    },
    {
      "entropy": 5.862604665756225,
      "epoch": 1.3986716770077776,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048084742189712147,
      "loss": 5.3476,
      "mean_token_accuracy": 0.20106639117002487,
      "num_tokens": 32415325.0,
      "step": 16005
    },
    {
      "entropy": 5.929654884338379,
      "epoch": 1.3991086253604825,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004808348391999112,
      "loss": 5.4255,
      "mean_token_accuracy": 0.1893130749464035,
      "num_tokens": 32425810.0,
      "step": 16010
    },
    {
      "entropy": 5.905332946777344,
      "epoch": 1.399545573713187,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004808222525546808,
      "loss": 5.4056,
      "mean_token_accuracy": 0.18739503026008605,
      "num_tokens": 32436799.0,
      "step": 16015
    },
    {
      "entropy": 5.87523455619812,
      "epoch": 1.399982522065892,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048080966196167163,
      "loss": 5.3353,
      "mean_token_accuracy": 0.19179599732160568,
      "num_tokens": 32446409.0,
      "step": 16020
    },
    {
      "entropy": 5.9264532089233395,
      "epoch": 1.4004194704185966,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004807970674211255,
      "loss": 5.3968,
      "mean_token_accuracy": 0.1849864348769188,
      "num_tokens": 32456268.0,
      "step": 16025
    },
    {
      "entropy": 5.837501668930054,
      "epoch": 1.4008564187713013,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048078446893328364,
      "loss": 5.4746,
      "mean_token_accuracy": 0.18347837924957275,
      "num_tokens": 32467622.0,
      "step": 16030
    },
    {
      "entropy": 5.929563760757446,
      "epoch": 1.401293367124006,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004807718664983879,
      "loss": 5.4025,
      "mean_token_accuracy": 0.19031818062067032,
      "num_tokens": 32477753.0,
      "step": 16035
    },
    {
      "entropy": 5.912538337707519,
      "epoch": 1.4017303154767107,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048075926011667997,
      "loss": 5.2926,
      "mean_token_accuracy": 0.19888316094875336,
      "num_tokens": 32487758.0,
      "step": 16040
    },
    {
      "entropy": 5.857427072525025,
      "epoch": 1.4021672638294154,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048074664978840147,
      "loss": 5.3333,
      "mean_token_accuracy": 0.19584411084651948,
      "num_tokens": 32497725.0,
      "step": 16045
    },
    {
      "entropy": 5.853944873809814,
      "epoch": 1.40260421218212,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004807340355137945,
      "loss": 5.2597,
      "mean_token_accuracy": 0.20242150425910949,
      "num_tokens": 32507515.0,
      "step": 16050
    },
    {
      "entropy": 5.854101228713989,
      "epoch": 1.4030411605348247,
      "grad_norm": 1.046875,
      "learning_rate": 0.000480721417293101,
      "loss": 5.3224,
      "mean_token_accuracy": 0.19746222645044326,
      "num_tokens": 32516943.0,
      "step": 16055
    },
    {
      "entropy": 5.805536413192749,
      "epoch": 1.4034781088875294,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004807087951265628,
      "loss": 5.2496,
      "mean_token_accuracy": 0.20551445335149765,
      "num_tokens": 32527223.0,
      "step": 16060
    },
    {
      "entropy": 5.859762716293335,
      "epoch": 1.4039150572402341,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004806961690144222,
      "loss": 5.3542,
      "mean_token_accuracy": 0.19369794875383378,
      "num_tokens": 32536668.0,
      "step": 16065
    },
    {
      "entropy": 5.892323160171509,
      "epoch": 1.4043520055929388,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004806835389569213,
      "loss": 5.3769,
      "mean_token_accuracy": 0.19433140903711318,
      "num_tokens": 32546566.0,
      "step": 16070
    },
    {
      "entropy": 5.849116611480713,
      "epoch": 1.4047889539456437,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048067090495430235,
      "loss": 5.2825,
      "mean_token_accuracy": 0.19675515294075013,
      "num_tokens": 32555927.0,
      "step": 16075
    },
    {
      "entropy": 5.869230556488037,
      "epoch": 1.4052259022983482,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048065826700680763,
      "loss": 5.3326,
      "mean_token_accuracy": 0.18960774689912796,
      "num_tokens": 32566569.0,
      "step": 16080
    },
    {
      "entropy": 5.826457405090332,
      "epoch": 1.4056628506510531,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048064562511467956,
      "loss": 5.3049,
      "mean_token_accuracy": 0.20044668018817902,
      "num_tokens": 32577076.0,
      "step": 16085
    },
    {
      "entropy": 5.863542366027832,
      "epoch": 1.4060997990037578,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048063297927816067,
      "loss": 5.4,
      "mean_token_accuracy": 0.18576691448688507,
      "num_tokens": 32587926.0,
      "step": 16090
    },
    {
      "entropy": 5.910733556747436,
      "epoch": 1.4065367473564625,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004806203294974935,
      "loss": 5.4332,
      "mean_token_accuracy": 0.19269084483385085,
      "num_tokens": 32599012.0,
      "step": 16095
    },
    {
      "entropy": 5.9569957733154295,
      "epoch": 1.4069736957091672,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048060767577292067,
      "loss": 5.3363,
      "mean_token_accuracy": 0.19091223329305648,
      "num_tokens": 32609094.0,
      "step": 16100
    },
    {
      "entropy": 5.855603265762329,
      "epoch": 1.4074106440618719,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004805950181046849,
      "loss": 5.3993,
      "mean_token_accuracy": 0.18990612179040908,
      "num_tokens": 32620356.0,
      "step": 16105
    },
    {
      "entropy": 5.90706238746643,
      "epoch": 1.4078475924145766,
      "grad_norm": 1.078125,
      "learning_rate": 0.000480582356493029,
      "loss": 5.3834,
      "mean_token_accuracy": 0.18922373950481414,
      "num_tokens": 32629841.0,
      "step": 16110
    },
    {
      "entropy": 5.9127432346344,
      "epoch": 1.4082845407672813,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004805696909381957,
      "loss": 5.4247,
      "mean_token_accuracy": 0.1879512995481491,
      "num_tokens": 32639006.0,
      "step": 16115
    },
    {
      "entropy": 5.897290420532227,
      "epoch": 1.408721489119986,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048055702144042794,
      "loss": 5.409,
      "mean_token_accuracy": 0.18783899396657944,
      "num_tokens": 32648419.0,
      "step": 16120
    },
    {
      "entropy": 5.8655084609985355,
      "epoch": 1.4091584374726907,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004805443479999689,
      "loss": 5.4022,
      "mean_token_accuracy": 0.19718537330627442,
      "num_tokens": 32658050.0,
      "step": 16125
    },
    {
      "entropy": 5.8802728176116945,
      "epoch": 1.4095953858253956,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004805316706170615,
      "loss": 5.3739,
      "mean_token_accuracy": 0.19416674375534057,
      "num_tokens": 32668176.0,
      "step": 16130
    },
    {
      "entropy": 5.825069904327393,
      "epoch": 1.4100323341781,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048051898929194907,
      "loss": 5.3089,
      "mean_token_accuracy": 0.19709164649248123,
      "num_tokens": 32678517.0,
      "step": 16135
    },
    {
      "entropy": 5.909453725814819,
      "epoch": 1.410469282530805,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004805063040248747,
      "loss": 5.3538,
      "mean_token_accuracy": 0.19456642419099807,
      "num_tokens": 32688459.0,
      "step": 16140
    },
    {
      "entropy": 5.904028797149659,
      "epoch": 1.4109062308835096,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004804936148160817,
      "loss": 5.3527,
      "mean_token_accuracy": 0.19631440490484237,
      "num_tokens": 32698481.0,
      "step": 16145
    },
    {
      "entropy": 5.839969444274902,
      "epoch": 1.4113431792362143,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004804809216658135,
      "loss": 5.3921,
      "mean_token_accuracy": 0.18698856234550476,
      "num_tokens": 32709031.0,
      "step": 16150
    },
    {
      "entropy": 5.821101093292237,
      "epoch": 1.411780127588919,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048046822457431347,
      "loss": 5.3501,
      "mean_token_accuracy": 0.1989353895187378,
      "num_tokens": 32720293.0,
      "step": 16155
    },
    {
      "entropy": 5.908024740219116,
      "epoch": 1.4122170759416237,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048045552354182536,
      "loss": 5.3966,
      "mean_token_accuracy": 0.1914379343390465,
      "num_tokens": 32730108.0,
      "step": 16160
    },
    {
      "entropy": 5.909698057174682,
      "epoch": 1.4126540242943284,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048044281856859255,
      "loss": 5.3912,
      "mean_token_accuracy": 0.1911873012781143,
      "num_tokens": 32740614.0,
      "step": 16165
    },
    {
      "entropy": 5.915780305862427,
      "epoch": 1.413090972647033,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004804301096548589,
      "loss": 5.3626,
      "mean_token_accuracy": 0.19116658121347427,
      "num_tokens": 32751155.0,
      "step": 16170
    },
    {
      "entropy": 5.847481966018677,
      "epoch": 1.4135279209997378,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004804173968008681,
      "loss": 5.2543,
      "mean_token_accuracy": 0.20120693296194075,
      "num_tokens": 32761067.0,
      "step": 16175
    },
    {
      "entropy": 5.763520622253418,
      "epoch": 1.4139648693524425,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048040468000686395,
      "loss": 5.2485,
      "mean_token_accuracy": 0.20808626264333724,
      "num_tokens": 32771357.0,
      "step": 16180
    },
    {
      "entropy": 5.905440378189087,
      "epoch": 1.4144018177051472,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004803919592730904,
      "loss": 5.422,
      "mean_token_accuracy": 0.19150423556566237,
      "num_tokens": 32781116.0,
      "step": 16185
    },
    {
      "entropy": 5.919905090332032,
      "epoch": 1.4148387660578519,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004803792345997915,
      "loss": 5.3974,
      "mean_token_accuracy": 0.19152844101190566,
      "num_tokens": 32791660.0,
      "step": 16190
    },
    {
      "entropy": 5.843469095230103,
      "epoch": 1.4152757144105568,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048036650598721125,
      "loss": 5.365,
      "mean_token_accuracy": 0.18820559829473496,
      "num_tokens": 32801413.0,
      "step": 16195
    },
    {
      "entropy": 5.912697744369507,
      "epoch": 1.4157126627632615,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048035377343559377,
      "loss": 5.428,
      "mean_token_accuracy": 0.18568021655082703,
      "num_tokens": 32811293.0,
      "step": 16200
    },
    {
      "entropy": 5.89748067855835,
      "epoch": 1.4161496111159662,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048034103694518336,
      "loss": 5.3835,
      "mean_token_accuracy": 0.1950879141688347,
      "num_tokens": 32821609.0,
      "step": 16205
    },
    {
      "entropy": 5.8838451385498045,
      "epoch": 1.4165865594686708,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048032829651622425,
      "loss": 5.3513,
      "mean_token_accuracy": 0.19342914074659348,
      "num_tokens": 32831219.0,
      "step": 16210
    },
    {
      "entropy": 5.90845251083374,
      "epoch": 1.4170235078213755,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048031555214896076,
      "loss": 5.3988,
      "mean_token_accuracy": 0.18542817384004592,
      "num_tokens": 32841085.0,
      "step": 16215
    },
    {
      "entropy": 5.91574649810791,
      "epoch": 1.4174604561740802,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004803028038436375,
      "loss": 5.3429,
      "mean_token_accuracy": 0.1974698543548584,
      "num_tokens": 32851592.0,
      "step": 16220
    },
    {
      "entropy": 5.912979602813721,
      "epoch": 1.417897404526785,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048029005160049886,
      "loss": 5.3513,
      "mean_token_accuracy": 0.18902435302734374,
      "num_tokens": 32862124.0,
      "step": 16225
    },
    {
      "entropy": 5.7908836841583256,
      "epoch": 1.4183343528794896,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004802772954197894,
      "loss": 5.3397,
      "mean_token_accuracy": 0.19746158868074418,
      "num_tokens": 32872760.0,
      "step": 16230
    },
    {
      "entropy": 5.88410210609436,
      "epoch": 1.4187713012321943,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048026453530175393,
      "loss": 5.4232,
      "mean_token_accuracy": 0.1877978503704071,
      "num_tokens": 32883542.0,
      "step": 16235
    },
    {
      "entropy": 5.874264287948608,
      "epoch": 1.419208249584899,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048025177124663704,
      "loss": 5.3422,
      "mean_token_accuracy": 0.19256283789873124,
      "num_tokens": 32893922.0,
      "step": 16240
    },
    {
      "entropy": 5.896006774902344,
      "epoch": 1.4196451979376037,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048023900325468367,
      "loss": 5.4146,
      "mean_token_accuracy": 0.19137748181819916,
      "num_tokens": 32903907.0,
      "step": 16245
    },
    {
      "entropy": 5.91215295791626,
      "epoch": 1.4200821462903086,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048022623132613863,
      "loss": 5.3802,
      "mean_token_accuracy": 0.19023822546005248,
      "num_tokens": 32914656.0,
      "step": 16250
    },
    {
      "entropy": 5.904300594329834,
      "epoch": 1.420519094643013,
      "grad_norm": 1.03125,
      "learning_rate": 0.000480213455461247,
      "loss": 5.3973,
      "mean_token_accuracy": 0.18687004148960112,
      "num_tokens": 32925250.0,
      "step": 16255
    },
    {
      "entropy": 5.87320909500122,
      "epoch": 1.420956042995718,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048020067566025373,
      "loss": 5.3863,
      "mean_token_accuracy": 0.19277166128158568,
      "num_tokens": 32935435.0,
      "step": 16260
    },
    {
      "entropy": 5.897303199768066,
      "epoch": 1.4213929913484227,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000480187891923404,
      "loss": 5.3185,
      "mean_token_accuracy": 0.1948848322033882,
      "num_tokens": 32946166.0,
      "step": 16265
    },
    {
      "entropy": 5.848087930679322,
      "epoch": 1.4218299397011274,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004801751042509429,
      "loss": 5.32,
      "mean_token_accuracy": 0.19442242234945298,
      "num_tokens": 32955751.0,
      "step": 16270
    },
    {
      "entropy": 5.8773698806762695,
      "epoch": 1.422266888053832,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004801623126431159,
      "loss": 5.4021,
      "mean_token_accuracy": 0.19028325825929643,
      "num_tokens": 32966121.0,
      "step": 16275
    },
    {
      "entropy": 5.822104549407959,
      "epoch": 1.4227038364065367,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004801495171001681,
      "loss": 5.3545,
      "mean_token_accuracy": 0.19969561249017714,
      "num_tokens": 32976950.0,
      "step": 16280
    },
    {
      "entropy": 5.888471221923828,
      "epoch": 1.4231407847592414,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004801367176223452,
      "loss": 5.3517,
      "mean_token_accuracy": 0.19630062878131865,
      "num_tokens": 32987278.0,
      "step": 16285
    },
    {
      "entropy": 5.918195199966431,
      "epoch": 1.4235777331119461,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004801239142098926,
      "loss": 5.4074,
      "mean_token_accuracy": 0.19031696915626525,
      "num_tokens": 32997050.0,
      "step": 16290
    },
    {
      "entropy": 5.916104936599732,
      "epoch": 1.4240146814646508,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004801111068630557,
      "loss": 5.3749,
      "mean_token_accuracy": 0.18942983597517013,
      "num_tokens": 33007799.0,
      "step": 16295
    },
    {
      "entropy": 5.9701337814331055,
      "epoch": 1.4244516298173555,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048009829558208044,
      "loss": 5.3733,
      "mean_token_accuracy": 0.19611490666866302,
      "num_tokens": 33017476.0,
      "step": 16300
    },
    {
      "entropy": 5.878015184402466,
      "epoch": 1.4248885781700604,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048008548036721226,
      "loss": 5.3359,
      "mean_token_accuracy": 0.19428729712963105,
      "num_tokens": 33027617.0,
      "step": 16305
    },
    {
      "entropy": 5.839101934432984,
      "epoch": 1.425325526522765,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004800726612186972,
      "loss": 5.367,
      "mean_token_accuracy": 0.19354716539382935,
      "num_tokens": 33037257.0,
      "step": 16310
    },
    {
      "entropy": 5.821336269378662,
      "epoch": 1.4257624748754698,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004800598381367811,
      "loss": 5.3249,
      "mean_token_accuracy": 0.19702288806438445,
      "num_tokens": 33047406.0,
      "step": 16315
    },
    {
      "entropy": 5.869730758666992,
      "epoch": 1.4261994232281745,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004800470111217096,
      "loss": 5.3767,
      "mean_token_accuracy": 0.19729204028844832,
      "num_tokens": 33057975.0,
      "step": 16320
    },
    {
      "entropy": 5.942981863021851,
      "epoch": 1.4266363715808792,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048003418017372924,
      "loss": 5.4166,
      "mean_token_accuracy": 0.18901972621679305,
      "num_tokens": 33068645.0,
      "step": 16325
    },
    {
      "entropy": 5.930734777450562,
      "epoch": 1.4270733199335839,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004800213452930857,
      "loss": 5.3898,
      "mean_token_accuracy": 0.1900014564394951,
      "num_tokens": 33079147.0,
      "step": 16330
    },
    {
      "entropy": 5.880298137664795,
      "epoch": 1.4275102682862886,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004800085064800255,
      "loss": 5.3278,
      "mean_token_accuracy": 0.1896761476993561,
      "num_tokens": 33089229.0,
      "step": 16335
    },
    {
      "entropy": 5.87848687171936,
      "epoch": 1.4279472166389933,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004799956637347946,
      "loss": 5.3516,
      "mean_token_accuracy": 0.1926169514656067,
      "num_tokens": 33099688.0,
      "step": 16340
    },
    {
      "entropy": 5.877392911911011,
      "epoch": 1.428384164991698,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004799828170576395,
      "loss": 5.3921,
      "mean_token_accuracy": 0.19336025416851044,
      "num_tokens": 33110019.0,
      "step": 16345
    },
    {
      "entropy": 5.78552017211914,
      "epoch": 1.4288211133444026,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004799699664488065,
      "loss": 5.2616,
      "mean_token_accuracy": 0.20470748841762543,
      "num_tokens": 33120141.0,
      "step": 16350
    },
    {
      "entropy": 5.877651739120483,
      "epoch": 1.4292580616971073,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004799571119085422,
      "loss": 5.3548,
      "mean_token_accuracy": 0.19421638399362565,
      "num_tokens": 33130047.0,
      "step": 16355
    },
    {
      "entropy": 5.9096263408660885,
      "epoch": 1.429695010049812,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004799442534370931,
      "loss": 5.3947,
      "mean_token_accuracy": 0.18761702626943588,
      "num_tokens": 33141234.0,
      "step": 16360
    },
    {
      "entropy": 5.884143114089966,
      "epoch": 1.4301319584025167,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004799313910347057,
      "loss": 5.3827,
      "mean_token_accuracy": 0.2003782093524933,
      "num_tokens": 33150788.0,
      "step": 16365
    },
    {
      "entropy": 5.859095191955566,
      "epoch": 1.4305689067552216,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000479918524701627,
      "loss": 5.3491,
      "mean_token_accuracy": 0.20064250826835633,
      "num_tokens": 33160928.0,
      "step": 16370
    },
    {
      "entropy": 5.960884714126587,
      "epoch": 1.431005855107926,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047990565443810357,
      "loss": 5.4621,
      "mean_token_accuracy": 0.17971984893083573,
      "num_tokens": 33171333.0,
      "step": 16375
    },
    {
      "entropy": 5.795910215377807,
      "epoch": 1.431442803460631,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047989278024438233,
      "loss": 5.2589,
      "mean_token_accuracy": 0.2011488765478134,
      "num_tokens": 33180396.0,
      "step": 16380
    },
    {
      "entropy": 5.860289001464844,
      "epoch": 1.4318797518133357,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004798799021207102,
      "loss": 5.3345,
      "mean_token_accuracy": 0.19769225120544434,
      "num_tokens": 33189571.0,
      "step": 16385
    },
    {
      "entropy": 5.877098560333252,
      "epoch": 1.4323167001660404,
      "grad_norm": 1.125,
      "learning_rate": 0.00047986702006733417,
      "loss": 5.2993,
      "mean_token_accuracy": 0.19957837015390395,
      "num_tokens": 33198979.0,
      "step": 16390
    },
    {
      "entropy": 5.879195213317871,
      "epoch": 1.432753648518745,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047985413408450136,
      "loss": 5.3663,
      "mean_token_accuracy": 0.19856209605932235,
      "num_tokens": 33208862.0,
      "step": 16395
    },
    {
      "entropy": 5.848235321044922,
      "epoch": 1.4331905968714498,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004798412441724589,
      "loss": 5.3772,
      "mean_token_accuracy": 0.19045196920633317,
      "num_tokens": 33218804.0,
      "step": 16400
    },
    {
      "entropy": 5.90876874923706,
      "epoch": 1.4336275452241545,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047982835033145413,
      "loss": 5.3794,
      "mean_token_accuracy": 0.18670645207166672,
      "num_tokens": 33228783.0,
      "step": 16405
    },
    {
      "entropy": 5.955608749389649,
      "epoch": 1.4340644935768592,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047981545256173415,
      "loss": 5.4254,
      "mean_token_accuracy": 0.18686060905456542,
      "num_tokens": 33241029.0,
      "step": 16410
    },
    {
      "entropy": 5.947125005722046,
      "epoch": 1.4345014419295639,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004798025508635465,
      "loss": 5.4318,
      "mean_token_accuracy": 0.19017109721899034,
      "num_tokens": 33250383.0,
      "step": 16415
    },
    {
      "entropy": 5.776224565505982,
      "epoch": 1.4349383902822685,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004797896452371386,
      "loss": 5.2431,
      "mean_token_accuracy": 0.20624830722808837,
      "num_tokens": 33260424.0,
      "step": 16420
    },
    {
      "entropy": 5.905255699157715,
      "epoch": 1.4353753386349735,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047977673568275805,
      "loss": 5.3043,
      "mean_token_accuracy": 0.1901807814836502,
      "num_tokens": 33270200.0,
      "step": 16425
    },
    {
      "entropy": 5.9430379390716555,
      "epoch": 1.435812286987678,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004797638222006524,
      "loss": 5.4896,
      "mean_token_accuracy": 0.1862714871764183,
      "num_tokens": 33279850.0,
      "step": 16430
    },
    {
      "entropy": 5.845584440231323,
      "epoch": 1.4362492353403828,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004797509047910693,
      "loss": 5.3398,
      "mean_token_accuracy": 0.19635631740093232,
      "num_tokens": 33289053.0,
      "step": 16435
    },
    {
      "entropy": 5.891699504852295,
      "epoch": 1.4366861836930875,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047973798345425655,
      "loss": 5.3589,
      "mean_token_accuracy": 0.19078613966703414,
      "num_tokens": 33298988.0,
      "step": 16440
    },
    {
      "entropy": 5.888004636764526,
      "epoch": 1.4371231320457922,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047972505819046207,
      "loss": 5.3868,
      "mean_token_accuracy": 0.19635367691516875,
      "num_tokens": 33308683.0,
      "step": 16445
    },
    {
      "entropy": 5.830250883102417,
      "epoch": 1.437560080398497,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004797121289999336,
      "loss": 5.3498,
      "mean_token_accuracy": 0.18841940313577651,
      "num_tokens": 33319230.0,
      "step": 16450
    },
    {
      "entropy": 5.946500778198242,
      "epoch": 1.4379970287512016,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004796991958829193,
      "loss": 5.3876,
      "mean_token_accuracy": 0.1896067351102829,
      "num_tokens": 33329277.0,
      "step": 16455
    },
    {
      "entropy": 5.918843746185303,
      "epoch": 1.4384339771039063,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004796862588396672,
      "loss": 5.3583,
      "mean_token_accuracy": 0.19105735719203948,
      "num_tokens": 33338644.0,
      "step": 16460
    },
    {
      "entropy": 5.820420360565185,
      "epoch": 1.438870925456611,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047967331787042536,
      "loss": 5.2853,
      "mean_token_accuracy": 0.20011228024959565,
      "num_tokens": 33348996.0,
      "step": 16465
    },
    {
      "entropy": 5.900073575973511,
      "epoch": 1.4393078738093157,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047966037297544195,
      "loss": 5.3864,
      "mean_token_accuracy": 0.1897851213812828,
      "num_tokens": 33358581.0,
      "step": 16470
    },
    {
      "entropy": 5.931667137145996,
      "epoch": 1.4397448221620204,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047964742415496553,
      "loss": 5.4001,
      "mean_token_accuracy": 0.19026722609996796,
      "num_tokens": 33368925.0,
      "step": 16475
    },
    {
      "entropy": 5.8991883277893065,
      "epoch": 1.4401817705147253,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004796344714092441,
      "loss": 5.4077,
      "mean_token_accuracy": 0.19133184105157852,
      "num_tokens": 33378918.0,
      "step": 16480
    },
    {
      "entropy": 5.874284982681274,
      "epoch": 1.4406187188674298,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004796215147385264,
      "loss": 5.3563,
      "mean_token_accuracy": 0.1965407148003578,
      "num_tokens": 33389039.0,
      "step": 16485
    },
    {
      "entropy": 5.919140720367432,
      "epoch": 1.4410556672201347,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004796085541430608,
      "loss": 5.3205,
      "mean_token_accuracy": 0.19816785007715226,
      "num_tokens": 33398893.0,
      "step": 16490
    },
    {
      "entropy": 5.893600797653198,
      "epoch": 1.4414926155728394,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004795955896230959,
      "loss": 5.4099,
      "mean_token_accuracy": 0.1892208307981491,
      "num_tokens": 33408188.0,
      "step": 16495
    },
    {
      "entropy": 5.829650497436523,
      "epoch": 1.441929563925544,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047958262117888035,
      "loss": 5.3292,
      "mean_token_accuracy": 0.2016781449317932,
      "num_tokens": 33418926.0,
      "step": 16500
    },
    {
      "entropy": 5.9180787086486815,
      "epoch": 1.4423665122782487,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047956964881066297,
      "loss": 5.3818,
      "mean_token_accuracy": 0.19171223193407058,
      "num_tokens": 33429790.0,
      "step": 16505
    },
    {
      "entropy": 5.877075099945069,
      "epoch": 1.4428034606309534,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004795566725186925,
      "loss": 5.3828,
      "mean_token_accuracy": 0.1950563296675682,
      "num_tokens": 33440470.0,
      "step": 16510
    },
    {
      "entropy": 5.910140752792358,
      "epoch": 1.4432404089836581,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004795436923032179,
      "loss": 5.3672,
      "mean_token_accuracy": 0.19308619499206542,
      "num_tokens": 33449875.0,
      "step": 16515
    },
    {
      "entropy": 5.9273286819458,
      "epoch": 1.4436773573363628,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004795307081644881,
      "loss": 5.3413,
      "mean_token_accuracy": 0.18853641748428346,
      "num_tokens": 33458841.0,
      "step": 16520
    },
    {
      "entropy": 5.859242916107178,
      "epoch": 1.4441143056890675,
      "grad_norm": 1.125,
      "learning_rate": 0.0004795177201027521,
      "loss": 5.2946,
      "mean_token_accuracy": 0.19736545383930207,
      "num_tokens": 33468656.0,
      "step": 16525
    },
    {
      "entropy": 5.875080680847168,
      "epoch": 1.4445512540417722,
      "grad_norm": 1.0,
      "learning_rate": 0.00047950472811825916,
      "loss": 5.3564,
      "mean_token_accuracy": 0.19143709689378738,
      "num_tokens": 33478977.0,
      "step": 16530
    },
    {
      "entropy": 5.923673963546753,
      "epoch": 1.444988202394477,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004794917322112583,
      "loss": 5.3217,
      "mean_token_accuracy": 0.19242318123579025,
      "num_tokens": 33488418.0,
      "step": 16535
    },
    {
      "entropy": 5.827993297576905,
      "epoch": 1.4454251507471816,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004794787323819989,
      "loss": 5.3093,
      "mean_token_accuracy": 0.19972742050886155,
      "num_tokens": 33499125.0,
      "step": 16540
    },
    {
      "entropy": 5.873738050460815,
      "epoch": 1.4458620990998865,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047946572863073023,
      "loss": 5.3575,
      "mean_token_accuracy": 0.19021617174148558,
      "num_tokens": 33508826.0,
      "step": 16545
    },
    {
      "entropy": 5.824547910690308,
      "epoch": 1.446299047452591,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047945272095770174,
      "loss": 5.2701,
      "mean_token_accuracy": 0.20738946348428727,
      "num_tokens": 33518398.0,
      "step": 16550
    },
    {
      "entropy": 5.83190655708313,
      "epoch": 1.4467359958052959,
      "grad_norm": 1.046875,
      "learning_rate": 0.000479439709363163,
      "loss": 5.3417,
      "mean_token_accuracy": 0.19234306663274764,
      "num_tokens": 33528289.0,
      "step": 16555
    },
    {
      "entropy": 5.921662378311157,
      "epoch": 1.4471729441580006,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004794266938473635,
      "loss": 5.4395,
      "mean_token_accuracy": 0.1857820451259613,
      "num_tokens": 33539228.0,
      "step": 16560
    },
    {
      "entropy": 5.921424102783203,
      "epoch": 1.4476098925107053,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047941367441055283,
      "loss": 5.4264,
      "mean_token_accuracy": 0.19161428958177568,
      "num_tokens": 33551015.0,
      "step": 16565
    },
    {
      "entropy": 5.899002742767334,
      "epoch": 1.44804684086341,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047940065105298073,
      "loss": 5.3644,
      "mean_token_accuracy": 0.19469457864761353,
      "num_tokens": 33560374.0,
      "step": 16570
    },
    {
      "entropy": 5.798738479614258,
      "epoch": 1.4484837892161146,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004793876237748972,
      "loss": 5.2389,
      "mean_token_accuracy": 0.2090599790215492,
      "num_tokens": 33569916.0,
      "step": 16575
    },
    {
      "entropy": 5.764926767349243,
      "epoch": 1.4489207375688193,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004793745925765518,
      "loss": 5.2561,
      "mean_token_accuracy": 0.20401863902807235,
      "num_tokens": 33579930.0,
      "step": 16580
    },
    {
      "entropy": 5.842911958694458,
      "epoch": 1.449357685921524,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004793615574581947,
      "loss": 5.3767,
      "mean_token_accuracy": 0.19127658158540725,
      "num_tokens": 33588612.0,
      "step": 16585
    },
    {
      "entropy": 5.958376502990722,
      "epoch": 1.4497946342742287,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047934851842007585,
      "loss": 5.414,
      "mean_token_accuracy": 0.1859847530722618,
      "num_tokens": 33599253.0,
      "step": 16590
    },
    {
      "entropy": 5.913934326171875,
      "epoch": 1.4502315826269334,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004793354754624453,
      "loss": 5.2928,
      "mean_token_accuracy": 0.1969093933701515,
      "num_tokens": 33609484.0,
      "step": 16595
    },
    {
      "entropy": 5.9088832378387455,
      "epoch": 1.4506685309796383,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004793224285855533,
      "loss": 5.3011,
      "mean_token_accuracy": 0.19684510827064514,
      "num_tokens": 33619660.0,
      "step": 16600
    },
    {
      "entropy": 5.838850975036621,
      "epoch": 1.4511054793323428,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047930937778965,
      "loss": 5.4122,
      "mean_token_accuracy": 0.18864979594945908,
      "num_tokens": 33629505.0,
      "step": 16605
    },
    {
      "entropy": 5.87839937210083,
      "epoch": 1.4515424276850477,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004792963230749858,
      "loss": 5.4266,
      "mean_token_accuracy": 0.1921192854642868,
      "num_tokens": 33639680.0,
      "step": 16610
    },
    {
      "entropy": 5.945688390731812,
      "epoch": 1.4519793760377524,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047928326444181117,
      "loss": 5.4155,
      "mean_token_accuracy": 0.18601533025503159,
      "num_tokens": 33649200.0,
      "step": 16615
    },
    {
      "entropy": 5.891090440750122,
      "epoch": 1.452416324390457,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047927020189037634,
      "loss": 5.3557,
      "mean_token_accuracy": 0.18863812685012818,
      "num_tokens": 33658386.0,
      "step": 16620
    },
    {
      "entropy": 5.871538543701172,
      "epoch": 1.4528532727431618,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004792571354209321,
      "loss": 5.3164,
      "mean_token_accuracy": 0.1987977907061577,
      "num_tokens": 33669139.0,
      "step": 16625
    },
    {
      "entropy": 5.832577276229858,
      "epoch": 1.4532902210958665,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004792440650337289,
      "loss": 5.395,
      "mean_token_accuracy": 0.19593448489904403,
      "num_tokens": 33679629.0,
      "step": 16630
    },
    {
      "entropy": 5.9313633918762205,
      "epoch": 1.4537271694485712,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004792309907290175,
      "loss": 5.4006,
      "mean_token_accuracy": 0.18841576129198073,
      "num_tokens": 33689633.0,
      "step": 16635
    },
    {
      "entropy": 5.943309688568116,
      "epoch": 1.4541641178012759,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047921791250704875,
      "loss": 5.3659,
      "mean_token_accuracy": 0.1851710744202137,
      "num_tokens": 33699719.0,
      "step": 16640
    },
    {
      "entropy": 5.9048563003540036,
      "epoch": 1.4546010661539805,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047920483036807346,
      "loss": 5.4117,
      "mean_token_accuracy": 0.19227744042873382,
      "num_tokens": 33709389.0,
      "step": 16645
    },
    {
      "entropy": 5.899011898040771,
      "epoch": 1.4550380145066852,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047919174431234246,
      "loss": 5.4059,
      "mean_token_accuracy": 0.1871466279029846,
      "num_tokens": 33719063.0,
      "step": 16650
    },
    {
      "entropy": 5.969896268844605,
      "epoch": 1.4554749628593902,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004791786543401068,
      "loss": 5.4814,
      "mean_token_accuracy": 0.18160057365894317,
      "num_tokens": 33729798.0,
      "step": 16655
    },
    {
      "entropy": 5.912513399124146,
      "epoch": 1.4559119112120946,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004791655604516176,
      "loss": 5.3706,
      "mean_token_accuracy": 0.1842681124806404,
      "num_tokens": 33740012.0,
      "step": 16660
    },
    {
      "entropy": 5.903078937530518,
      "epoch": 1.4563488595647995,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047915246264712595,
      "loss": 5.3455,
      "mean_token_accuracy": 0.19351834505796434,
      "num_tokens": 33749478.0,
      "step": 16665
    },
    {
      "entropy": 5.7911216735839846,
      "epoch": 1.4567858079175042,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004791393609268831,
      "loss": 5.2311,
      "mean_token_accuracy": 0.2053156837821007,
      "num_tokens": 33758935.0,
      "step": 16670
    },
    {
      "entropy": 5.866981649398804,
      "epoch": 1.457222756270209,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004791262552911404,
      "loss": 5.3135,
      "mean_token_accuracy": 0.1955024406313896,
      "num_tokens": 33769035.0,
      "step": 16675
    },
    {
      "entropy": 5.877590799331665,
      "epoch": 1.4576597046229136,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047911314574014904,
      "loss": 5.3377,
      "mean_token_accuracy": 0.19487644582986832,
      "num_tokens": 33779288.0,
      "step": 16680
    },
    {
      "entropy": 5.902356767654419,
      "epoch": 1.4580966529756183,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004791000322741606,
      "loss": 5.3594,
      "mean_token_accuracy": 0.1927088439464569,
      "num_tokens": 33789138.0,
      "step": 16685
    },
    {
      "entropy": 5.936922931671143,
      "epoch": 1.458533601328323,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004790869148934267,
      "loss": 5.3806,
      "mean_token_accuracy": 0.18927017599344254,
      "num_tokens": 33798649.0,
      "step": 16690
    },
    {
      "entropy": 5.807880067825318,
      "epoch": 1.4589705496810277,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047907379359819877,
      "loss": 5.2577,
      "mean_token_accuracy": 0.20591161400079727,
      "num_tokens": 33808742.0,
      "step": 16695
    },
    {
      "entropy": 5.826742029190063,
      "epoch": 1.4594074980337324,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004790606683887286,
      "loss": 5.3041,
      "mean_token_accuracy": 0.19582216441631317,
      "num_tokens": 33819507.0,
      "step": 16700
    },
    {
      "entropy": 5.916469573974609,
      "epoch": 1.459844446386437,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047904753926526777,
      "loss": 5.3625,
      "mean_token_accuracy": 0.18949666917324065,
      "num_tokens": 33829722.0,
      "step": 16705
    },
    {
      "entropy": 5.909551048278809,
      "epoch": 1.4602813947391418,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004790344062280683,
      "loss": 5.3955,
      "mean_token_accuracy": 0.19379607439041138,
      "num_tokens": 33839350.0,
      "step": 16710
    },
    {
      "entropy": 5.855269813537598,
      "epoch": 1.4607183430918464,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047902126927738196,
      "loss": 5.3362,
      "mean_token_accuracy": 0.19303398877382277,
      "num_tokens": 33849233.0,
      "step": 16715
    },
    {
      "entropy": 5.874124765396118,
      "epoch": 1.4611552914445514,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047900812841346086,
      "loss": 5.3936,
      "mean_token_accuracy": 0.18890943378210068,
      "num_tokens": 33859996.0,
      "step": 16720
    },
    {
      "entropy": 5.947896242141724,
      "epoch": 1.4615922397972558,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004789949836365569,
      "loss": 5.3925,
      "mean_token_accuracy": 0.19671765416860582,
      "num_tokens": 33870780.0,
      "step": 16725
    },
    {
      "entropy": 5.9162639617919925,
      "epoch": 1.4620291881499607,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004789818349469222,
      "loss": 5.3991,
      "mean_token_accuracy": 0.19325575679540635,
      "num_tokens": 33880182.0,
      "step": 16730
    },
    {
      "entropy": 5.873405504226684,
      "epoch": 1.4624661365026654,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004789686823448091,
      "loss": 5.4002,
      "mean_token_accuracy": 0.18442019075155258,
      "num_tokens": 33889464.0,
      "step": 16735
    },
    {
      "entropy": 5.868384885787964,
      "epoch": 1.4629030848553701,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004789555258304698,
      "loss": 5.3498,
      "mean_token_accuracy": 0.18686940819025039,
      "num_tokens": 33899530.0,
      "step": 16740
    },
    {
      "entropy": 5.92289891242981,
      "epoch": 1.4633400332080748,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004789423654041566,
      "loss": 5.4036,
      "mean_token_accuracy": 0.1889926850795746,
      "num_tokens": 33910094.0,
      "step": 16745
    },
    {
      "entropy": 5.929456758499145,
      "epoch": 1.4637769815607795,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047892920106612204,
      "loss": 5.3239,
      "mean_token_accuracy": 0.1927153542637825,
      "num_tokens": 33921181.0,
      "step": 16750
    },
    {
      "entropy": 5.87033519744873,
      "epoch": 1.4642139299134842,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004789160328166184,
      "loss": 5.4553,
      "mean_token_accuracy": 0.1882624074816704,
      "num_tokens": 33931567.0,
      "step": 16755
    },
    {
      "entropy": 5.886786651611328,
      "epoch": 1.464650878266189,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047890286065589856,
      "loss": 5.3929,
      "mean_token_accuracy": 0.19348010569810867,
      "num_tokens": 33941513.0,
      "step": 16760
    },
    {
      "entropy": 5.883497810363769,
      "epoch": 1.4650878266188936,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00047888968458421487,
      "loss": 5.3292,
      "mean_token_accuracy": 0.19393365681171418,
      "num_tokens": 33953400.0,
      "step": 16765
    },
    {
      "entropy": 5.917049741744995,
      "epoch": 1.4655247749715983,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004788765046018204,
      "loss": 5.4043,
      "mean_token_accuracy": 0.18817954510450363,
      "num_tokens": 33964394.0,
      "step": 16770
    },
    {
      "entropy": 5.922627973556518,
      "epoch": 1.4659617233243032,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047886332070896747,
      "loss": 5.5042,
      "mean_token_accuracy": 0.1822545349597931,
      "num_tokens": 33975266.0,
      "step": 16775
    },
    {
      "entropy": 5.945690155029297,
      "epoch": 1.4663986716770077,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004788501329059094,
      "loss": 5.4521,
      "mean_token_accuracy": 0.1843678116798401,
      "num_tokens": 33986159.0,
      "step": 16780
    },
    {
      "entropy": 5.917521619796753,
      "epoch": 1.4668356200297126,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047883694119289896,
      "loss": 5.3932,
      "mean_token_accuracy": 0.19189659357070923,
      "num_tokens": 33997792.0,
      "step": 16785
    },
    {
      "entropy": 5.87490758895874,
      "epoch": 1.4672725683824173,
      "grad_norm": 1.0,
      "learning_rate": 0.00047882374557018916,
      "loss": 5.373,
      "mean_token_accuracy": 0.1921267881989479,
      "num_tokens": 34007945.0,
      "step": 16790
    },
    {
      "entropy": 5.933843755722046,
      "epoch": 1.467709516735122,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047881054603803307,
      "loss": 5.4402,
      "mean_token_accuracy": 0.19515731930732727,
      "num_tokens": 34018399.0,
      "step": 16795
    },
    {
      "entropy": 5.946059465408325,
      "epoch": 1.4681464650878266,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004787973425966839,
      "loss": 5.4633,
      "mean_token_accuracy": 0.18366601318120956,
      "num_tokens": 34028267.0,
      "step": 16800
    },
    {
      "entropy": 5.9642235279083256,
      "epoch": 1.4685834134405313,
      "grad_norm": 1.09375,
      "learning_rate": 0.000478784135246395,
      "loss": 5.4827,
      "mean_token_accuracy": 0.1841556340456009,
      "num_tokens": 34037904.0,
      "step": 16805
    },
    {
      "entropy": 5.834139108657837,
      "epoch": 1.469020361793236,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004787709239874196,
      "loss": 5.334,
      "mean_token_accuracy": 0.19967958629131316,
      "num_tokens": 34049203.0,
      "step": 16810
    },
    {
      "entropy": 5.889572095870972,
      "epoch": 1.4694573101459407,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000478757708820011,
      "loss": 5.4203,
      "mean_token_accuracy": 0.19524312466382981,
      "num_tokens": 34059670.0,
      "step": 16815
    },
    {
      "entropy": 5.874208974838257,
      "epoch": 1.4698942584986454,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004787444897444229,
      "loss": 5.3347,
      "mean_token_accuracy": 0.19880213737487792,
      "num_tokens": 34069730.0,
      "step": 16820
    },
    {
      "entropy": 5.9357822895050045,
      "epoch": 1.47033120685135,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004787312667609086,
      "loss": 5.3572,
      "mean_token_accuracy": 0.19326461404561995,
      "num_tokens": 34079315.0,
      "step": 16825
    },
    {
      "entropy": 5.90643081665039,
      "epoch": 1.470768155204055,
      "grad_norm": 1.0,
      "learning_rate": 0.000478718039869722,
      "loss": 5.368,
      "mean_token_accuracy": 0.19092688709497452,
      "num_tokens": 34089431.0,
      "step": 16830
    },
    {
      "entropy": 5.919514560699463,
      "epoch": 1.4712051035567595,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047870480907111657,
      "loss": 5.3214,
      "mean_token_accuracy": 0.19398991465568544,
      "num_tokens": 34100120.0,
      "step": 16835
    },
    {
      "entropy": 5.922703981399536,
      "epoch": 1.4716420519094644,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004786915743653461,
      "loss": 5.365,
      "mean_token_accuracy": 0.19734181761741637,
      "num_tokens": 34110271.0,
      "step": 16840
    },
    {
      "entropy": 5.913754606246949,
      "epoch": 1.472079000262169,
      "grad_norm": 1.0,
      "learning_rate": 0.00047867833575266456,
      "loss": 5.4403,
      "mean_token_accuracy": 0.18584922403097154,
      "num_tokens": 34120240.0,
      "step": 16845
    },
    {
      "entropy": 5.907829141616821,
      "epoch": 1.4725159486148738,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004786650932333259,
      "loss": 5.3672,
      "mean_token_accuracy": 0.19543719589710234,
      "num_tokens": 34130813.0,
      "step": 16850
    },
    {
      "entropy": 5.939738178253174,
      "epoch": 1.4729528969675785,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004786518468075839,
      "loss": 5.427,
      "mean_token_accuracy": 0.1943126365542412,
      "num_tokens": 34141169.0,
      "step": 16855
    },
    {
      "entropy": 5.8646503448486325,
      "epoch": 1.4733898453202832,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004786385964756928,
      "loss": 5.3821,
      "mean_token_accuracy": 0.1869334802031517,
      "num_tokens": 34151869.0,
      "step": 16860
    },
    {
      "entropy": 5.951348543167114,
      "epoch": 1.4738267936729879,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004786253422379068,
      "loss": 5.4918,
      "mean_token_accuracy": 0.17965234220027923,
      "num_tokens": 34162608.0,
      "step": 16865
    },
    {
      "entropy": 5.888792610168457,
      "epoch": 1.4742637420256925,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047861208409448,
      "loss": 5.4002,
      "mean_token_accuracy": 0.19053869545459748,
      "num_tokens": 34172648.0,
      "step": 16870
    },
    {
      "entropy": 5.905248355865479,
      "epoch": 1.4747006903783972,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004785988220456666,
      "loss": 5.3778,
      "mean_token_accuracy": 0.19408624917268752,
      "num_tokens": 34183603.0,
      "step": 16875
    },
    {
      "entropy": 5.936775350570679,
      "epoch": 1.475137638731102,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004785855560917213,
      "loss": 5.4264,
      "mean_token_accuracy": 0.18515047430992126,
      "num_tokens": 34193830.0,
      "step": 16880
    },
    {
      "entropy": 5.869189739227295,
      "epoch": 1.4755745870838066,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047857228623289825,
      "loss": 5.2463,
      "mean_token_accuracy": 0.20833783596754074,
      "num_tokens": 34203074.0,
      "step": 16885
    },
    {
      "entropy": 5.8460283279418945,
      "epoch": 1.4760115354365113,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047855901246945205,
      "loss": 5.4617,
      "mean_token_accuracy": 0.18323195427656175,
      "num_tokens": 34212958.0,
      "step": 16890
    },
    {
      "entropy": 5.908617734909058,
      "epoch": 1.4764484837892162,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004785457348016374,
      "loss": 5.3975,
      "mean_token_accuracy": 0.18510571420192717,
      "num_tokens": 34223077.0,
      "step": 16895
    },
    {
      "entropy": 5.957742643356323,
      "epoch": 1.4768854321419207,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047853245322970885,
      "loss": 5.3887,
      "mean_token_accuracy": 0.196946357190609,
      "num_tokens": 34233386.0,
      "step": 16900
    },
    {
      "entropy": 5.921525478363037,
      "epoch": 1.4773223804946256,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004785191677539212,
      "loss": 5.4015,
      "mean_token_accuracy": 0.18910367786884308,
      "num_tokens": 34244488.0,
      "step": 16905
    },
    {
      "entropy": 5.880860042572022,
      "epoch": 1.4777593288473303,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047850587837452936,
      "loss": 5.3965,
      "mean_token_accuracy": 0.18914542645215987,
      "num_tokens": 34254047.0,
      "step": 16910
    },
    {
      "entropy": 5.881480884552002,
      "epoch": 1.478196277200035,
      "grad_norm": 1.0,
      "learning_rate": 0.0004784925850917881,
      "loss": 5.3754,
      "mean_token_accuracy": 0.18625262677669524,
      "num_tokens": 34264297.0,
      "step": 16915
    },
    {
      "entropy": 5.953115034103393,
      "epoch": 1.4786332255527397,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004784792879059524,
      "loss": 5.4423,
      "mean_token_accuracy": 0.18821997791528702,
      "num_tokens": 34274945.0,
      "step": 16920
    },
    {
      "entropy": 5.882370185852051,
      "epoch": 1.4790701739054444,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047846598681727733,
      "loss": 5.2953,
      "mean_token_accuracy": 0.19185276329517365,
      "num_tokens": 34284914.0,
      "step": 16925
    },
    {
      "entropy": 5.870051765441895,
      "epoch": 1.479507122258149,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000478452681826018,
      "loss": 5.3789,
      "mean_token_accuracy": 0.19382943361997604,
      "num_tokens": 34294277.0,
      "step": 16930
    },
    {
      "entropy": 5.9569618701934814,
      "epoch": 1.4799440706108538,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047843937293242967,
      "loss": 5.4067,
      "mean_token_accuracy": 0.19203140139579772,
      "num_tokens": 34304865.0,
      "step": 16935
    },
    {
      "entropy": 5.768768835067749,
      "epoch": 1.4803810189635584,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047842606013676753,
      "loss": 5.2507,
      "mean_token_accuracy": 0.2071198269724846,
      "num_tokens": 34315404.0,
      "step": 16940
    },
    {
      "entropy": 5.978053951263428,
      "epoch": 1.4808179673162631,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047841274343928697,
      "loss": 5.3575,
      "mean_token_accuracy": 0.18734903782606124,
      "num_tokens": 34325448.0,
      "step": 16945
    },
    {
      "entropy": 5.88825306892395,
      "epoch": 1.481254915668968,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004783994228402434,
      "loss": 5.3825,
      "mean_token_accuracy": 0.20219199359416962,
      "num_tokens": 34335266.0,
      "step": 16950
    },
    {
      "entropy": 5.89263277053833,
      "epoch": 1.4816918640216725,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004783860983398924,
      "loss": 5.348,
      "mean_token_accuracy": 0.19658633321523666,
      "num_tokens": 34345847.0,
      "step": 16955
    },
    {
      "entropy": 5.85954065322876,
      "epoch": 1.4821288123743774,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004783727699384894,
      "loss": 5.2974,
      "mean_token_accuracy": 0.19412160962820052,
      "num_tokens": 34355718.0,
      "step": 16960
    },
    {
      "entropy": 5.85218768119812,
      "epoch": 1.4825657607270821,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004783594376362901,
      "loss": 5.3842,
      "mean_token_accuracy": 0.19343351572752,
      "num_tokens": 34366100.0,
      "step": 16965
    },
    {
      "entropy": 5.93750262260437,
      "epoch": 1.4830027090797868,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004783461014335502,
      "loss": 5.4463,
      "mean_token_accuracy": 0.18366983383893967,
      "num_tokens": 34376977.0,
      "step": 16970
    },
    {
      "entropy": 5.957103204727173,
      "epoch": 1.4834396574324915,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004783327613305256,
      "loss": 5.397,
      "mean_token_accuracy": 0.18791622519493104,
      "num_tokens": 34387782.0,
      "step": 16975
    },
    {
      "entropy": 5.887605953216553,
      "epoch": 1.4838766057851962,
      "grad_norm": 1.015625,
      "learning_rate": 0.000478319417327472,
      "loss": 5.3398,
      "mean_token_accuracy": 0.19444614350795747,
      "num_tokens": 34397321.0,
      "step": 16980
    },
    {
      "entropy": 5.864364147186279,
      "epoch": 1.484313554137901,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004783060694246455,
      "loss": 5.3959,
      "mean_token_accuracy": 0.1900751695036888,
      "num_tokens": 34407594.0,
      "step": 16985
    },
    {
      "entropy": 5.946622610092163,
      "epoch": 1.4847505024906056,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004782927176223021,
      "loss": 5.4037,
      "mean_token_accuracy": 0.18883211314678192,
      "num_tokens": 34417337.0,
      "step": 16990
    },
    {
      "entropy": 5.857937669754028,
      "epoch": 1.4851874508433103,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004782793619206977,
      "loss": 5.2648,
      "mean_token_accuracy": 0.200723597407341,
      "num_tokens": 34427643.0,
      "step": 16995
    },
    {
      "entropy": 5.879633045196533,
      "epoch": 1.485624399196015,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004782660023200887,
      "loss": 5.3996,
      "mean_token_accuracy": 0.19086676687002183,
      "num_tokens": 34437224.0,
      "step": 17000
    },
    {
      "entropy": 5.957306957244873,
      "epoch": 1.4860613475487199,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004782526388207313,
      "loss": 5.3889,
      "mean_token_accuracy": 0.18514008224010467,
      "num_tokens": 34447376.0,
      "step": 17005
    },
    {
      "entropy": 5.867077302932739,
      "epoch": 1.4864982959014243,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004782392714228818,
      "loss": 5.4131,
      "mean_token_accuracy": 0.19030866920948028,
      "num_tokens": 34457649.0,
      "step": 17010
    },
    {
      "entropy": 5.940423631668091,
      "epoch": 1.4869352442541293,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047822590012679654,
      "loss": 5.4077,
      "mean_token_accuracy": 0.19559329003095627,
      "num_tokens": 34467437.0,
      "step": 17015
    },
    {
      "entropy": 5.92473578453064,
      "epoch": 1.487372192606834,
      "grad_norm": 1.0,
      "learning_rate": 0.00047821252493273204,
      "loss": 5.3151,
      "mean_token_accuracy": 0.19274616092443467,
      "num_tokens": 34477615.0,
      "step": 17020
    },
    {
      "entropy": 5.8514961242675785,
      "epoch": 1.4878091409595386,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004781991458409448,
      "loss": 5.3357,
      "mean_token_accuracy": 0.1940949708223343,
      "num_tokens": 34487619.0,
      "step": 17025
    },
    {
      "entropy": 5.87247486114502,
      "epoch": 1.4882460893122433,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047818576285169154,
      "loss": 5.3574,
      "mean_token_accuracy": 0.19202938973903655,
      "num_tokens": 34498837.0,
      "step": 17030
    },
    {
      "entropy": 5.878036642074585,
      "epoch": 1.488683037664948,
      "grad_norm": 0.921875,
      "learning_rate": 0.00047817237596522885,
      "loss": 5.372,
      "mean_token_accuracy": 0.19528424143791198,
      "num_tokens": 34508972.0,
      "step": 17035
    },
    {
      "entropy": 5.945930004119873,
      "epoch": 1.4891199860176527,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004781589851818135,
      "loss": 5.3722,
      "mean_token_accuracy": 0.18947663456201552,
      "num_tokens": 34519289.0,
      "step": 17040
    },
    {
      "entropy": 5.855012464523315,
      "epoch": 1.4895569343703574,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047814559050170236,
      "loss": 5.3086,
      "mean_token_accuracy": 0.19824141561985015,
      "num_tokens": 34528384.0,
      "step": 17045
    },
    {
      "entropy": 5.8573805809021,
      "epoch": 1.489993882723062,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047813219192515235,
      "loss": 5.3516,
      "mean_token_accuracy": 0.18578649461269378,
      "num_tokens": 34537442.0,
      "step": 17050
    },
    {
      "entropy": 5.910864162445068,
      "epoch": 1.4904308310757668,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004781187894524205,
      "loss": 5.3709,
      "mean_token_accuracy": 0.19523007571697235,
      "num_tokens": 34547960.0,
      "step": 17055
    },
    {
      "entropy": 5.914022636413574,
      "epoch": 1.4908677794284715,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047810538308376376,
      "loss": 5.3122,
      "mean_token_accuracy": 0.19192912429571152,
      "num_tokens": 34556730.0,
      "step": 17060
    },
    {
      "entropy": 5.905202007293701,
      "epoch": 1.4913047277811762,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004780919728194394,
      "loss": 5.358,
      "mean_token_accuracy": 0.19097459614276885,
      "num_tokens": 34566751.0,
      "step": 17065
    },
    {
      "entropy": 5.936365413665771,
      "epoch": 1.491741676133881,
      "grad_norm": 1.125,
      "learning_rate": 0.00047807855865970454,
      "loss": 5.344,
      "mean_token_accuracy": 0.1989758253097534,
      "num_tokens": 34576417.0,
      "step": 17070
    },
    {
      "entropy": 5.935174751281738,
      "epoch": 1.4921786244865856,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004780651406048166,
      "loss": 5.4528,
      "mean_token_accuracy": 0.1875170052051544,
      "num_tokens": 34586976.0,
      "step": 17075
    },
    {
      "entropy": 5.910993909835815,
      "epoch": 1.4926155728392905,
      "grad_norm": 1.0,
      "learning_rate": 0.0004780517186550328,
      "loss": 5.3571,
      "mean_token_accuracy": 0.19539913535118103,
      "num_tokens": 34597374.0,
      "step": 17080
    },
    {
      "entropy": 5.843881702423095,
      "epoch": 1.4930525211919952,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004780382928106107,
      "loss": 5.3272,
      "mean_token_accuracy": 0.19554341286420823,
      "num_tokens": 34607586.0,
      "step": 17085
    },
    {
      "entropy": 5.930488443374633,
      "epoch": 1.4934894695446999,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047802486307180767,
      "loss": 5.4121,
      "mean_token_accuracy": 0.192698834836483,
      "num_tokens": 34617152.0,
      "step": 17090
    },
    {
      "entropy": 5.886059141159057,
      "epoch": 1.4939264178974045,
      "grad_norm": 1.0,
      "learning_rate": 0.00047801142943888135,
      "loss": 5.3221,
      "mean_token_accuracy": 0.2058628812432289,
      "num_tokens": 34627421.0,
      "step": 17095
    },
    {
      "entropy": 5.877628564834595,
      "epoch": 1.4943633662501092,
      "grad_norm": 1.0,
      "learning_rate": 0.00047799799191208955,
      "loss": 5.3351,
      "mean_token_accuracy": 0.19613291174173356,
      "num_tokens": 34637687.0,
      "step": 17100
    },
    {
      "entropy": 5.918317222595215,
      "epoch": 1.494800314602814,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004779845504916898,
      "loss": 5.4292,
      "mean_token_accuracy": 0.18591466993093492,
      "num_tokens": 34647884.0,
      "step": 17105
    },
    {
      "entropy": 5.902310991287232,
      "epoch": 1.4952372629555186,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004779711051779401,
      "loss": 5.3571,
      "mean_token_accuracy": 0.19086277186870576,
      "num_tokens": 34658658.0,
      "step": 17110
    },
    {
      "entropy": 5.889243459701538,
      "epoch": 1.4956742113082233,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004779576559710982,
      "loss": 5.3602,
      "mean_token_accuracy": 0.1945526510477066,
      "num_tokens": 34668728.0,
      "step": 17115
    },
    {
      "entropy": 5.938075113296509,
      "epoch": 1.496111159660928,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047794420287142203,
      "loss": 5.3915,
      "mean_token_accuracy": 0.18668634295463563,
      "num_tokens": 34678576.0,
      "step": 17120
    },
    {
      "entropy": 5.997232961654663,
      "epoch": 1.496548108013633,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047793074587916984,
      "loss": 5.4401,
      "mean_token_accuracy": 0.188953261077404,
      "num_tokens": 34688920.0,
      "step": 17125
    },
    {
      "entropy": 5.919175338745117,
      "epoch": 1.4969850563663374,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047791728499459955,
      "loss": 5.3481,
      "mean_token_accuracy": 0.19388152062892913,
      "num_tokens": 34699348.0,
      "step": 17130
    },
    {
      "entropy": 5.8173665523529055,
      "epoch": 1.4974220047190423,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004779038202179694,
      "loss": 5.291,
      "mean_token_accuracy": 0.20066154152154922,
      "num_tokens": 34708991.0,
      "step": 17135
    },
    {
      "entropy": 5.829899597167969,
      "epoch": 1.497858953071747,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047789035154953776,
      "loss": 5.3745,
      "mean_token_accuracy": 0.18618644326925277,
      "num_tokens": 34720869.0,
      "step": 17140
    },
    {
      "entropy": 5.920046043395996,
      "epoch": 1.4982959014244517,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004778768789895627,
      "loss": 5.3409,
      "mean_token_accuracy": 0.19303058385848998,
      "num_tokens": 34731025.0,
      "step": 17145
    },
    {
      "entropy": 5.922177076339722,
      "epoch": 1.4987328497771564,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004778634025383029,
      "loss": 5.3917,
      "mean_token_accuracy": 0.19000910371541976,
      "num_tokens": 34741846.0,
      "step": 17150
    },
    {
      "entropy": 5.826422214508057,
      "epoch": 1.499169798129861,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047784992219601673,
      "loss": 5.2783,
      "mean_token_accuracy": 0.20830559432506562,
      "num_tokens": 34753294.0,
      "step": 17155
    },
    {
      "entropy": 5.947305488586426,
      "epoch": 1.4996067464825658,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004778364379629627,
      "loss": 5.3831,
      "mean_token_accuracy": 0.19146943539381028,
      "num_tokens": 34763997.0,
      "step": 17160
    },
    {
      "entropy": 5.848710441589356,
      "epoch": 1.5000436948352704,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047782294983939955,
      "loss": 5.2716,
      "mean_token_accuracy": 0.19675895422697068,
      "num_tokens": 34772925.0,
      "step": 17165
    },
    {
      "entropy": 5.901733160018921,
      "epoch": 1.5004806431879751,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004778094578255859,
      "loss": 5.3851,
      "mean_token_accuracy": 0.1916409134864807,
      "num_tokens": 34782608.0,
      "step": 17170
    },
    {
      "entropy": 5.88265471458435,
      "epoch": 1.5009175915406798,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004777959619217807,
      "loss": 5.3631,
      "mean_token_accuracy": 0.19385848939418793,
      "num_tokens": 34792533.0,
      "step": 17175
    },
    {
      "entropy": 5.955074596405029,
      "epoch": 1.5013545398933847,
      "grad_norm": 1.125,
      "learning_rate": 0.00047778246212824263,
      "loss": 5.448,
      "mean_token_accuracy": 0.18678167313337327,
      "num_tokens": 34803069.0,
      "step": 17180
    },
    {
      "entropy": 5.913841676712036,
      "epoch": 1.5017914882460892,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004777689584452306,
      "loss": 5.3966,
      "mean_token_accuracy": 0.19314165860414506,
      "num_tokens": 34811925.0,
      "step": 17185
    },
    {
      "entropy": 5.868616247177124,
      "epoch": 1.5022284365987941,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004777554508730038,
      "loss": 5.3232,
      "mean_token_accuracy": 0.1939409852027893,
      "num_tokens": 34821511.0,
      "step": 17190
    },
    {
      "entropy": 5.89548077583313,
      "epoch": 1.5026653849514986,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004777419394118212,
      "loss": 5.4335,
      "mean_token_accuracy": 0.1886785328388214,
      "num_tokens": 34831683.0,
      "step": 17195
    },
    {
      "entropy": 5.892634057998658,
      "epoch": 1.5031023333042035,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004777284240619419,
      "loss": 5.3855,
      "mean_token_accuracy": 0.19575888514518738,
      "num_tokens": 34841578.0,
      "step": 17200
    },
    {
      "entropy": 5.925187253952027,
      "epoch": 1.5035392816569082,
      "grad_norm": 1.0,
      "learning_rate": 0.0004777149048236252,
      "loss": 5.3801,
      "mean_token_accuracy": 0.19049941450357438,
      "num_tokens": 34851959.0,
      "step": 17205
    },
    {
      "entropy": 5.876517248153687,
      "epoch": 1.503976230009613,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047770138169713047,
      "loss": 5.3816,
      "mean_token_accuracy": 0.196961210668087,
      "num_tokens": 34862557.0,
      "step": 17210
    },
    {
      "entropy": 5.929167222976685,
      "epoch": 1.5044131783623176,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047768785468271704,
      "loss": 5.4058,
      "mean_token_accuracy": 0.18997585028409958,
      "num_tokens": 34872394.0,
      "step": 17215
    },
    {
      "entropy": 5.937861442565918,
      "epoch": 1.5048501267150223,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047767432378064426,
      "loss": 5.3939,
      "mean_token_accuracy": 0.19019131511449813,
      "num_tokens": 34881938.0,
      "step": 17220
    },
    {
      "entropy": 5.894475507736206,
      "epoch": 1.505287075067727,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047766078899117174,
      "loss": 5.3277,
      "mean_token_accuracy": 0.19712415486574172,
      "num_tokens": 34892769.0,
      "step": 17225
    },
    {
      "entropy": 5.971549701690674,
      "epoch": 1.5057240234204317,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047764725031455927,
      "loss": 5.4754,
      "mean_token_accuracy": 0.1774889960885048,
      "num_tokens": 34903643.0,
      "step": 17230
    },
    {
      "entropy": 5.948525047302246,
      "epoch": 1.5061609717731366,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004776337077510662,
      "loss": 5.36,
      "mean_token_accuracy": 0.18524433225393294,
      "num_tokens": 34913558.0,
      "step": 17235
    },
    {
      "entropy": 5.919990539550781,
      "epoch": 1.506597920125841,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047762016130095246,
      "loss": 5.3422,
      "mean_token_accuracy": 0.19147340804338456,
      "num_tokens": 34923653.0,
      "step": 17240
    },
    {
      "entropy": 5.887195730209351,
      "epoch": 1.507034868478546,
      "grad_norm": 1.546875,
      "learning_rate": 0.00047760661096447784,
      "loss": 5.3844,
      "mean_token_accuracy": 0.19217015504837037,
      "num_tokens": 34933381.0,
      "step": 17245
    },
    {
      "entropy": 5.882370376586914,
      "epoch": 1.5074718168312504,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004775930567419023,
      "loss": 5.3003,
      "mean_token_accuracy": 0.19662448316812514,
      "num_tokens": 34942486.0,
      "step": 17250
    },
    {
      "entropy": 5.913878917694092,
      "epoch": 1.5079087651839553,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004775794986334857,
      "loss": 5.4324,
      "mean_token_accuracy": 0.19103975147008895,
      "num_tokens": 34952291.0,
      "step": 17255
    },
    {
      "entropy": 5.949154806137085,
      "epoch": 1.5083457135366598,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004775659366394883,
      "loss": 5.4029,
      "mean_token_accuracy": 0.19205063432455063,
      "num_tokens": 34963842.0,
      "step": 17260
    },
    {
      "entropy": 5.892694759368896,
      "epoch": 1.5087826618893647,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047755237076016997,
      "loss": 5.3249,
      "mean_token_accuracy": 0.1986763820052147,
      "num_tokens": 34973057.0,
      "step": 17265
    },
    {
      "entropy": 6.001256370544434,
      "epoch": 1.5092196102420694,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004775388009957911,
      "loss": 5.5165,
      "mean_token_accuracy": 0.17957509607076644,
      "num_tokens": 34983257.0,
      "step": 17270
    },
    {
      "entropy": 5.965215826034546,
      "epoch": 1.509656558594774,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047752522734661193,
      "loss": 5.3309,
      "mean_token_accuracy": 0.19560019820928573,
      "num_tokens": 34993066.0,
      "step": 17275
    },
    {
      "entropy": 5.959238147735595,
      "epoch": 1.5100935069474788,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047751164981289264,
      "loss": 5.3184,
      "mean_token_accuracy": 0.19239215105772017,
      "num_tokens": 35002830.0,
      "step": 17280
    },
    {
      "entropy": 5.821845865249633,
      "epoch": 1.5105304553001835,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004774980683948939,
      "loss": 5.3186,
      "mean_token_accuracy": 0.19191874861717223,
      "num_tokens": 35011837.0,
      "step": 17285
    },
    {
      "entropy": 5.899555969238281,
      "epoch": 1.5109674036528882,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047748448309287606,
      "loss": 5.3458,
      "mean_token_accuracy": 0.19205600023269653,
      "num_tokens": 35021589.0,
      "step": 17290
    },
    {
      "entropy": 5.971734237670899,
      "epoch": 1.5114043520055929,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047747089390709973,
      "loss": 5.4393,
      "mean_token_accuracy": 0.1913031041622162,
      "num_tokens": 35032207.0,
      "step": 17295
    },
    {
      "entropy": 5.958012247085572,
      "epoch": 1.5118413003582978,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004774573008378256,
      "loss": 5.4378,
      "mean_token_accuracy": 0.18277907818555833,
      "num_tokens": 35042661.0,
      "step": 17300
    },
    {
      "entropy": 5.92912278175354,
      "epoch": 1.5122782487110022,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004774437038853143,
      "loss": 5.3596,
      "mean_token_accuracy": 0.19516084641218184,
      "num_tokens": 35053050.0,
      "step": 17305
    },
    {
      "entropy": 5.96362042427063,
      "epoch": 1.5127151970637072,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047743010304982674,
      "loss": 5.4082,
      "mean_token_accuracy": 0.18751831501722335,
      "num_tokens": 35064381.0,
      "step": 17310
    },
    {
      "entropy": 5.900276041030883,
      "epoch": 1.5131521454164116,
      "grad_norm": 1.0,
      "learning_rate": 0.0004774164983316237,
      "loss": 5.3343,
      "mean_token_accuracy": 0.19146289825439453,
      "num_tokens": 35074787.0,
      "step": 17315
    },
    {
      "entropy": 5.7684159755706785,
      "epoch": 1.5135890937691165,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004774028897309662,
      "loss": 5.3246,
      "mean_token_accuracy": 0.1980694502592087,
      "num_tokens": 35085801.0,
      "step": 17320
    },
    {
      "entropy": 5.909919929504395,
      "epoch": 1.5140260421218212,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047738927724811514,
      "loss": 5.3869,
      "mean_token_accuracy": 0.1929144009947777,
      "num_tokens": 35096800.0,
      "step": 17325
    },
    {
      "entropy": 5.858640146255493,
      "epoch": 1.514462990474526,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004773756608833318,
      "loss": 5.2562,
      "mean_token_accuracy": 0.20504631400108336,
      "num_tokens": 35106735.0,
      "step": 17330
    },
    {
      "entropy": 5.830012035369873,
      "epoch": 1.5148999388272306,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047736204063687717,
      "loss": 5.3224,
      "mean_token_accuracy": 0.1979971319437027,
      "num_tokens": 35116221.0,
      "step": 17335
    },
    {
      "entropy": 5.865318155288696,
      "epoch": 1.5153368871799353,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047734841650901255,
      "loss": 5.3043,
      "mean_token_accuracy": 0.19962907433509827,
      "num_tokens": 35124896.0,
      "step": 17340
    },
    {
      "entropy": 5.919596147537232,
      "epoch": 1.51577383553264,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004773347884999993,
      "loss": 5.3701,
      "mean_token_accuracy": 0.19166366010904312,
      "num_tokens": 35134576.0,
      "step": 17345
    },
    {
      "entropy": 5.87563910484314,
      "epoch": 1.5162107838853447,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004773211566100988,
      "loss": 5.271,
      "mean_token_accuracy": 0.20484990775585174,
      "num_tokens": 35144210.0,
      "step": 17350
    },
    {
      "entropy": 5.8482842445373535,
      "epoch": 1.5166477322380496,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004773075208395725,
      "loss": 5.2971,
      "mean_token_accuracy": 0.20105746537446975,
      "num_tokens": 35153669.0,
      "step": 17355
    },
    {
      "entropy": 5.889918613433838,
      "epoch": 1.517084680590754,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004772938811886818,
      "loss": 5.41,
      "mean_token_accuracy": 0.18726354092359543,
      "num_tokens": 35163469.0,
      "step": 17360
    },
    {
      "entropy": 5.951386880874634,
      "epoch": 1.517521628943459,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004772802376576887,
      "loss": 5.4017,
      "mean_token_accuracy": 0.19011984765529633,
      "num_tokens": 35173479.0,
      "step": 17365
    },
    {
      "entropy": 5.903867387771607,
      "epoch": 1.5179585772961635,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004772665902468545,
      "loss": 5.3419,
      "mean_token_accuracy": 0.19468404352664948,
      "num_tokens": 35184570.0,
      "step": 17370
    },
    {
      "entropy": 5.91821322441101,
      "epoch": 1.5183955256488684,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004772529389564412,
      "loss": 5.3891,
      "mean_token_accuracy": 0.18678501695394517,
      "num_tokens": 35193714.0,
      "step": 17375
    },
    {
      "entropy": 5.871516418457031,
      "epoch": 1.518832474001573,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004772392837867104,
      "loss": 5.3969,
      "mean_token_accuracy": 0.1937032401561737,
      "num_tokens": 35203806.0,
      "step": 17380
    },
    {
      "entropy": 5.936268520355225,
      "epoch": 1.5192694223542778,
      "grad_norm": 1.125,
      "learning_rate": 0.00047722562473792433,
      "loss": 5.4137,
      "mean_token_accuracy": 0.1896337538957596,
      "num_tokens": 35213677.0,
      "step": 17385
    },
    {
      "entropy": 5.8915032863616945,
      "epoch": 1.5197063707069824,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047721196181034465,
      "loss": 5.2662,
      "mean_token_accuracy": 0.19353970587253572,
      "num_tokens": 35223663.0,
      "step": 17390
    },
    {
      "entropy": 5.892671585083008,
      "epoch": 1.5201433190596871,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047719829500423374,
      "loss": 5.3175,
      "mean_token_accuracy": 0.1880458787083626,
      "num_tokens": 35234551.0,
      "step": 17395
    },
    {
      "entropy": 5.874663543701172,
      "epoch": 1.5205802674123918,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047718462431985354,
      "loss": 5.3458,
      "mean_token_accuracy": 0.19164895117282868,
      "num_tokens": 35244350.0,
      "step": 17400
    },
    {
      "entropy": 5.8735603332519535,
      "epoch": 1.5210172157650965,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004771709497574662,
      "loss": 5.3018,
      "mean_token_accuracy": 0.20125180184841157,
      "num_tokens": 35253970.0,
      "step": 17405
    },
    {
      "entropy": 5.860834360122681,
      "epoch": 1.5214541641178014,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047715727131733424,
      "loss": 5.3183,
      "mean_token_accuracy": 0.1974247694015503,
      "num_tokens": 35263316.0,
      "step": 17410
    },
    {
      "entropy": 5.8445152759552,
      "epoch": 1.521891112470506,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047714358899971984,
      "loss": 5.262,
      "mean_token_accuracy": 0.20193752348423005,
      "num_tokens": 35273407.0,
      "step": 17415
    },
    {
      "entropy": 5.881416463851929,
      "epoch": 1.5223280608232108,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004771299028048855,
      "loss": 5.4296,
      "mean_token_accuracy": 0.18679196536540985,
      "num_tokens": 35284225.0,
      "step": 17420
    },
    {
      "entropy": 5.9134046077728275,
      "epoch": 1.5227650091759153,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047711621273309365,
      "loss": 5.2711,
      "mean_token_accuracy": 0.19767058938741683,
      "num_tokens": 35294747.0,
      "step": 17425
    },
    {
      "entropy": 5.900652265548706,
      "epoch": 1.5232019575286202,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000477102518784607,
      "loss": 5.3673,
      "mean_token_accuracy": 0.19901274442672728,
      "num_tokens": 35304856.0,
      "step": 17430
    },
    {
      "entropy": 5.939558124542236,
      "epoch": 1.5236389058813247,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047708882095968817,
      "loss": 5.4121,
      "mean_token_accuracy": 0.19614295810461044,
      "num_tokens": 35315550.0,
      "step": 17435
    },
    {
      "entropy": 5.929317235946655,
      "epoch": 1.5240758542340296,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004770751192585997,
      "loss": 5.4485,
      "mean_token_accuracy": 0.18357206583023072,
      "num_tokens": 35325970.0,
      "step": 17440
    },
    {
      "entropy": 5.833967542648315,
      "epoch": 1.5245128025867343,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004770614136816047,
      "loss": 5.3375,
      "mean_token_accuracy": 0.19029786437749863,
      "num_tokens": 35337277.0,
      "step": 17445
    },
    {
      "entropy": 5.937223720550537,
      "epoch": 1.524949750939439,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047704770422896586,
      "loss": 5.4097,
      "mean_token_accuracy": 0.1894491046667099,
      "num_tokens": 35347686.0,
      "step": 17450
    },
    {
      "entropy": 5.969204711914062,
      "epoch": 1.5253866992921437,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047703399090094623,
      "loss": 5.3298,
      "mean_token_accuracy": 0.19759221524000167,
      "num_tokens": 35358727.0,
      "step": 17455
    },
    {
      "entropy": 5.84347357749939,
      "epoch": 1.5258236476448483,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004770202736978087,
      "loss": 5.3802,
      "mean_token_accuracy": 0.1948215126991272,
      "num_tokens": 35369741.0,
      "step": 17460
    },
    {
      "entropy": 5.9003321647644045,
      "epoch": 1.526260595997553,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004770065526198166,
      "loss": 5.429,
      "mean_token_accuracy": 0.19388871341943742,
      "num_tokens": 35380213.0,
      "step": 17465
    },
    {
      "entropy": 5.900565147399902,
      "epoch": 1.5266975443502577,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004769928276672328,
      "loss": 5.3617,
      "mean_token_accuracy": 0.19699898213148118,
      "num_tokens": 35389937.0,
      "step": 17470
    },
    {
      "entropy": 5.915965175628662,
      "epoch": 1.5271344927029626,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004769790988403208,
      "loss": 5.3564,
      "mean_token_accuracy": 0.19320784658193588,
      "num_tokens": 35400534.0,
      "step": 17475
    },
    {
      "entropy": 5.891364669799804,
      "epoch": 1.5275714410556671,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004769653661393438,
      "loss": 5.3767,
      "mean_token_accuracy": 0.19301858991384507,
      "num_tokens": 35410607.0,
      "step": 17480
    },
    {
      "entropy": 5.9297257423400875,
      "epoch": 1.528008389408372,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004769516295645654,
      "loss": 5.3724,
      "mean_token_accuracy": 0.19150681495666505,
      "num_tokens": 35420242.0,
      "step": 17485
    },
    {
      "entropy": 5.88666467666626,
      "epoch": 1.5284453377610765,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047693788911624876,
      "loss": 5.3178,
      "mean_token_accuracy": 0.19924603700637816,
      "num_tokens": 35429469.0,
      "step": 17490
    },
    {
      "entropy": 5.861633348464966,
      "epoch": 1.5288822861137814,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004769241447946577,
      "loss": 5.3778,
      "mean_token_accuracy": 0.19314075112342835,
      "num_tokens": 35439656.0,
      "step": 17495
    },
    {
      "entropy": 5.927529668807983,
      "epoch": 1.529319234466486,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004769103966000557,
      "loss": 5.2646,
      "mean_token_accuracy": 0.19377469569444655,
      "num_tokens": 35448600.0,
      "step": 17500
    },
    {
      "entropy": 5.863404226303101,
      "epoch": 1.5297561828191908,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004768966445327065,
      "loss": 5.3505,
      "mean_token_accuracy": 0.2011193409562111,
      "num_tokens": 35458124.0,
      "step": 17505
    },
    {
      "entropy": 5.90255069732666,
      "epoch": 1.5301931311718955,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004768828885928738,
      "loss": 5.3202,
      "mean_token_accuracy": 0.19443731606006623,
      "num_tokens": 35467616.0,
      "step": 17510
    },
    {
      "entropy": 5.912509727478027,
      "epoch": 1.5306300795246002,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004768691287808216,
      "loss": 5.4015,
      "mean_token_accuracy": 0.19519911855459213,
      "num_tokens": 35478356.0,
      "step": 17515
    },
    {
      "entropy": 5.924255323410034,
      "epoch": 1.5310670278773049,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004768553650968137,
      "loss": 5.3763,
      "mean_token_accuracy": 0.1950480595231056,
      "num_tokens": 35490213.0,
      "step": 17520
    },
    {
      "entropy": 5.844392490386963,
      "epoch": 1.5315039762300096,
      "grad_norm": 1.125,
      "learning_rate": 0.0004768415975411142,
      "loss": 5.3003,
      "mean_token_accuracy": 0.20153789669275285,
      "num_tokens": 35499982.0,
      "step": 17525
    },
    {
      "entropy": 5.94242491722107,
      "epoch": 1.5319409245827145,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047682782611398693,
      "loss": 5.4067,
      "mean_token_accuracy": 0.19115403294563293,
      "num_tokens": 35510572.0,
      "step": 17530
    },
    {
      "entropy": 5.971250915527344,
      "epoch": 1.532377872935419,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047681405081569637,
      "loss": 5.4737,
      "mean_token_accuracy": 0.18341293334960937,
      "num_tokens": 35521560.0,
      "step": 17535
    },
    {
      "entropy": 5.956853580474854,
      "epoch": 1.5328148212881239,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047680027164650644,
      "loss": 5.428,
      "mean_token_accuracy": 0.18603553026914596,
      "num_tokens": 35530879.0,
      "step": 17540
    },
    {
      "entropy": 5.909808731079101,
      "epoch": 1.5332517696408283,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047678648860668163,
      "loss": 5.4095,
      "mean_token_accuracy": 0.18803575187921523,
      "num_tokens": 35541210.0,
      "step": 17545
    },
    {
      "entropy": 5.87404842376709,
      "epoch": 1.5336887179935332,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047677270169648616,
      "loss": 5.3556,
      "mean_token_accuracy": 0.19636611193418502,
      "num_tokens": 35550149.0,
      "step": 17550
    },
    {
      "entropy": 5.92678804397583,
      "epoch": 1.534125666346238,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047675891091618464,
      "loss": 5.3383,
      "mean_token_accuracy": 0.18747488707304,
      "num_tokens": 35561191.0,
      "step": 17555
    },
    {
      "entropy": 5.904361724853516,
      "epoch": 1.5345626146989426,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047674511626604144,
      "loss": 5.2947,
      "mean_token_accuracy": 0.19879430532455444,
      "num_tokens": 35571636.0,
      "step": 17560
    },
    {
      "entropy": 5.908290386199951,
      "epoch": 1.5349995630516473,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004767313177463212,
      "loss": 5.3393,
      "mean_token_accuracy": 0.19077035933732986,
      "num_tokens": 35583191.0,
      "step": 17565
    },
    {
      "entropy": 5.872945833206177,
      "epoch": 1.535436511404352,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004767175153572885,
      "loss": 5.3128,
      "mean_token_accuracy": 0.1940495863556862,
      "num_tokens": 35592875.0,
      "step": 17570
    },
    {
      "entropy": 5.85155291557312,
      "epoch": 1.5358734597570567,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004767037090992083,
      "loss": 5.2864,
      "mean_token_accuracy": 0.20377043932676314,
      "num_tokens": 35602158.0,
      "step": 17575
    },
    {
      "entropy": 5.814711618423462,
      "epoch": 1.5363104081097614,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047668989897234516,
      "loss": 5.337,
      "mean_token_accuracy": 0.19658266007900238,
      "num_tokens": 35612077.0,
      "step": 17580
    },
    {
      "entropy": 5.884006214141846,
      "epoch": 1.5367473564624663,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047667608497696415,
      "loss": 5.2731,
      "mean_token_accuracy": 0.19873480796813964,
      "num_tokens": 35621856.0,
      "step": 17585
    },
    {
      "entropy": 5.89922399520874,
      "epoch": 1.5371843048151708,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047666226711333014,
      "loss": 5.3791,
      "mean_token_accuracy": 0.19969033002853392,
      "num_tokens": 35633160.0,
      "step": 17590
    },
    {
      "entropy": 5.958705997467041,
      "epoch": 1.5376212531678757,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004766484453817081,
      "loss": 5.3883,
      "mean_token_accuracy": 0.18742993623018264,
      "num_tokens": 35642477.0,
      "step": 17595
    },
    {
      "entropy": 5.90261549949646,
      "epoch": 1.5380582015205801,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004766346197823633,
      "loss": 5.3364,
      "mean_token_accuracy": 0.19184394776821137,
      "num_tokens": 35651938.0,
      "step": 17600
    },
    {
      "entropy": 5.9062735080719,
      "epoch": 1.538495149873285,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047662079031556085,
      "loss": 5.3657,
      "mean_token_accuracy": 0.18748097717761994,
      "num_tokens": 35662881.0,
      "step": 17605
    },
    {
      "entropy": 5.898350954055786,
      "epoch": 1.5389320982259895,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004766069569815659,
      "loss": 5.3407,
      "mean_token_accuracy": 0.18979430198669434,
      "num_tokens": 35673436.0,
      "step": 17610
    },
    {
      "entropy": 5.87291259765625,
      "epoch": 1.5393690465786944,
      "grad_norm": 1.09375,
      "learning_rate": 0.000476593119780644,
      "loss": 5.3226,
      "mean_token_accuracy": 0.19359501600265502,
      "num_tokens": 35683411.0,
      "step": 17615
    },
    {
      "entropy": 5.856565523147583,
      "epoch": 1.5398059949313991,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047657927871306037,
      "loss": 5.3187,
      "mean_token_accuracy": 0.1982378602027893,
      "num_tokens": 35693806.0,
      "step": 17620
    },
    {
      "entropy": 5.876311111450195,
      "epoch": 1.5402429432841038,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004765654337790805,
      "loss": 5.354,
      "mean_token_accuracy": 0.19455063641071318,
      "num_tokens": 35705015.0,
      "step": 17625
    },
    {
      "entropy": 5.975933456420899,
      "epoch": 1.5406798916368085,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004765515849789701,
      "loss": 5.3864,
      "mean_token_accuracy": 0.19196631461381913,
      "num_tokens": 35715668.0,
      "step": 17630
    },
    {
      "entropy": 5.947570943832398,
      "epoch": 1.5411168399895132,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047653773231299455,
      "loss": 5.3458,
      "mean_token_accuracy": 0.19706546366214753,
      "num_tokens": 35726381.0,
      "step": 17635
    },
    {
      "entropy": 5.8925737857818605,
      "epoch": 1.541553788342218,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047652387578141973,
      "loss": 5.3495,
      "mean_token_accuracy": 0.19527925103902816,
      "num_tokens": 35736223.0,
      "step": 17640
    },
    {
      "entropy": 5.904705286026001,
      "epoch": 1.5419907366949226,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047651001538451146,
      "loss": 5.3597,
      "mean_token_accuracy": 0.18943448811769487,
      "num_tokens": 35746959.0,
      "step": 17645
    },
    {
      "entropy": 5.9006634712219235,
      "epoch": 1.5424276850476275,
      "grad_norm": 1.0,
      "learning_rate": 0.00047649615112253547,
      "loss": 5.3407,
      "mean_token_accuracy": 0.19884119033813477,
      "num_tokens": 35756975.0,
      "step": 17650
    },
    {
      "entropy": 5.882176685333252,
      "epoch": 1.542864633400332,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047648228299575766,
      "loss": 5.3122,
      "mean_token_accuracy": 0.1957428976893425,
      "num_tokens": 35765890.0,
      "step": 17655
    },
    {
      "entropy": 5.875711059570312,
      "epoch": 1.5433015817530369,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047646841100444415,
      "loss": 5.2599,
      "mean_token_accuracy": 0.2032760798931122,
      "num_tokens": 35775630.0,
      "step": 17660
    },
    {
      "entropy": 5.847629833221435,
      "epoch": 1.5437385301057414,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047645453514886095,
      "loss": 5.3886,
      "mean_token_accuracy": 0.1955067530274391,
      "num_tokens": 35786053.0,
      "step": 17665
    },
    {
      "entropy": 5.89942774772644,
      "epoch": 1.5441754784584463,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004764406554292742,
      "loss": 5.3746,
      "mean_token_accuracy": 0.19023712426424028,
      "num_tokens": 35795736.0,
      "step": 17670
    },
    {
      "entropy": 5.961433029174804,
      "epoch": 1.544612426811151,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004764267718459501,
      "loss": 5.3088,
      "mean_token_accuracy": 0.19603450894355773,
      "num_tokens": 35805887.0,
      "step": 17675
    },
    {
      "entropy": 5.854416847229004,
      "epoch": 1.5450493751638557,
      "grad_norm": 1.046875,
      "learning_rate": 0.000476412884399155,
      "loss": 5.3587,
      "mean_token_accuracy": 0.1933973416686058,
      "num_tokens": 35815625.0,
      "step": 17680
    },
    {
      "entropy": 5.885452699661255,
      "epoch": 1.5454863235165603,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004763989930891552,
      "loss": 5.3302,
      "mean_token_accuracy": 0.19068955332040788,
      "num_tokens": 35825617.0,
      "step": 17685
    },
    {
      "entropy": 5.913095474243164,
      "epoch": 1.545923271869265,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047638509791621714,
      "loss": 5.3591,
      "mean_token_accuracy": 0.18815550506114959,
      "num_tokens": 35835538.0,
      "step": 17690
    },
    {
      "entropy": 5.904006433486939,
      "epoch": 1.5463602202219697,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047637119888060747,
      "loss": 5.2569,
      "mean_token_accuracy": 0.20512516051530838,
      "num_tokens": 35846302.0,
      "step": 17695
    },
    {
      "entropy": 5.952295970916748,
      "epoch": 1.5467971685746744,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047635729598259266,
      "loss": 5.4558,
      "mean_token_accuracy": 0.184280164539814,
      "num_tokens": 35857384.0,
      "step": 17700
    },
    {
      "entropy": 5.979367399215699,
      "epoch": 1.5472341169273793,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004763433892224394,
      "loss": 5.3963,
      "mean_token_accuracy": 0.18361231535673142,
      "num_tokens": 35867496.0,
      "step": 17705
    },
    {
      "entropy": 5.905223417282104,
      "epoch": 1.5476710652800838,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004763294786004144,
      "loss": 5.3812,
      "mean_token_accuracy": 0.19333763867616655,
      "num_tokens": 35878051.0,
      "step": 17710
    },
    {
      "entropy": 5.8805769920349125,
      "epoch": 1.5481080136327887,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004763155641167846,
      "loss": 5.3237,
      "mean_token_accuracy": 0.19716420620679856,
      "num_tokens": 35887926.0,
      "step": 17715
    },
    {
      "entropy": 5.860585451126099,
      "epoch": 1.5485449619854932,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047630164577181666,
      "loss": 5.323,
      "mean_token_accuracy": 0.19271173626184462,
      "num_tokens": 35897982.0,
      "step": 17720
    },
    {
      "entropy": 5.879752159118652,
      "epoch": 1.548981910338198,
      "grad_norm": 1.25,
      "learning_rate": 0.0004762877235657777,
      "loss": 5.3524,
      "mean_token_accuracy": 0.1991825819015503,
      "num_tokens": 35908805.0,
      "step": 17725
    },
    {
      "entropy": 5.90902361869812,
      "epoch": 1.5494188586909028,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004762737974989348,
      "loss": 5.3307,
      "mean_token_accuracy": 0.1993390366435051,
      "num_tokens": 35918739.0,
      "step": 17730
    },
    {
      "entropy": 5.899259424209594,
      "epoch": 1.5498558070436075,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000476259867571555,
      "loss": 5.4323,
      "mean_token_accuracy": 0.19072485715150833,
      "num_tokens": 35928592.0,
      "step": 17735
    },
    {
      "entropy": 5.805506038665771,
      "epoch": 1.5502927553963122,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004762459337839055,
      "loss": 5.3207,
      "mean_token_accuracy": 0.19300913363695144,
      "num_tokens": 35938287.0,
      "step": 17740
    },
    {
      "entropy": 5.934724521636963,
      "epoch": 1.5507297037490169,
      "grad_norm": 1.125,
      "learning_rate": 0.00047623199613625343,
      "loss": 5.3383,
      "mean_token_accuracy": 0.18955149054527282,
      "num_tokens": 35947890.0,
      "step": 17745
    },
    {
      "entropy": 5.862792539596557,
      "epoch": 1.5511666521017216,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004762180546288663,
      "loss": 5.3268,
      "mean_token_accuracy": 0.2028201177716255,
      "num_tokens": 35956513.0,
      "step": 17750
    },
    {
      "entropy": 5.8051951885223385,
      "epoch": 1.5516036004544262,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004762041092620114,
      "loss": 5.3354,
      "mean_token_accuracy": 0.19995701909065247,
      "num_tokens": 35966837.0,
      "step": 17755
    },
    {
      "entropy": 5.956231307983399,
      "epoch": 1.5520405488071312,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047619016003595625,
      "loss": 5.4617,
      "mean_token_accuracy": 0.18655049055814743,
      "num_tokens": 35977076.0,
      "step": 17760
    },
    {
      "entropy": 5.975378751754761,
      "epoch": 1.5524774971598356,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047617620695096854,
      "loss": 5.3067,
      "mean_token_accuracy": 0.19733960181474686,
      "num_tokens": 35986862.0,
      "step": 17765
    },
    {
      "entropy": 5.937920331954956,
      "epoch": 1.5529144455125405,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047616225000731566,
      "loss": 5.4087,
      "mean_token_accuracy": 0.18183398693799974,
      "num_tokens": 35997488.0,
      "step": 17770
    },
    {
      "entropy": 5.8509416580200195,
      "epoch": 1.553351393865245,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004761482892052655,
      "loss": 5.3409,
      "mean_token_accuracy": 0.1901787355542183,
      "num_tokens": 36008045.0,
      "step": 17775
    },
    {
      "entropy": 5.92145528793335,
      "epoch": 1.55378834221795,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047613432454508565,
      "loss": 5.3393,
      "mean_token_accuracy": 0.19117463231086732,
      "num_tokens": 36017864.0,
      "step": 17780
    },
    {
      "entropy": 5.957915782928467,
      "epoch": 1.5542252905706544,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047612035602704407,
      "loss": 5.3203,
      "mean_token_accuracy": 0.19447557330131532,
      "num_tokens": 36027797.0,
      "step": 17785
    },
    {
      "entropy": 5.928832244873047,
      "epoch": 1.5546622389233593,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047610638365140867,
      "loss": 5.3586,
      "mean_token_accuracy": 0.1883138135075569,
      "num_tokens": 36037627.0,
      "step": 17790
    },
    {
      "entropy": 5.880336761474609,
      "epoch": 1.555099187276064,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047609240741844757,
      "loss": 5.3301,
      "mean_token_accuracy": 0.18909080028533937,
      "num_tokens": 36046324.0,
      "step": 17795
    },
    {
      "entropy": 5.877044725418091,
      "epoch": 1.5555361356287687,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004760784273284286,
      "loss": 5.3016,
      "mean_token_accuracy": 0.19933241605758667,
      "num_tokens": 36056212.0,
      "step": 17800
    },
    {
      "entropy": 5.9144690990447994,
      "epoch": 1.5559730839814734,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004760644433816201,
      "loss": 5.4494,
      "mean_token_accuracy": 0.18175553679466247,
      "num_tokens": 36066745.0,
      "step": 17805
    },
    {
      "entropy": 5.9456164836883545,
      "epoch": 1.556410032334178,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004760504555782902,
      "loss": 5.3222,
      "mean_token_accuracy": 0.1927734062075615,
      "num_tokens": 36077415.0,
      "step": 17810
    },
    {
      "entropy": 5.895496940612793,
      "epoch": 1.5568469806868828,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047603646391870715,
      "loss": 5.3444,
      "mean_token_accuracy": 0.19898289442062378,
      "num_tokens": 36088381.0,
      "step": 17815
    },
    {
      "entropy": 5.8447731018066404,
      "epoch": 1.5572839290395875,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047602246840313945,
      "loss": 5.3192,
      "mean_token_accuracy": 0.19926446080207824,
      "num_tokens": 36098333.0,
      "step": 17820
    },
    {
      "entropy": 5.913682746887207,
      "epoch": 1.5577208773922924,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047600846903185544,
      "loss": 5.3568,
      "mean_token_accuracy": 0.19547896832227707,
      "num_tokens": 36107843.0,
      "step": 17825
    },
    {
      "entropy": 5.877012395858765,
      "epoch": 1.5581578257449968,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004759944658051237,
      "loss": 5.3544,
      "mean_token_accuracy": 0.19559666216373445,
      "num_tokens": 36116948.0,
      "step": 17830
    },
    {
      "entropy": 5.909966945648193,
      "epoch": 1.5585947740977018,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004759804587232127,
      "loss": 5.3523,
      "mean_token_accuracy": 0.19740303307771684,
      "num_tokens": 36126808.0,
      "step": 17835
    },
    {
      "entropy": 5.940655946731567,
      "epoch": 1.5590317224504062,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047596644778639133,
      "loss": 5.4531,
      "mean_token_accuracy": 0.1868113398551941,
      "num_tokens": 36136630.0,
      "step": 17840
    },
    {
      "entropy": 6.004293489456177,
      "epoch": 1.5594686708031111,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004759524329949281,
      "loss": 5.3794,
      "mean_token_accuracy": 0.18518418520689012,
      "num_tokens": 36146422.0,
      "step": 17845
    },
    {
      "entropy": 5.925592947006225,
      "epoch": 1.5599056191558158,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004759384143490919,
      "loss": 5.3621,
      "mean_token_accuracy": 0.19011625796556472,
      "num_tokens": 36156169.0,
      "step": 17850
    },
    {
      "entropy": 5.879165315628052,
      "epoch": 1.5603425675085205,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004759243918491517,
      "loss": 5.3634,
      "mean_token_accuracy": 0.19730480164289474,
      "num_tokens": 36165173.0,
      "step": 17855
    },
    {
      "entropy": 5.942328786849975,
      "epoch": 1.5607795158612252,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047591036549537633,
      "loss": 5.407,
      "mean_token_accuracy": 0.1919306606054306,
      "num_tokens": 36175852.0,
      "step": 17860
    },
    {
      "entropy": 5.91231484413147,
      "epoch": 1.56121646421393,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004758963352880348,
      "loss": 5.3821,
      "mean_token_accuracy": 0.1887059509754181,
      "num_tokens": 36185288.0,
      "step": 17865
    },
    {
      "entropy": 5.921192836761475,
      "epoch": 1.5616534125666346,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004758823012273964,
      "loss": 5.3634,
      "mean_token_accuracy": 0.1899438664317131,
      "num_tokens": 36196374.0,
      "step": 17870
    },
    {
      "entropy": 5.89608359336853,
      "epoch": 1.5620903609193393,
      "grad_norm": 1.0,
      "learning_rate": 0.00047586826331373025,
      "loss": 5.3877,
      "mean_token_accuracy": 0.19689127206802368,
      "num_tokens": 36206714.0,
      "step": 17875
    },
    {
      "entropy": 5.9496448040008545,
      "epoch": 1.5625273092720442,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004758542215473054,
      "loss": 5.469,
      "mean_token_accuracy": 0.1812836393713951,
      "num_tokens": 36217891.0,
      "step": 17880
    },
    {
      "entropy": 5.903634691238404,
      "epoch": 1.5629642576247487,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004758401759283915,
      "loss": 5.3222,
      "mean_token_accuracy": 0.19527608752250672,
      "num_tokens": 36226776.0,
      "step": 17885
    },
    {
      "entropy": 5.904152679443359,
      "epoch": 1.5634012059774536,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047582612645725764,
      "loss": 5.3946,
      "mean_token_accuracy": 0.19034274220466613,
      "num_tokens": 36237362.0,
      "step": 17890
    },
    {
      "entropy": 5.892521667480469,
      "epoch": 1.563838154330158,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047581207313417346,
      "loss": 5.299,
      "mean_token_accuracy": 0.1962823674082756,
      "num_tokens": 36246264.0,
      "step": 17895
    },
    {
      "entropy": 5.933979606628418,
      "epoch": 1.564275102682863,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047579801595940853,
      "loss": 5.4011,
      "mean_token_accuracy": 0.1895760715007782,
      "num_tokens": 36255676.0,
      "step": 17900
    },
    {
      "entropy": 5.905929517745972,
      "epoch": 1.5647120510355674,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047578395493323245,
      "loss": 5.2423,
      "mean_token_accuracy": 0.20391708612442017,
      "num_tokens": 36264346.0,
      "step": 17905
    },
    {
      "entropy": 5.875597763061523,
      "epoch": 1.5651489993882723,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047576989005591495,
      "loss": 5.3567,
      "mean_token_accuracy": 0.19015281349420549,
      "num_tokens": 36273569.0,
      "step": 17910
    },
    {
      "entropy": 5.833621597290039,
      "epoch": 1.565585947740977,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004757558213277257,
      "loss": 5.2741,
      "mean_token_accuracy": 0.19822679162025453,
      "num_tokens": 36283708.0,
      "step": 17915
    },
    {
      "entropy": 5.885605573654175,
      "epoch": 1.5660228960936817,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047574174874893456,
      "loss": 5.3463,
      "mean_token_accuracy": 0.19500282853841783,
      "num_tokens": 36293184.0,
      "step": 17920
    },
    {
      "entropy": 5.911890745162964,
      "epoch": 1.5664598444463864,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004757276723198115,
      "loss": 5.4375,
      "mean_token_accuracy": 0.18188079297542573,
      "num_tokens": 36303071.0,
      "step": 17925
    },
    {
      "entropy": 5.930815315246582,
      "epoch": 1.566896792799091,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004757135920406266,
      "loss": 5.286,
      "mean_token_accuracy": 0.20310503095388413,
      "num_tokens": 36313245.0,
      "step": 17930
    },
    {
      "entropy": 5.892140960693359,
      "epoch": 1.567333741151796,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004756995079116498,
      "loss": 5.328,
      "mean_token_accuracy": 0.1891069605946541,
      "num_tokens": 36322711.0,
      "step": 17935
    },
    {
      "entropy": 5.834352731704712,
      "epoch": 1.5677706895045005,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047568541993315124,
      "loss": 5.3421,
      "mean_token_accuracy": 0.19100269079208373,
      "num_tokens": 36333667.0,
      "step": 17940
    },
    {
      "entropy": 5.982916164398193,
      "epoch": 1.5682076378572054,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047567132810540113,
      "loss": 5.4057,
      "mean_token_accuracy": 0.18766377568244935,
      "num_tokens": 36343660.0,
      "step": 17945
    },
    {
      "entropy": 5.953520202636719,
      "epoch": 1.5686445862099099,
      "grad_norm": 1.0,
      "learning_rate": 0.00047565723242866983,
      "loss": 5.4233,
      "mean_token_accuracy": 0.18948468565940857,
      "num_tokens": 36353581.0,
      "step": 17950
    },
    {
      "entropy": 5.908800649642944,
      "epoch": 1.5690815345626148,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047564313290322774,
      "loss": 5.3038,
      "mean_token_accuracy": 0.19238700419664384,
      "num_tokens": 36363818.0,
      "step": 17955
    },
    {
      "entropy": 5.874623155593872,
      "epoch": 1.5695184829153193,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004756290295293451,
      "loss": 5.3585,
      "mean_token_accuracy": 0.1978289633989334,
      "num_tokens": 36373590.0,
      "step": 17960
    },
    {
      "entropy": 5.90105094909668,
      "epoch": 1.5699554312680242,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004756149223072926,
      "loss": 5.4185,
      "mean_token_accuracy": 0.18512791097164155,
      "num_tokens": 36383556.0,
      "step": 17965
    },
    {
      "entropy": 5.916012191772461,
      "epoch": 1.5703923796207289,
      "grad_norm": 1.0,
      "learning_rate": 0.0004756008112373408,
      "loss": 5.3625,
      "mean_token_accuracy": 0.1873842731118202,
      "num_tokens": 36393194.0,
      "step": 17970
    },
    {
      "entropy": 5.874263191223145,
      "epoch": 1.5708293279734336,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004755866963197603,
      "loss": 5.3332,
      "mean_token_accuracy": 0.19659625589847565,
      "num_tokens": 36402962.0,
      "step": 17975
    },
    {
      "entropy": 5.952871465682984,
      "epoch": 1.5712662763261382,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004755725775548219,
      "loss": 5.3416,
      "mean_token_accuracy": 0.195272533595562,
      "num_tokens": 36411932.0,
      "step": 17980
    },
    {
      "entropy": 5.902034711837769,
      "epoch": 1.571703224678843,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047555845494279633,
      "loss": 5.3346,
      "mean_token_accuracy": 0.19627313762903215,
      "num_tokens": 36421363.0,
      "step": 17985
    },
    {
      "entropy": 5.912841558456421,
      "epoch": 1.5721401730315476,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004755443284839545,
      "loss": 5.3412,
      "mean_token_accuracy": 0.19433388262987136,
      "num_tokens": 36431422.0,
      "step": 17990
    },
    {
      "entropy": 5.879149103164673,
      "epoch": 1.5725771213842523,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047553019817856736,
      "loss": 5.3048,
      "mean_token_accuracy": 0.2017539396882057,
      "num_tokens": 36441004.0,
      "step": 17995
    },
    {
      "entropy": 5.968699169158936,
      "epoch": 1.5730140697369572,
      "grad_norm": 1.125,
      "learning_rate": 0.00047551606402690595,
      "loss": 5.427,
      "mean_token_accuracy": 0.1861295908689499,
      "num_tokens": 36451474.0,
      "step": 18000
    },
    {
      "epoch": 1.5730140697369572,
      "eval_entropy": 5.706996285943715,
      "eval_loss": 5.392451286315918,
      "eval_mean_token_accuracy": 0.20040247674692468,
      "eval_num_tokens": 36451474.0,
      "eval_runtime": 21.5369,
      "eval_samples_per_second": 1718.632,
      "eval_steps_per_second": 214.841,
      "step": 18000
    },
    {
      "entropy": 5.895614385604858,
      "epoch": 1.5734510180896617,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004755019260292414,
      "loss": 5.3301,
      "mean_token_accuracy": 0.1960537001490593,
      "num_tokens": 36461341.0,
      "step": 18005
    },
    {
      "entropy": 5.900636529922485,
      "epoch": 1.5738879664423666,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004754877841858448,
      "loss": 5.4457,
      "mean_token_accuracy": 0.18880501836538316,
      "num_tokens": 36471679.0,
      "step": 18010
    },
    {
      "entropy": 5.942022800445557,
      "epoch": 1.574324914795071,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004754736384969875,
      "loss": 5.3692,
      "mean_token_accuracy": 0.19350694119930267,
      "num_tokens": 36481457.0,
      "step": 18015
    },
    {
      "entropy": 5.925584316253662,
      "epoch": 1.574761863147776,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004754594889629407,
      "loss": 5.3286,
      "mean_token_accuracy": 0.19190168678760527,
      "num_tokens": 36490596.0,
      "step": 18020
    },
    {
      "entropy": 5.939559650421143,
      "epoch": 1.5751988115004807,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004754453355839759,
      "loss": 5.3957,
      "mean_token_accuracy": 0.19167802333831788,
      "num_tokens": 36501027.0,
      "step": 18025
    },
    {
      "entropy": 5.907642889022827,
      "epoch": 1.5756357598531854,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047543117836036446,
      "loss": 5.316,
      "mean_token_accuracy": 0.19303235709667205,
      "num_tokens": 36512089.0,
      "step": 18030
    },
    {
      "entropy": 5.894006204605103,
      "epoch": 1.57607270820589,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047541701729237815,
      "loss": 5.3203,
      "mean_token_accuracy": 0.19646432846784592,
      "num_tokens": 36522521.0,
      "step": 18035
    },
    {
      "entropy": 5.98519229888916,
      "epoch": 1.5765096565585948,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004754028523802883,
      "loss": 5.4335,
      "mean_token_accuracy": 0.18336056023836136,
      "num_tokens": 36533111.0,
      "step": 18040
    },
    {
      "entropy": 5.856820201873779,
      "epoch": 1.5769466049112995,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004753886836243668,
      "loss": 5.313,
      "mean_token_accuracy": 0.20637662410736085,
      "num_tokens": 36542477.0,
      "step": 18045
    },
    {
      "entropy": 5.891405248641968,
      "epoch": 1.5773835532640041,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047537451102488526,
      "loss": 5.3492,
      "mean_token_accuracy": 0.19914858490228654,
      "num_tokens": 36552137.0,
      "step": 18050
    },
    {
      "entropy": 5.929056787490845,
      "epoch": 1.577820501616709,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004753603345821157,
      "loss": 5.3708,
      "mean_token_accuracy": 0.18732179701328278,
      "num_tokens": 36562440.0,
      "step": 18055
    },
    {
      "entropy": 5.905518913269043,
      "epoch": 1.5782574499694135,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047534615429632986,
      "loss": 5.3381,
      "mean_token_accuracy": 0.19851451218128205,
      "num_tokens": 36571863.0,
      "step": 18060
    },
    {
      "entropy": 5.904295492172241,
      "epoch": 1.5786943983221184,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004753319701677999,
      "loss": 5.3245,
      "mean_token_accuracy": 0.19593575596809387,
      "num_tokens": 36581409.0,
      "step": 18065
    },
    {
      "entropy": 5.914949131011963,
      "epoch": 1.579131346674823,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047531778219679773,
      "loss": 5.3323,
      "mean_token_accuracy": 0.19080870896577834,
      "num_tokens": 36592779.0,
      "step": 18070
    },
    {
      "entropy": 5.90907564163208,
      "epoch": 1.5795682950275278,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047530359038359547,
      "loss": 5.3441,
      "mean_token_accuracy": 0.2051027685403824,
      "num_tokens": 36601173.0,
      "step": 18075
    },
    {
      "entropy": 5.911975479125976,
      "epoch": 1.5800052433802323,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004752893947284654,
      "loss": 5.3693,
      "mean_token_accuracy": 0.19352277964353562,
      "num_tokens": 36610147.0,
      "step": 18080
    },
    {
      "entropy": 5.868476295471192,
      "epoch": 1.5804421917329372,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004752751952316799,
      "loss": 5.3563,
      "mean_token_accuracy": 0.1920127421617508,
      "num_tokens": 36620093.0,
      "step": 18085
    },
    {
      "entropy": 5.9370043754577635,
      "epoch": 1.580879140085642,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047526099189351115,
      "loss": 5.3683,
      "mean_token_accuracy": 0.19532284140586853,
      "num_tokens": 36631833.0,
      "step": 18090
    },
    {
      "entropy": 5.923976802825928,
      "epoch": 1.5813160884383466,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004752467847142316,
      "loss": 5.3234,
      "mean_token_accuracy": 0.18936334699392318,
      "num_tokens": 36641123.0,
      "step": 18095
    },
    {
      "entropy": 5.902895450592041,
      "epoch": 1.5817530367910513,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047523257369411383,
      "loss": 5.389,
      "mean_token_accuracy": 0.19220559448003768,
      "num_tokens": 36652310.0,
      "step": 18100
    },
    {
      "entropy": 5.93191351890564,
      "epoch": 1.582189985143756,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004752183588334303,
      "loss": 5.425,
      "mean_token_accuracy": 0.18831028044223785,
      "num_tokens": 36662783.0,
      "step": 18105
    },
    {
      "entropy": 5.87323899269104,
      "epoch": 1.5826269334964607,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047520414013245383,
      "loss": 5.336,
      "mean_token_accuracy": 0.1992632731795311,
      "num_tokens": 36672933.0,
      "step": 18110
    },
    {
      "entropy": 5.920408248901367,
      "epoch": 1.5830638818491654,
      "grad_norm": 1.125,
      "learning_rate": 0.0004751899175914571,
      "loss": 5.419,
      "mean_token_accuracy": 0.18871867954730986,
      "num_tokens": 36682570.0,
      "step": 18115
    },
    {
      "entropy": 5.877653789520264,
      "epoch": 1.5835008302018703,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004751756912107127,
      "loss": 5.3162,
      "mean_token_accuracy": 0.19725985080003738,
      "num_tokens": 36693008.0,
      "step": 18120
    },
    {
      "entropy": 5.85341796875,
      "epoch": 1.5839377785545747,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047516146099049376,
      "loss": 5.2045,
      "mean_token_accuracy": 0.20365933030843736,
      "num_tokens": 36702611.0,
      "step": 18125
    },
    {
      "entropy": 5.873370599746704,
      "epoch": 1.5843747269072797,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047514722693107304,
      "loss": 5.3939,
      "mean_token_accuracy": 0.1926397204399109,
      "num_tokens": 36712470.0,
      "step": 18130
    },
    {
      "entropy": 5.975137424468994,
      "epoch": 1.5848116752599841,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004751329890327237,
      "loss": 5.4063,
      "mean_token_accuracy": 0.19199775904417038,
      "num_tokens": 36722578.0,
      "step": 18135
    },
    {
      "entropy": 6.001350545883179,
      "epoch": 1.585248623612689,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004751187472957187,
      "loss": 5.365,
      "mean_token_accuracy": 0.19383874535560608,
      "num_tokens": 36732334.0,
      "step": 18140
    },
    {
      "entropy": 5.885690498352051,
      "epoch": 1.5856855719653937,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004751045017203313,
      "loss": 5.4145,
      "mean_token_accuracy": 0.18766795247793197,
      "num_tokens": 36742328.0,
      "step": 18145
    },
    {
      "entropy": 5.864083003997803,
      "epoch": 1.5861225203180984,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004750902523068347,
      "loss": 5.37,
      "mean_token_accuracy": 0.19408053755760193,
      "num_tokens": 36754363.0,
      "step": 18150
    },
    {
      "entropy": 5.965652656555176,
      "epoch": 1.586559468670803,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004750759990555022,
      "loss": 5.3758,
      "mean_token_accuracy": 0.19101385325193404,
      "num_tokens": 36765588.0,
      "step": 18155
    },
    {
      "entropy": 5.911399459838867,
      "epoch": 1.5869964170235078,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004750617419666072,
      "loss": 5.3173,
      "mean_token_accuracy": 0.19593722969293595,
      "num_tokens": 36774851.0,
      "step": 18160
    },
    {
      "entropy": 5.861891078948974,
      "epoch": 1.5874333653762125,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047504748104042324,
      "loss": 5.3788,
      "mean_token_accuracy": 0.19290250092744826,
      "num_tokens": 36785836.0,
      "step": 18165
    },
    {
      "entropy": 5.8876540184021,
      "epoch": 1.5878703137289172,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047503321627722375,
      "loss": 5.3096,
      "mean_token_accuracy": 0.19669491052627563,
      "num_tokens": 36795367.0,
      "step": 18170
    },
    {
      "entropy": 5.955416107177735,
      "epoch": 1.588307262081622,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047501894767728225,
      "loss": 5.3189,
      "mean_token_accuracy": 0.19646521508693696,
      "num_tokens": 36804294.0,
      "step": 18175
    },
    {
      "entropy": 5.933022832870483,
      "epoch": 1.5887442104343266,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047500467524087266,
      "loss": 5.3723,
      "mean_token_accuracy": 0.18751894384622575,
      "num_tokens": 36813711.0,
      "step": 18180
    },
    {
      "entropy": 5.87294487953186,
      "epoch": 1.5891811587870315,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004749903989682685,
      "loss": 5.3391,
      "mean_token_accuracy": 0.19203144907951356,
      "num_tokens": 36824498.0,
      "step": 18185
    },
    {
      "entropy": 5.901075744628907,
      "epoch": 1.589618107139736,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047497611885974366,
      "loss": 5.3209,
      "mean_token_accuracy": 0.19447920918464662,
      "num_tokens": 36833329.0,
      "step": 18190
    },
    {
      "entropy": 5.937035131454468,
      "epoch": 1.5900550554924409,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047496183491557215,
      "loss": 5.365,
      "mean_token_accuracy": 0.19037197679281234,
      "num_tokens": 36842783.0,
      "step": 18195
    },
    {
      "entropy": 5.968135833740234,
      "epoch": 1.5904920038451456,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047494754713602787,
      "loss": 5.3286,
      "mean_token_accuracy": 0.19241310358047486,
      "num_tokens": 36852785.0,
      "step": 18200
    },
    {
      "entropy": 5.9530518531799315,
      "epoch": 1.5909289521978502,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047493325552138484,
      "loss": 5.3879,
      "mean_token_accuracy": 0.1876951277256012,
      "num_tokens": 36862337.0,
      "step": 18205
    },
    {
      "entropy": 5.956246423721313,
      "epoch": 1.591365900550555,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004749189600719172,
      "loss": 5.4079,
      "mean_token_accuracy": 0.18943759351968764,
      "num_tokens": 36872026.0,
      "step": 18210
    },
    {
      "entropy": 5.910048484802246,
      "epoch": 1.5918028489032596,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004749046607878992,
      "loss": 5.3808,
      "mean_token_accuracy": 0.19663942009210586,
      "num_tokens": 36882520.0,
      "step": 18215
    },
    {
      "entropy": 5.860343408584595,
      "epoch": 1.5922397972559643,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047489035766960504,
      "loss": 5.2813,
      "mean_token_accuracy": 0.20259670168161392,
      "num_tokens": 36892630.0,
      "step": 18220
    },
    {
      "entropy": 5.936383867263794,
      "epoch": 1.592676745608669,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047487605071730904,
      "loss": 5.3726,
      "mean_token_accuracy": 0.1938191458582878,
      "num_tokens": 36902692.0,
      "step": 18225
    },
    {
      "entropy": 5.9708973407745365,
      "epoch": 1.593113693961374,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047486173993128566,
      "loss": 5.4752,
      "mean_token_accuracy": 0.17902841567993164,
      "num_tokens": 36913581.0,
      "step": 18230
    },
    {
      "entropy": 5.918566799163818,
      "epoch": 1.5935506423140784,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004748474253118094,
      "loss": 5.3625,
      "mean_token_accuracy": 0.19219231903553008,
      "num_tokens": 36924031.0,
      "step": 18235
    },
    {
      "entropy": 5.906092977523803,
      "epoch": 1.5939875906667833,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004748331068591548,
      "loss": 5.3676,
      "mean_token_accuracy": 0.1904296800494194,
      "num_tokens": 36934622.0,
      "step": 18240
    },
    {
      "entropy": 5.939258813858032,
      "epoch": 1.5944245390194878,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004748187845735965,
      "loss": 5.3773,
      "mean_token_accuracy": 0.18965503871440886,
      "num_tokens": 36944374.0,
      "step": 18245
    },
    {
      "entropy": 5.952177619934082,
      "epoch": 1.5948614873721927,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047480445845540927,
      "loss": 5.3214,
      "mean_token_accuracy": 0.19959311485290526,
      "num_tokens": 36954371.0,
      "step": 18250
    },
    {
      "entropy": 5.900774097442627,
      "epoch": 1.5952984357248972,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047479012850486777,
      "loss": 5.3494,
      "mean_token_accuracy": 0.19635445177555083,
      "num_tokens": 36964217.0,
      "step": 18255
    },
    {
      "entropy": 5.967082405090332,
      "epoch": 1.595735384077602,
      "grad_norm": 1.0,
      "learning_rate": 0.00047477579472224687,
      "loss": 5.4,
      "mean_token_accuracy": 0.18863581866025925,
      "num_tokens": 36974580.0,
      "step": 18260
    },
    {
      "entropy": 5.83411545753479,
      "epoch": 1.5961723324303068,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004747614571078216,
      "loss": 5.212,
      "mean_token_accuracy": 0.21212231367826462,
      "num_tokens": 36983434.0,
      "step": 18265
    },
    {
      "entropy": 5.899393033981323,
      "epoch": 1.5966092807830115,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004747471156618669,
      "loss": 5.4435,
      "mean_token_accuracy": 0.18636171966791154,
      "num_tokens": 36993813.0,
      "step": 18270
    },
    {
      "entropy": 5.911933851242066,
      "epoch": 1.5970462291357161,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004747327703846579,
      "loss": 5.3453,
      "mean_token_accuracy": 0.19506884068250657,
      "num_tokens": 37003855.0,
      "step": 18275
    },
    {
      "entropy": 5.897821664810181,
      "epoch": 1.5974831774884208,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047471842127646966,
      "loss": 5.3218,
      "mean_token_accuracy": 0.19648411124944687,
      "num_tokens": 37013660.0,
      "step": 18280
    },
    {
      "entropy": 5.849512958526612,
      "epoch": 1.5979201258411255,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047470406833757746,
      "loss": 5.265,
      "mean_token_accuracy": 0.19645503759384156,
      "num_tokens": 37023888.0,
      "step": 18285
    },
    {
      "entropy": 5.987795066833496,
      "epoch": 1.5983570741938302,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047468971156825656,
      "loss": 5.474,
      "mean_token_accuracy": 0.1769721806049347,
      "num_tokens": 37034260.0,
      "step": 18290
    },
    {
      "entropy": 5.939581346511841,
      "epoch": 1.5987940225465351,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004746753509687824,
      "loss": 5.3711,
      "mean_token_accuracy": 0.1919611006975174,
      "num_tokens": 37045300.0,
      "step": 18295
    },
    {
      "entropy": 5.917181491851807,
      "epoch": 1.5992309708992396,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004746609865394303,
      "loss": 5.3251,
      "mean_token_accuracy": 0.19688171595335008,
      "num_tokens": 37054898.0,
      "step": 18300
    },
    {
      "entropy": 5.865383577346802,
      "epoch": 1.5996679192519445,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004746466182804759,
      "loss": 5.2886,
      "mean_token_accuracy": 0.20007946193218232,
      "num_tokens": 37064519.0,
      "step": 18305
    },
    {
      "entropy": 5.855148839950561,
      "epoch": 1.600104867604649,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004746322461921948,
      "loss": 5.3831,
      "mean_token_accuracy": 0.19339389652013778,
      "num_tokens": 37074967.0,
      "step": 18310
    },
    {
      "entropy": 5.966103744506836,
      "epoch": 1.600541815957354,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004746178702748626,
      "loss": 5.369,
      "mean_token_accuracy": 0.1938752979040146,
      "num_tokens": 37084736.0,
      "step": 18315
    },
    {
      "entropy": 5.936101245880127,
      "epoch": 1.6009787643100586,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000474603490528755,
      "loss": 5.3296,
      "mean_token_accuracy": 0.19894316792488098,
      "num_tokens": 37094327.0,
      "step": 18320
    },
    {
      "entropy": 5.851759576797486,
      "epoch": 1.6014157126627633,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004745891069541479,
      "loss": 5.3249,
      "mean_token_accuracy": 0.19660812616348267,
      "num_tokens": 37104099.0,
      "step": 18325
    },
    {
      "entropy": 5.939263820648193,
      "epoch": 1.601852661015468,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047457471955131714,
      "loss": 5.3977,
      "mean_token_accuracy": 0.19250956177711487,
      "num_tokens": 37115337.0,
      "step": 18330
    },
    {
      "entropy": 5.977224731445313,
      "epoch": 1.6022896093681727,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004745603283205387,
      "loss": 5.343,
      "mean_token_accuracy": 0.1943281978368759,
      "num_tokens": 37125522.0,
      "step": 18335
    },
    {
      "entropy": 5.943755006790161,
      "epoch": 1.6027265577208774,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004745459332620885,
      "loss": 5.422,
      "mean_token_accuracy": 0.1845826804637909,
      "num_tokens": 37135847.0,
      "step": 18340
    },
    {
      "entropy": 5.863091611862183,
      "epoch": 1.603163506073582,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004745315343762428,
      "loss": 5.3283,
      "mean_token_accuracy": 0.19180596321821214,
      "num_tokens": 37145714.0,
      "step": 18345
    },
    {
      "entropy": 5.907475137710572,
      "epoch": 1.603600454426287,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047451713166327776,
      "loss": 5.3636,
      "mean_token_accuracy": 0.19266925156116485,
      "num_tokens": 37154921.0,
      "step": 18350
    },
    {
      "entropy": 5.89783935546875,
      "epoch": 1.6040374027789914,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047450272512346953,
      "loss": 5.3178,
      "mean_token_accuracy": 0.20479815751314162,
      "num_tokens": 37165303.0,
      "step": 18355
    },
    {
      "entropy": 5.88664059638977,
      "epoch": 1.6044743511316963,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047448831475709454,
      "loss": 5.3141,
      "mean_token_accuracy": 0.19798549860715867,
      "num_tokens": 37175014.0,
      "step": 18360
    },
    {
      "entropy": 5.857017469406128,
      "epoch": 1.6049112994844008,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004744739005644291,
      "loss": 5.3267,
      "mean_token_accuracy": 0.19859050661325456,
      "num_tokens": 37185830.0,
      "step": 18365
    },
    {
      "entropy": 5.880775260925293,
      "epoch": 1.6053482478371057,
      "grad_norm": 1.0,
      "learning_rate": 0.0004744594825457498,
      "loss": 5.2429,
      "mean_token_accuracy": 0.2020293727517128,
      "num_tokens": 37196264.0,
      "step": 18370
    },
    {
      "entropy": 5.923350381851196,
      "epoch": 1.6057851961898104,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047444506070133305,
      "loss": 5.3807,
      "mean_token_accuracy": 0.19468964785337448,
      "num_tokens": 37206338.0,
      "step": 18375
    },
    {
      "entropy": 5.873742532730103,
      "epoch": 1.606222144542515,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004744306350314555,
      "loss": 5.3253,
      "mean_token_accuracy": 0.2009591653943062,
      "num_tokens": 37216231.0,
      "step": 18380
    },
    {
      "entropy": 5.952521848678589,
      "epoch": 1.6066590928952198,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004744162055363939,
      "loss": 5.4531,
      "mean_token_accuracy": 0.18138249665498735,
      "num_tokens": 37227209.0,
      "step": 18385
    },
    {
      "entropy": 5.983677911758423,
      "epoch": 1.6070960412479245,
      "grad_norm": 1.109375,
      "learning_rate": 0.000474401772216425,
      "loss": 5.447,
      "mean_token_accuracy": 0.18832987248897554,
      "num_tokens": 37238056.0,
      "step": 18390
    },
    {
      "entropy": 5.933745574951172,
      "epoch": 1.6075329896006292,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004743873350718257,
      "loss": 5.3468,
      "mean_token_accuracy": 0.19368500560522078,
      "num_tokens": 37247864.0,
      "step": 18395
    },
    {
      "entropy": 5.872524452209473,
      "epoch": 1.6079699379533339,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004743728941028727,
      "loss": 5.2911,
      "mean_token_accuracy": 0.19647256880998612,
      "num_tokens": 37256418.0,
      "step": 18400
    },
    {
      "entropy": 5.8862110614776615,
      "epoch": 1.6084068863060388,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004743584493098431,
      "loss": 5.3854,
      "mean_token_accuracy": 0.1913978189229965,
      "num_tokens": 37266540.0,
      "step": 18405
    },
    {
      "entropy": 5.950182723999023,
      "epoch": 1.6088438346587433,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047434400069301414,
      "loss": 5.3596,
      "mean_token_accuracy": 0.189559605717659,
      "num_tokens": 37277011.0,
      "step": 18410
    },
    {
      "entropy": 5.899205160140991,
      "epoch": 1.6092807830114482,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004743295482526627,
      "loss": 5.3512,
      "mean_token_accuracy": 0.19226530492305755,
      "num_tokens": 37287203.0,
      "step": 18415
    },
    {
      "entropy": 5.908243560791016,
      "epoch": 1.6097177313641526,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004743150919890661,
      "loss": 5.3064,
      "mean_token_accuracy": 0.19964665174484253,
      "num_tokens": 37297293.0,
      "step": 18420
    },
    {
      "entropy": 5.951489114761353,
      "epoch": 1.6101546797168576,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047430063190250155,
      "loss": 5.3996,
      "mean_token_accuracy": 0.18811395317316054,
      "num_tokens": 37307882.0,
      "step": 18425
    },
    {
      "entropy": 5.898273181915283,
      "epoch": 1.610591628069562,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004742861679932464,
      "loss": 5.2766,
      "mean_token_accuracy": 0.20785792618989946,
      "num_tokens": 37317466.0,
      "step": 18430
    },
    {
      "entropy": 5.884138822555542,
      "epoch": 1.611028576422267,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047427170026157823,
      "loss": 5.4267,
      "mean_token_accuracy": 0.18434988707304,
      "num_tokens": 37327544.0,
      "step": 18435
    },
    {
      "entropy": 5.912940549850464,
      "epoch": 1.6114655247749716,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004742572287077744,
      "loss": 5.4042,
      "mean_token_accuracy": 0.18560616075992584,
      "num_tokens": 37338082.0,
      "step": 18440
    },
    {
      "entropy": 5.923292207717895,
      "epoch": 1.6119024731276763,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047424275333211256,
      "loss": 5.3424,
      "mean_token_accuracy": 0.19257732927799226,
      "num_tokens": 37348127.0,
      "step": 18445
    },
    {
      "entropy": 5.951858377456665,
      "epoch": 1.612339421480381,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004742282741348702,
      "loss": 5.415,
      "mean_token_accuracy": 0.18269358575344086,
      "num_tokens": 37359547.0,
      "step": 18450
    },
    {
      "entropy": 5.984919166564941,
      "epoch": 1.6127763698330857,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004742137911163253,
      "loss": 5.4646,
      "mean_token_accuracy": 0.19212558269500732,
      "num_tokens": 37369547.0,
      "step": 18455
    },
    {
      "entropy": 5.894956779479981,
      "epoch": 1.6132133181857904,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004741993042767554,
      "loss": 5.4473,
      "mean_token_accuracy": 0.18379407972097397,
      "num_tokens": 37380055.0,
      "step": 18460
    },
    {
      "entropy": 5.928543901443481,
      "epoch": 1.613650266538495,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047418481361643844,
      "loss": 5.4179,
      "mean_token_accuracy": 0.18643434941768647,
      "num_tokens": 37390179.0,
      "step": 18465
    },
    {
      "entropy": 5.9538657665252686,
      "epoch": 1.6140872148912,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047417031913565247,
      "loss": 5.3721,
      "mean_token_accuracy": 0.19514377266168595,
      "num_tokens": 37400663.0,
      "step": 18470
    },
    {
      "entropy": 5.953991508483886,
      "epoch": 1.6145241632439045,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004741558208346753,
      "loss": 5.3673,
      "mean_token_accuracy": 0.19902296513319015,
      "num_tokens": 37410988.0,
      "step": 18475
    },
    {
      "entropy": 5.954576206207276,
      "epoch": 1.6149611115966094,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047414131871378526,
      "loss": 5.4269,
      "mean_token_accuracy": 0.18880924433469773,
      "num_tokens": 37422092.0,
      "step": 18480
    },
    {
      "entropy": 5.870896339416504,
      "epoch": 1.6153980599493138,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004741268127732603,
      "loss": 5.3572,
      "mean_token_accuracy": 0.1954844206571579,
      "num_tokens": 37431637.0,
      "step": 18485
    },
    {
      "entropy": 5.962408447265625,
      "epoch": 1.6158350083020188,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047411230301337877,
      "loss": 5.4461,
      "mean_token_accuracy": 0.17834992408752443,
      "num_tokens": 37442678.0,
      "step": 18490
    },
    {
      "entropy": 5.954875946044922,
      "epoch": 1.6162719566547235,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004740977894344189,
      "loss": 5.3615,
      "mean_token_accuracy": 0.1915615439414978,
      "num_tokens": 37452324.0,
      "step": 18495
    },
    {
      "entropy": 5.974987506866455,
      "epoch": 1.6167089050074281,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004740832720366591,
      "loss": 5.3126,
      "mean_token_accuracy": 0.1891907960176468,
      "num_tokens": 37462940.0,
      "step": 18500
    },
    {
      "entropy": 5.909378719329834,
      "epoch": 1.6171458533601328,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004740687508203779,
      "loss": 5.4187,
      "mean_token_accuracy": 0.19214618653059007,
      "num_tokens": 37474121.0,
      "step": 18505
    },
    {
      "entropy": 5.923149156570434,
      "epoch": 1.6175828017128375,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004740542257858536,
      "loss": 5.3753,
      "mean_token_accuracy": 0.19024090766906737,
      "num_tokens": 37483439.0,
      "step": 18510
    },
    {
      "entropy": 5.952045965194702,
      "epoch": 1.6180197500655422,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047403969693336507,
      "loss": 5.3661,
      "mean_token_accuracy": 0.1920511469244957,
      "num_tokens": 37493804.0,
      "step": 18515
    },
    {
      "entropy": 5.918132400512695,
      "epoch": 1.618456698418247,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004740251642631909,
      "loss": 5.2687,
      "mean_token_accuracy": 0.19767033755779267,
      "num_tokens": 37504158.0,
      "step": 18520
    },
    {
      "entropy": 5.923590469360351,
      "epoch": 1.6188936467709518,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004740106277756097,
      "loss": 5.3626,
      "mean_token_accuracy": 0.1892780289053917,
      "num_tokens": 37515069.0,
      "step": 18525
    },
    {
      "entropy": 5.860126495361328,
      "epoch": 1.6193305951236563,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047399608747090033,
      "loss": 5.2807,
      "mean_token_accuracy": 0.1982532858848572,
      "num_tokens": 37525486.0,
      "step": 18530
    },
    {
      "entropy": 5.928969430923462,
      "epoch": 1.6197675434763612,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047398154334934187,
      "loss": 5.3977,
      "mean_token_accuracy": 0.18795075714588166,
      "num_tokens": 37535616.0,
      "step": 18535
    },
    {
      "entropy": 5.88739652633667,
      "epoch": 1.6202044918290657,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047396699541121296,
      "loss": 5.3014,
      "mean_token_accuracy": 0.20071678161621093,
      "num_tokens": 37546084.0,
      "step": 18540
    },
    {
      "entropy": 5.881149816513061,
      "epoch": 1.6206414401817706,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004739524436567929,
      "loss": 5.354,
      "mean_token_accuracy": 0.19755718261003494,
      "num_tokens": 37555865.0,
      "step": 18545
    },
    {
      "entropy": 5.866886186599731,
      "epoch": 1.6210783885344753,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004739378880863608,
      "loss": 5.3489,
      "mean_token_accuracy": 0.18973239958286287,
      "num_tokens": 37565985.0,
      "step": 18550
    },
    {
      "entropy": 5.9431986808776855,
      "epoch": 1.62151533688718,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004739233287001957,
      "loss": 5.426,
      "mean_token_accuracy": 0.19054572880268097,
      "num_tokens": 37575887.0,
      "step": 18555
    },
    {
      "entropy": 5.9268293380737305,
      "epoch": 1.6219522852398847,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047390876549857695,
      "loss": 5.276,
      "mean_token_accuracy": 0.2008074179291725,
      "num_tokens": 37585346.0,
      "step": 18560
    },
    {
      "entropy": 5.900470972061157,
      "epoch": 1.6223892335925894,
      "grad_norm": 1.0,
      "learning_rate": 0.00047389419848178384,
      "loss": 5.3446,
      "mean_token_accuracy": 0.19639372527599336,
      "num_tokens": 37595674.0,
      "step": 18565
    },
    {
      "entropy": 5.823383617401123,
      "epoch": 1.622826181945294,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047387962765009576,
      "loss": 5.3117,
      "mean_token_accuracy": 0.19761801064014434,
      "num_tokens": 37605537.0,
      "step": 18570
    },
    {
      "entropy": 5.973278141021728,
      "epoch": 1.6232631302979987,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047386505300379224,
      "loss": 5.4306,
      "mean_token_accuracy": 0.18718017935752868,
      "num_tokens": 37615801.0,
      "step": 18575
    },
    {
      "entropy": 5.936657905578613,
      "epoch": 1.6237000786507036,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004738504745431528,
      "loss": 5.2969,
      "mean_token_accuracy": 0.20527651011943818,
      "num_tokens": 37625582.0,
      "step": 18580
    },
    {
      "entropy": 5.875746726989746,
      "epoch": 1.6241370270034081,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047383589226845707,
      "loss": 5.4134,
      "mean_token_accuracy": 0.19119309931993483,
      "num_tokens": 37636075.0,
      "step": 18585
    },
    {
      "entropy": 5.83789291381836,
      "epoch": 1.624573975356113,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047382130617998477,
      "loss": 5.2826,
      "mean_token_accuracy": 0.20057970881462098,
      "num_tokens": 37645692.0,
      "step": 18590
    },
    {
      "entropy": 5.930126476287842,
      "epoch": 1.6250109237088175,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004738067162780156,
      "loss": 5.3707,
      "mean_token_accuracy": 0.19036000370979309,
      "num_tokens": 37655035.0,
      "step": 18595
    },
    {
      "entropy": 5.926767873764038,
      "epoch": 1.6254478720615224,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004737921225628294,
      "loss": 5.3098,
      "mean_token_accuracy": 0.19148503839969636,
      "num_tokens": 37665829.0,
      "step": 18600
    },
    {
      "entropy": 5.8574378967285154,
      "epoch": 1.6258848204142269,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047377752503470613,
      "loss": 5.3392,
      "mean_token_accuracy": 0.20566757768392563,
      "num_tokens": 37676318.0,
      "step": 18605
    },
    {
      "entropy": 5.895806550979614,
      "epoch": 1.6263217687669318,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047376292369392584,
      "loss": 5.3733,
      "mean_token_accuracy": 0.19361147433519363,
      "num_tokens": 37686977.0,
      "step": 18610
    },
    {
      "entropy": 5.8874427318573,
      "epoch": 1.6267587171196365,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047374831854076843,
      "loss": 5.3218,
      "mean_token_accuracy": 0.1972762778401375,
      "num_tokens": 37697004.0,
      "step": 18615
    },
    {
      "entropy": 5.936458683013916,
      "epoch": 1.6271956654723412,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047373370957551423,
      "loss": 5.355,
      "mean_token_accuracy": 0.19368371665477752,
      "num_tokens": 37707613.0,
      "step": 18620
    },
    {
      "entropy": 5.959619808197021,
      "epoch": 1.6276326138250459,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004737190967984433,
      "loss": 5.3491,
      "mean_token_accuracy": 0.1895310789346695,
      "num_tokens": 37717167.0,
      "step": 18625
    },
    {
      "entropy": 5.925066089630127,
      "epoch": 1.6280695621777506,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004737044802098359,
      "loss": 5.4371,
      "mean_token_accuracy": 0.18496766984462737,
      "num_tokens": 37727496.0,
      "step": 18630
    },
    {
      "entropy": 5.891392946243286,
      "epoch": 1.6285065105304553,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004736898598099725,
      "loss": 5.3521,
      "mean_token_accuracy": 0.19699207097291946,
      "num_tokens": 37736844.0,
      "step": 18635
    },
    {
      "entropy": 5.927703380584717,
      "epoch": 1.62894345888316,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047367523559913347,
      "loss": 5.3691,
      "mean_token_accuracy": 0.19027186632156373,
      "num_tokens": 37746027.0,
      "step": 18640
    },
    {
      "entropy": 5.941772985458374,
      "epoch": 1.6293804072358649,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047366060757759933,
      "loss": 5.3074,
      "mean_token_accuracy": 0.1976277694106102,
      "num_tokens": 37756499.0,
      "step": 18645
    },
    {
      "entropy": 5.919715499877929,
      "epoch": 1.6298173555885693,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047364597574565064,
      "loss": 5.3111,
      "mean_token_accuracy": 0.19833234697580338,
      "num_tokens": 37766533.0,
      "step": 18650
    },
    {
      "entropy": 5.87665581703186,
      "epoch": 1.6302543039412742,
      "grad_norm": 1.03125,
      "learning_rate": 0.000473631340103568,
      "loss": 5.3371,
      "mean_token_accuracy": 0.20190175026655197,
      "num_tokens": 37775937.0,
      "step": 18655
    },
    {
      "entropy": 5.9244935512542725,
      "epoch": 1.6306912522939787,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047361670065163225,
      "loss": 5.3282,
      "mean_token_accuracy": 0.20088298618793488,
      "num_tokens": 37786587.0,
      "step": 18660
    },
    {
      "entropy": 5.922546482086181,
      "epoch": 1.6311282006466836,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047360205739012406,
      "loss": 5.3929,
      "mean_token_accuracy": 0.1948801800608635,
      "num_tokens": 37797200.0,
      "step": 18665
    },
    {
      "entropy": 5.849664688110352,
      "epoch": 1.6315651489993883,
      "grad_norm": 1.125,
      "learning_rate": 0.0004735874103193244,
      "loss": 5.3402,
      "mean_token_accuracy": 0.19793972969055176,
      "num_tokens": 37806652.0,
      "step": 18670
    },
    {
      "entropy": 5.934342336654663,
      "epoch": 1.632002097352093,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004735727594395141,
      "loss": 5.416,
      "mean_token_accuracy": 0.18322045505046844,
      "num_tokens": 37817057.0,
      "step": 18675
    },
    {
      "entropy": 5.932273960113525,
      "epoch": 1.6324390457047977,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047355810475097423,
      "loss": 5.3377,
      "mean_token_accuracy": 0.190315842628479,
      "num_tokens": 37826216.0,
      "step": 18680
    },
    {
      "entropy": 5.950630807876587,
      "epoch": 1.6328759940575024,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047354344625398596,
      "loss": 5.4035,
      "mean_token_accuracy": 0.1880777359008789,
      "num_tokens": 37835834.0,
      "step": 18685
    },
    {
      "entropy": 5.8595606803894045,
      "epoch": 1.633312942410207,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004735287839488302,
      "loss": 5.3174,
      "mean_token_accuracy": 0.20235079228878022,
      "num_tokens": 37846252.0,
      "step": 18690
    },
    {
      "entropy": 5.856032466888427,
      "epoch": 1.6337498907629118,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047351411783578846,
      "loss": 5.3022,
      "mean_token_accuracy": 0.1946254700422287,
      "num_tokens": 37856020.0,
      "step": 18695
    },
    {
      "entropy": 5.933895254135132,
      "epoch": 1.6341868391156167,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004734994479151418,
      "loss": 5.3417,
      "mean_token_accuracy": 0.1912478670477867,
      "num_tokens": 37866132.0,
      "step": 18700
    },
    {
      "entropy": 5.918582439422607,
      "epoch": 1.6346237874683212,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004734847741871718,
      "loss": 5.3867,
      "mean_token_accuracy": 0.19220236837863922,
      "num_tokens": 37876779.0,
      "step": 18705
    },
    {
      "entropy": 5.913011646270752,
      "epoch": 1.635060735821026,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047347009665215983,
      "loss": 5.3704,
      "mean_token_accuracy": 0.19430431872606277,
      "num_tokens": 37886981.0,
      "step": 18710
    },
    {
      "entropy": 5.8723996639251705,
      "epoch": 1.6354976841737305,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004734554153103874,
      "loss": 5.3165,
      "mean_token_accuracy": 0.20260583907365798,
      "num_tokens": 37896974.0,
      "step": 18715
    },
    {
      "entropy": 5.960935497283936,
      "epoch": 1.6359346325264355,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047344073016213615,
      "loss": 5.4483,
      "mean_token_accuracy": 0.1880125880241394,
      "num_tokens": 37909020.0,
      "step": 18720
    },
    {
      "entropy": 5.933570528030396,
      "epoch": 1.63637158087914,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047342604120768764,
      "loss": 5.2843,
      "mean_token_accuracy": 0.19772206693887712,
      "num_tokens": 37919417.0,
      "step": 18725
    },
    {
      "entropy": 5.920751142501831,
      "epoch": 1.6368085292318448,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004734113484473237,
      "loss": 5.4397,
      "mean_token_accuracy": 0.18745054453611373,
      "num_tokens": 37929030.0,
      "step": 18730
    },
    {
      "entropy": 5.939340257644654,
      "epoch": 1.6372454775845495,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004733966518813262,
      "loss": 5.307,
      "mean_token_accuracy": 0.1966559872031212,
      "num_tokens": 37939157.0,
      "step": 18735
    },
    {
      "entropy": 5.890168952941894,
      "epoch": 1.6376824259372542,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047338195150997686,
      "loss": 5.3206,
      "mean_token_accuracy": 0.19096328914165497,
      "num_tokens": 37950544.0,
      "step": 18740
    },
    {
      "entropy": 5.865218210220337,
      "epoch": 1.638119374289959,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047336724733355786,
      "loss": 5.3,
      "mean_token_accuracy": 0.20406735092401504,
      "num_tokens": 37960059.0,
      "step": 18745
    },
    {
      "entropy": 5.884112310409546,
      "epoch": 1.6385563226426636,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047335253935235094,
      "loss": 5.3372,
      "mean_token_accuracy": 0.19595694690942764,
      "num_tokens": 37970669.0,
      "step": 18750
    },
    {
      "entropy": 5.884138345718384,
      "epoch": 1.6389932709953685,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047333782756663853,
      "loss": 5.3416,
      "mean_token_accuracy": 0.1993900328874588,
      "num_tokens": 37980088.0,
      "step": 18755
    },
    {
      "entropy": 5.923800134658814,
      "epoch": 1.639430219348073,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047332311197670264,
      "loss": 5.386,
      "mean_token_accuracy": 0.19393368065357208,
      "num_tokens": 37989433.0,
      "step": 18760
    },
    {
      "entropy": 5.91578106880188,
      "epoch": 1.639867167700778,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047330839258282544,
      "loss": 5.3569,
      "mean_token_accuracy": 0.19115270376205445,
      "num_tokens": 37999735.0,
      "step": 18765
    },
    {
      "entropy": 5.881199789047241,
      "epoch": 1.6403041160534824,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004732936693852895,
      "loss": 5.2484,
      "mean_token_accuracy": 0.2071860358119011,
      "num_tokens": 38008515.0,
      "step": 18770
    },
    {
      "entropy": 5.925659847259522,
      "epoch": 1.6407410644061873,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047327894238437696,
      "loss": 5.393,
      "mean_token_accuracy": 0.1930020660161972,
      "num_tokens": 38018778.0,
      "step": 18775
    },
    {
      "entropy": 5.964154386520386,
      "epoch": 1.6411780127588917,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047326421158037045,
      "loss": 5.3922,
      "mean_token_accuracy": 0.1954418271780014,
      "num_tokens": 38029062.0,
      "step": 18780
    },
    {
      "entropy": 5.86992826461792,
      "epoch": 1.6416149611115967,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047324947697355253,
      "loss": 5.2668,
      "mean_token_accuracy": 0.20118501484394075,
      "num_tokens": 38039945.0,
      "step": 18785
    },
    {
      "entropy": 5.970302295684815,
      "epoch": 1.6420519094643014,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047323473856420573,
      "loss": 5.3663,
      "mean_token_accuracy": 0.18911336362361908,
      "num_tokens": 38050213.0,
      "step": 18790
    },
    {
      "entropy": 5.886522722244263,
      "epoch": 1.642488857817006,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047321999635261276,
      "loss": 5.3007,
      "mean_token_accuracy": 0.1961507871747017,
      "num_tokens": 38060147.0,
      "step": 18795
    },
    {
      "entropy": 5.9101815700531,
      "epoch": 1.6429258061697107,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004732052503390563,
      "loss": 5.3785,
      "mean_token_accuracy": 0.18933226764202118,
      "num_tokens": 38070854.0,
      "step": 18800
    },
    {
      "entropy": 5.933733081817627,
      "epoch": 1.6433627545224154,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004731905005238194,
      "loss": 5.4004,
      "mean_token_accuracy": 0.1912371262907982,
      "num_tokens": 38080998.0,
      "step": 18805
    },
    {
      "entropy": 5.916021013259888,
      "epoch": 1.6437997028751201,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047317574690718486,
      "loss": 5.3779,
      "mean_token_accuracy": 0.1912556141614914,
      "num_tokens": 38091421.0,
      "step": 18810
    },
    {
      "entropy": 5.866908168792724,
      "epoch": 1.6442366512278248,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047316098948943557,
      "loss": 5.3322,
      "mean_token_accuracy": 0.1953820750117302,
      "num_tokens": 38101106.0,
      "step": 18815
    },
    {
      "entropy": 5.879564094543457,
      "epoch": 1.6446735995805297,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047314622827085466,
      "loss": 5.3337,
      "mean_token_accuracy": 0.19353198409080505,
      "num_tokens": 38111346.0,
      "step": 18820
    },
    {
      "entropy": 5.9992516994476315,
      "epoch": 1.6451105479332342,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047313146325172534,
      "loss": 5.4273,
      "mean_token_accuracy": 0.18604643195867537,
      "num_tokens": 38121798.0,
      "step": 18825
    },
    {
      "entropy": 5.956054830551148,
      "epoch": 1.645547496285939,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004731166944323307,
      "loss": 5.3863,
      "mean_token_accuracy": 0.18822738528251648,
      "num_tokens": 38134145.0,
      "step": 18830
    },
    {
      "entropy": 5.936854457855224,
      "epoch": 1.6459844446386436,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047310192181295406,
      "loss": 5.3218,
      "mean_token_accuracy": 0.19302600026130676,
      "num_tokens": 38144213.0,
      "step": 18835
    },
    {
      "entropy": 5.855650234222412,
      "epoch": 1.6464213929913485,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004730871453938788,
      "loss": 5.3324,
      "mean_token_accuracy": 0.19606308043003082,
      "num_tokens": 38154245.0,
      "step": 18840
    },
    {
      "entropy": 5.88578577041626,
      "epoch": 1.6468583413440532,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004730723651753882,
      "loss": 5.3472,
      "mean_token_accuracy": 0.196369406580925,
      "num_tokens": 38163969.0,
      "step": 18845
    },
    {
      "entropy": 5.903145217895508,
      "epoch": 1.6472952896967579,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004730575811577659,
      "loss": 5.28,
      "mean_token_accuracy": 0.1963663399219513,
      "num_tokens": 38173586.0,
      "step": 18850
    },
    {
      "entropy": 5.862619638442993,
      "epoch": 1.6477322380494626,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004730427933412954,
      "loss": 5.2953,
      "mean_token_accuracy": 0.1948613554239273,
      "num_tokens": 38182702.0,
      "step": 18855
    },
    {
      "entropy": 5.861179065704346,
      "epoch": 1.6481691864021673,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004730280017262604,
      "loss": 5.3402,
      "mean_token_accuracy": 0.19427991658449173,
      "num_tokens": 38191937.0,
      "step": 18860
    },
    {
      "entropy": 5.8628252983093265,
      "epoch": 1.648606134754872,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004730132063129445,
      "loss": 5.3137,
      "mean_token_accuracy": 0.1937575340270996,
      "num_tokens": 38201234.0,
      "step": 18865
    },
    {
      "entropy": 5.890830707550049,
      "epoch": 1.6490430831075766,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047299840710163157,
      "loss": 5.3374,
      "mean_token_accuracy": 0.19800386875867843,
      "num_tokens": 38211790.0,
      "step": 18870
    },
    {
      "entropy": 5.948133945465088,
      "epoch": 1.6494800314602815,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047298360409260543,
      "loss": 5.3214,
      "mean_token_accuracy": 0.19291991293430327,
      "num_tokens": 38221946.0,
      "step": 18875
    },
    {
      "entropy": 5.90933051109314,
      "epoch": 1.649916979812986,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047296879728615,
      "loss": 5.3681,
      "mean_token_accuracy": 0.19313209354877472,
      "num_tokens": 38232366.0,
      "step": 18880
    },
    {
      "entropy": 5.928026247024536,
      "epoch": 1.650353928165691,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047295398668254933,
      "loss": 5.4029,
      "mean_token_accuracy": 0.19180445820093156,
      "num_tokens": 38242504.0,
      "step": 18885
    },
    {
      "entropy": 5.9991028785705565,
      "epoch": 1.6507908765183954,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004729391722820875,
      "loss": 5.3955,
      "mean_token_accuracy": 0.19306564331054688,
      "num_tokens": 38251916.0,
      "step": 18890
    },
    {
      "entropy": 5.889881610870361,
      "epoch": 1.6512278248711003,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047292435408504855,
      "loss": 5.2981,
      "mean_token_accuracy": 0.19855620414018632,
      "num_tokens": 38262438.0,
      "step": 18895
    },
    {
      "entropy": 5.792263698577881,
      "epoch": 1.6516647732238048,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004729095320917169,
      "loss": 5.2933,
      "mean_token_accuracy": 0.207982137799263,
      "num_tokens": 38273095.0,
      "step": 18900
    },
    {
      "entropy": 5.947273206710816,
      "epoch": 1.6521017215765097,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004728947063023766,
      "loss": 5.3371,
      "mean_token_accuracy": 0.1908865362405777,
      "num_tokens": 38282827.0,
      "step": 18905
    },
    {
      "entropy": 5.999677944183349,
      "epoch": 1.6525386699292144,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004728798767173122,
      "loss": 5.3023,
      "mean_token_accuracy": 0.19088486731052398,
      "num_tokens": 38292827.0,
      "step": 18910
    },
    {
      "entropy": 5.890858507156372,
      "epoch": 1.652975618281919,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047286504333680804,
      "loss": 5.3746,
      "mean_token_accuracy": 0.1862487941980362,
      "num_tokens": 38302658.0,
      "step": 18915
    },
    {
      "entropy": 5.851995134353638,
      "epoch": 1.6534125666346238,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004728502061611488,
      "loss": 5.3403,
      "mean_token_accuracy": 0.19000650495290755,
      "num_tokens": 38312154.0,
      "step": 18920
    },
    {
      "entropy": 5.943078517913818,
      "epoch": 1.6538495149873285,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047283536519061875,
      "loss": 5.3198,
      "mean_token_accuracy": 0.1952947497367859,
      "num_tokens": 38321317.0,
      "step": 18925
    },
    {
      "entropy": 5.921110534667969,
      "epoch": 1.6542864633400332,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00047282052042550276,
      "loss": 5.3274,
      "mean_token_accuracy": 0.18637175261974334,
      "num_tokens": 38331632.0,
      "step": 18930
    },
    {
      "entropy": 5.909644222259521,
      "epoch": 1.6547234116927378,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047280567186608563,
      "loss": 5.3976,
      "mean_token_accuracy": 0.18568094968795776,
      "num_tokens": 38342072.0,
      "step": 18935
    },
    {
      "entropy": 5.948240041732788,
      "epoch": 1.6551603600454428,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000472790819512652,
      "loss": 5.3451,
      "mean_token_accuracy": 0.192420394718647,
      "num_tokens": 38351105.0,
      "step": 18940
    },
    {
      "entropy": 5.966451215744018,
      "epoch": 1.6555973083981472,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004727759633654868,
      "loss": 5.4105,
      "mean_token_accuracy": 0.18400340527296066,
      "num_tokens": 38361521.0,
      "step": 18945
    },
    {
      "entropy": 5.8698570251464846,
      "epoch": 1.6560342567508521,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004727611034248751,
      "loss": 5.2929,
      "mean_token_accuracy": 0.19383472055196763,
      "num_tokens": 38370720.0,
      "step": 18950
    },
    {
      "entropy": 5.885596561431885,
      "epoch": 1.6564712051035566,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004727462396911017,
      "loss": 5.3763,
      "mean_token_accuracy": 0.19439753144979477,
      "num_tokens": 38380434.0,
      "step": 18955
    },
    {
      "entropy": 5.832363748550415,
      "epoch": 1.6569081534562615,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047273137216445176,
      "loss": 5.2628,
      "mean_token_accuracy": 0.21261580288410187,
      "num_tokens": 38390503.0,
      "step": 18960
    },
    {
      "entropy": 5.923024892807007,
      "epoch": 1.6573451018089662,
      "grad_norm": 1.0,
      "learning_rate": 0.0004727165008452106,
      "loss": 5.4402,
      "mean_token_accuracy": 0.18309035003185273,
      "num_tokens": 38401757.0,
      "step": 18965
    },
    {
      "entropy": 5.891209983825684,
      "epoch": 1.657782050161671,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004727016257336633,
      "loss": 5.3777,
      "mean_token_accuracy": 0.19525395333766937,
      "num_tokens": 38412130.0,
      "step": 18970
    },
    {
      "entropy": 5.886040496826172,
      "epoch": 1.6582189985143756,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047268674683009526,
      "loss": 5.337,
      "mean_token_accuracy": 0.19560080468654634,
      "num_tokens": 38422155.0,
      "step": 18975
    },
    {
      "entropy": 5.852785301208496,
      "epoch": 1.6586559468670803,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047267186413479185,
      "loss": 5.2733,
      "mean_token_accuracy": 0.2027613937854767,
      "num_tokens": 38431621.0,
      "step": 18980
    },
    {
      "entropy": 5.967005252838135,
      "epoch": 1.659092895219785,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047265697764803843,
      "loss": 5.4133,
      "mean_token_accuracy": 0.18312666416168213,
      "num_tokens": 38442354.0,
      "step": 18985
    },
    {
      "entropy": 5.895773696899414,
      "epoch": 1.6595298435724897,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047264208737012064,
      "loss": 5.3578,
      "mean_token_accuracy": 0.19051515758037568,
      "num_tokens": 38452932.0,
      "step": 18990
    },
    {
      "entropy": 5.913449716567993,
      "epoch": 1.6599667919251946,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004726271933013241,
      "loss": 5.3572,
      "mean_token_accuracy": 0.19448905885219575,
      "num_tokens": 38464442.0,
      "step": 18995
    },
    {
      "entropy": 5.929831409454346,
      "epoch": 1.660403740277899,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047261229544193434,
      "loss": 5.366,
      "mean_token_accuracy": 0.1934272289276123,
      "num_tokens": 38475718.0,
      "step": 19000
    },
    {
      "entropy": 5.925448083877564,
      "epoch": 1.660840688630604,
      "grad_norm": 1.0,
      "learning_rate": 0.00047259739379223726,
      "loss": 5.3437,
      "mean_token_accuracy": 0.18971968144178392,
      "num_tokens": 38485091.0,
      "step": 19005
    },
    {
      "entropy": 5.883081912994385,
      "epoch": 1.6612776369833084,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047258248835251867,
      "loss": 5.3565,
      "mean_token_accuracy": 0.1935834676027298,
      "num_tokens": 38495748.0,
      "step": 19010
    },
    {
      "entropy": 5.936135339736938,
      "epoch": 1.6617145853360134,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047256757912306435,
      "loss": 5.4051,
      "mean_token_accuracy": 0.1924675539135933,
      "num_tokens": 38506152.0,
      "step": 19015
    },
    {
      "entropy": 5.934273815155029,
      "epoch": 1.662151533688718,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004725526661041604,
      "loss": 5.2926,
      "mean_token_accuracy": 0.19426233768463136,
      "num_tokens": 38515722.0,
      "step": 19020
    },
    {
      "entropy": 5.8587440013885494,
      "epoch": 1.6625884820414227,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004725377492960927,
      "loss": 5.3278,
      "mean_token_accuracy": 0.198077292740345,
      "num_tokens": 38526270.0,
      "step": 19025
    },
    {
      "entropy": 5.9017129898071286,
      "epoch": 1.6630254303941274,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004725228286991475,
      "loss": 5.3233,
      "mean_token_accuracy": 0.19188827127218247,
      "num_tokens": 38537436.0,
      "step": 19030
    },
    {
      "entropy": 5.948308086395263,
      "epoch": 1.6634623787468321,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004725079043136111,
      "loss": 5.3468,
      "mean_token_accuracy": 0.20038892775774003,
      "num_tokens": 38547298.0,
      "step": 19035
    },
    {
      "entropy": 5.922237586975098,
      "epoch": 1.6638993270995368,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004724929761397694,
      "loss": 5.3557,
      "mean_token_accuracy": 0.192820905148983,
      "num_tokens": 38556465.0,
      "step": 19040
    },
    {
      "entropy": 5.876933956146241,
      "epoch": 1.6643362754522415,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000472478044177909,
      "loss": 5.38,
      "mean_token_accuracy": 0.19337158948183059,
      "num_tokens": 38568009.0,
      "step": 19045
    },
    {
      "entropy": 5.932672023773193,
      "epoch": 1.6647732238049464,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004724631084283163,
      "loss": 5.4009,
      "mean_token_accuracy": 0.19206545352935792,
      "num_tokens": 38578778.0,
      "step": 19050
    },
    {
      "entropy": 5.858993148803711,
      "epoch": 1.6652101721576509,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047244816889127763,
      "loss": 5.3101,
      "mean_token_accuracy": 0.19866302609443665,
      "num_tokens": 38588991.0,
      "step": 19055
    },
    {
      "entropy": 5.913332509994507,
      "epoch": 1.6656471205103558,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004724332255670797,
      "loss": 5.3097,
      "mean_token_accuracy": 0.19836800992488862,
      "num_tokens": 38598800.0,
      "step": 19060
    },
    {
      "entropy": 5.891294622421265,
      "epoch": 1.6660840688630603,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047241827845600896,
      "loss": 5.3348,
      "mean_token_accuracy": 0.1931643709540367,
      "num_tokens": 38608680.0,
      "step": 19065
    },
    {
      "entropy": 5.887544059753418,
      "epoch": 1.6665210172157652,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047240332755835217,
      "loss": 5.365,
      "mean_token_accuracy": 0.19061765521764756,
      "num_tokens": 38618402.0,
      "step": 19070
    },
    {
      "entropy": 6.03393030166626,
      "epoch": 1.6669579655684696,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047238837287439625,
      "loss": 5.3658,
      "mean_token_accuracy": 0.19136208444833755,
      "num_tokens": 38628370.0,
      "step": 19075
    },
    {
      "entropy": 5.9251946926116945,
      "epoch": 1.6673949139211746,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004723734144044278,
      "loss": 5.3052,
      "mean_token_accuracy": 0.199597629904747,
      "num_tokens": 38638138.0,
      "step": 19080
    },
    {
      "entropy": 5.939941692352295,
      "epoch": 1.6678318622738793,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047235845214873385,
      "loss": 5.3942,
      "mean_token_accuracy": 0.19106968641281127,
      "num_tokens": 38649412.0,
      "step": 19085
    },
    {
      "entropy": 5.95612530708313,
      "epoch": 1.668268810626584,
      "grad_norm": 0.921875,
      "learning_rate": 0.00047234348610760145,
      "loss": 5.3113,
      "mean_token_accuracy": 0.19061520844697952,
      "num_tokens": 38660048.0,
      "step": 19090
    },
    {
      "entropy": 5.901933574676514,
      "epoch": 1.6687057589792886,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004723285162813175,
      "loss": 5.2996,
      "mean_token_accuracy": 0.20195935368537904,
      "num_tokens": 38671246.0,
      "step": 19095
    },
    {
      "entropy": 5.905249214172363,
      "epoch": 1.6691427073319933,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004723135426701693,
      "loss": 5.3576,
      "mean_token_accuracy": 0.19758587926626206,
      "num_tokens": 38680880.0,
      "step": 19100
    },
    {
      "entropy": 5.871839809417724,
      "epoch": 1.669579655684698,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004722985652744438,
      "loss": 5.4061,
      "mean_token_accuracy": 0.19375296533107758,
      "num_tokens": 38691440.0,
      "step": 19105
    },
    {
      "entropy": 5.956132316589356,
      "epoch": 1.6700166040374027,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047228358409442863,
      "loss": 5.3491,
      "mean_token_accuracy": 0.1979931890964508,
      "num_tokens": 38700846.0,
      "step": 19110
    },
    {
      "entropy": 5.976755857467651,
      "epoch": 1.6704535523901076,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047226859913041084,
      "loss": 5.3665,
      "mean_token_accuracy": 0.1990732491016388,
      "num_tokens": 38712130.0,
      "step": 19115
    },
    {
      "entropy": 5.979550075531006,
      "epoch": 1.670890500742812,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047225361038267795,
      "loss": 5.4202,
      "mean_token_accuracy": 0.19446709156036376,
      "num_tokens": 38722482.0,
      "step": 19120
    },
    {
      "entropy": 5.845263290405273,
      "epoch": 1.671327449095517,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047223861785151744,
      "loss": 5.3207,
      "mean_token_accuracy": 0.19815725535154344,
      "num_tokens": 38731936.0,
      "step": 19125
    },
    {
      "entropy": 5.863184404373169,
      "epoch": 1.6717643974482215,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004722236215372169,
      "loss": 5.2582,
      "mean_token_accuracy": 0.19762666076421737,
      "num_tokens": 38741444.0,
      "step": 19130
    },
    {
      "entropy": 5.9993022918701175,
      "epoch": 1.6722013458009264,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000472208621440064,
      "loss": 5.4068,
      "mean_token_accuracy": 0.1802867904305458,
      "num_tokens": 38751454.0,
      "step": 19135
    },
    {
      "entropy": 5.985609722137451,
      "epoch": 1.672638294153631,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004721936175603463,
      "loss": 5.3627,
      "mean_token_accuracy": 0.19016223102808,
      "num_tokens": 38761373.0,
      "step": 19140
    },
    {
      "entropy": 5.9501955032348635,
      "epoch": 1.6730752425063358,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047217860989835184,
      "loss": 5.4235,
      "mean_token_accuracy": 0.1938718006014824,
      "num_tokens": 38772047.0,
      "step": 19145
    },
    {
      "entropy": 5.949259424209595,
      "epoch": 1.6735121908590405,
      "grad_norm": 1.25,
      "learning_rate": 0.0004721635984543682,
      "loss": 5.2965,
      "mean_token_accuracy": 0.20174793303012847,
      "num_tokens": 38781393.0,
      "step": 19150
    },
    {
      "entropy": 5.950770378112793,
      "epoch": 1.6739491392117452,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004721485832286835,
      "loss": 5.3205,
      "mean_token_accuracy": 0.1953478366136551,
      "num_tokens": 38791104.0,
      "step": 19155
    },
    {
      "entropy": 5.879067754745483,
      "epoch": 1.6743860875644498,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004721335642215856,
      "loss": 5.4011,
      "mean_token_accuracy": 0.19291983544826508,
      "num_tokens": 38801855.0,
      "step": 19160
    },
    {
      "entropy": 5.891265106201172,
      "epoch": 1.6748230359171545,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004721185414333627,
      "loss": 5.2922,
      "mean_token_accuracy": 0.20018799155950545,
      "num_tokens": 38811983.0,
      "step": 19165
    },
    {
      "entropy": 5.927467584609985,
      "epoch": 1.6752599842698594,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004721035148643029,
      "loss": 5.3345,
      "mean_token_accuracy": 0.19032654762268067,
      "num_tokens": 38821762.0,
      "step": 19170
    },
    {
      "entropy": 5.8646674156188965,
      "epoch": 1.675696932622564,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004720884845146944,
      "loss": 5.3102,
      "mean_token_accuracy": 0.19735245108604432,
      "num_tokens": 38831840.0,
      "step": 19175
    },
    {
      "entropy": 5.893741464614868,
      "epoch": 1.6761338809752688,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047207345038482545,
      "loss": 5.2906,
      "mean_token_accuracy": 0.2004453256726265,
      "num_tokens": 38841506.0,
      "step": 19180
    },
    {
      "entropy": 5.9476100444793705,
      "epoch": 1.6765708293279733,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004720584124749844,
      "loss": 5.3921,
      "mean_token_accuracy": 0.19148998856544494,
      "num_tokens": 38851793.0,
      "step": 19185
    },
    {
      "entropy": 5.865274286270141,
      "epoch": 1.6770077776806782,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047204337078545985,
      "loss": 5.3015,
      "mean_token_accuracy": 0.19971559047698975,
      "num_tokens": 38861958.0,
      "step": 19190
    },
    {
      "entropy": 5.9249656200408936,
      "epoch": 1.677444726033383,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004720283253165402,
      "loss": 5.3064,
      "mean_token_accuracy": 0.19781468510627748,
      "num_tokens": 38872181.0,
      "step": 19195
    },
    {
      "entropy": 5.941156196594238,
      "epoch": 1.6778816743860876,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047201327606851394,
      "loss": 5.3491,
      "mean_token_accuracy": 0.19894104450941086,
      "num_tokens": 38882423.0,
      "step": 19200
    },
    {
      "entropy": 5.917841196060181,
      "epoch": 1.6783186227387923,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004719982230416699,
      "loss": 5.3566,
      "mean_token_accuracy": 0.19289954006671906,
      "num_tokens": 38892833.0,
      "step": 19205
    },
    {
      "entropy": 5.97090973854065,
      "epoch": 1.678755571091497,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004719831662362967,
      "loss": 5.4089,
      "mean_token_accuracy": 0.19201980233192445,
      "num_tokens": 38904037.0,
      "step": 19210
    },
    {
      "entropy": 5.896849060058594,
      "epoch": 1.6791925194442017,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004719681056526831,
      "loss": 5.3328,
      "mean_token_accuracy": 0.19788239300251007,
      "num_tokens": 38914194.0,
      "step": 19215
    },
    {
      "entropy": 5.842992448806763,
      "epoch": 1.6796294677969064,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047195304129111804,
      "loss": 5.3719,
      "mean_token_accuracy": 0.1884673446416855,
      "num_tokens": 38925354.0,
      "step": 19220
    },
    {
      "entropy": 5.978987503051758,
      "epoch": 1.6800664161496113,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047193797315189047,
      "loss": 5.3941,
      "mean_token_accuracy": 0.19128120094537734,
      "num_tokens": 38936170.0,
      "step": 19225
    },
    {
      "entropy": 6.0059105396270756,
      "epoch": 1.6805033645023157,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004719229012352894,
      "loss": 5.3723,
      "mean_token_accuracy": 0.1933169350028038,
      "num_tokens": 38945316.0,
      "step": 19230
    },
    {
      "entropy": 5.963331985473633,
      "epoch": 1.6809403128550207,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047190782554160386,
      "loss": 5.3606,
      "mean_token_accuracy": 0.19202713817358016,
      "num_tokens": 38955444.0,
      "step": 19235
    },
    {
      "entropy": 5.873808479309082,
      "epoch": 1.6813772612077251,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004718927460711231,
      "loss": 5.3189,
      "mean_token_accuracy": 0.20434165447950364,
      "num_tokens": 38965202.0,
      "step": 19240
    },
    {
      "entropy": 5.899515151977539,
      "epoch": 1.68181420956043,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047187766282413623,
      "loss": 5.3483,
      "mean_token_accuracy": 0.19664521068334578,
      "num_tokens": 38975822.0,
      "step": 19245
    },
    {
      "entropy": 5.933702707290649,
      "epoch": 1.6822511579131345,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004718625758009327,
      "loss": 5.3415,
      "mean_token_accuracy": 0.19507901072502137,
      "num_tokens": 38985890.0,
      "step": 19250
    },
    {
      "entropy": 5.910011577606201,
      "epoch": 1.6826881062658394,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047184748500180175,
      "loss": 5.3081,
      "mean_token_accuracy": 0.20093366503715515,
      "num_tokens": 38996374.0,
      "step": 19255
    },
    {
      "entropy": 5.907700300216675,
      "epoch": 1.6831250546185441,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004718323904270329,
      "loss": 5.3392,
      "mean_token_accuracy": 0.1956373780965805,
      "num_tokens": 39006371.0,
      "step": 19260
    },
    {
      "entropy": 5.918034982681275,
      "epoch": 1.6835620029712488,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004718172920769157,
      "loss": 5.2988,
      "mean_token_accuracy": 0.19733487367630004,
      "num_tokens": 39015514.0,
      "step": 19265
    },
    {
      "entropy": 5.944803380966187,
      "epoch": 1.6839989513239535,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004718021899517397,
      "loss": 5.3502,
      "mean_token_accuracy": 0.19775994569063188,
      "num_tokens": 39025700.0,
      "step": 19270
    },
    {
      "entropy": 5.872176265716552,
      "epoch": 1.6844358996766582,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047178708405179454,
      "loss": 5.3196,
      "mean_token_accuracy": 0.20453754663467408,
      "num_tokens": 39036157.0,
      "step": 19275
    },
    {
      "entropy": 5.966128969192505,
      "epoch": 1.6848728480293629,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047177197437737003,
      "loss": 5.4782,
      "mean_token_accuracy": 0.18780629485845565,
      "num_tokens": 39046044.0,
      "step": 19280
    },
    {
      "entropy": 5.949373531341553,
      "epoch": 1.6853097963820676,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004717568609287559,
      "loss": 5.3949,
      "mean_token_accuracy": 0.18495397865772248,
      "num_tokens": 39056034.0,
      "step": 19285
    },
    {
      "entropy": 5.8996528625488285,
      "epoch": 1.6857467447347725,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004717417437062421,
      "loss": 5.3456,
      "mean_token_accuracy": 0.1967225566506386,
      "num_tokens": 39065687.0,
      "step": 19290
    },
    {
      "entropy": 5.9524641036987305,
      "epoch": 1.686183693087477,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047172662271011855,
      "loss": 5.3517,
      "mean_token_accuracy": 0.19141679555177687,
      "num_tokens": 39077107.0,
      "step": 19295
    },
    {
      "entropy": 5.949234962463379,
      "epoch": 1.6866206414401819,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047171149794067533,
      "loss": 5.3419,
      "mean_token_accuracy": 0.19734421670436858,
      "num_tokens": 39087220.0,
      "step": 19300
    },
    {
      "entropy": 5.913172483444214,
      "epoch": 1.6870575897928863,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004716963693982025,
      "loss": 5.2964,
      "mean_token_accuracy": 0.1979448229074478,
      "num_tokens": 39097706.0,
      "step": 19305
    },
    {
      "entropy": 5.9279498100280765,
      "epoch": 1.6874945381455912,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004716812370829902,
      "loss": 5.3608,
      "mean_token_accuracy": 0.19019176959991455,
      "num_tokens": 39108044.0,
      "step": 19310
    },
    {
      "entropy": 5.934221076965332,
      "epoch": 1.687931486498296,
      "grad_norm": 1.125,
      "learning_rate": 0.0004716661009953287,
      "loss": 5.3692,
      "mean_token_accuracy": 0.18748905807733535,
      "num_tokens": 39116917.0,
      "step": 19315
    },
    {
      "entropy": 5.8858812808990475,
      "epoch": 1.6883684348510006,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004716509611355084,
      "loss": 5.3314,
      "mean_token_accuracy": 0.20164989531040192,
      "num_tokens": 39126770.0,
      "step": 19320
    },
    {
      "entropy": 5.881637477874756,
      "epoch": 1.6888053832037053,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047163581750381956,
      "loss": 5.3508,
      "mean_token_accuracy": 0.18900355100631713,
      "num_tokens": 39137406.0,
      "step": 19325
    },
    {
      "entropy": 5.936739110946656,
      "epoch": 1.68924233155641,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047162067010055275,
      "loss": 5.3405,
      "mean_token_accuracy": 0.19583241939544677,
      "num_tokens": 39146950.0,
      "step": 19330
    },
    {
      "entropy": 5.956313562393189,
      "epoch": 1.6896792799091147,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004716055189259984,
      "loss": 5.2802,
      "mean_token_accuracy": 0.19611471593379975,
      "num_tokens": 39157382.0,
      "step": 19335
    },
    {
      "entropy": 5.900577211380005,
      "epoch": 1.6901162282618194,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047159036398044727,
      "loss": 5.3187,
      "mean_token_accuracy": 0.19703029543161393,
      "num_tokens": 39166641.0,
      "step": 19340
    },
    {
      "entropy": 5.914486980438232,
      "epoch": 1.6905531766145243,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004715752052641899,
      "loss": 5.3902,
      "mean_token_accuracy": 0.19560399204492568,
      "num_tokens": 39176731.0,
      "step": 19345
    },
    {
      "entropy": 5.9629960536956785,
      "epoch": 1.6909901249672288,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004715600427775171,
      "loss": 5.4023,
      "mean_token_accuracy": 0.18968440890312194,
      "num_tokens": 39187202.0,
      "step": 19350
    },
    {
      "entropy": 5.963880300521851,
      "epoch": 1.6914270733199337,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004715448765207196,
      "loss": 5.385,
      "mean_token_accuracy": 0.19411734491586685,
      "num_tokens": 39197302.0,
      "step": 19355
    },
    {
      "entropy": 5.962043333053589,
      "epoch": 1.6918640216726382,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004715297064940885,
      "loss": 5.3648,
      "mean_token_accuracy": 0.19435934871435165,
      "num_tokens": 39208382.0,
      "step": 19360
    },
    {
      "entropy": 5.916847372055054,
      "epoch": 1.692300970025343,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004715145326979146,
      "loss": 5.3415,
      "mean_token_accuracy": 0.19160952121019365,
      "num_tokens": 39218529.0,
      "step": 19365
    },
    {
      "entropy": 5.892445659637451,
      "epoch": 1.6927379183780478,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047149935513248896,
      "loss": 5.3869,
      "mean_token_accuracy": 0.19605011492967606,
      "num_tokens": 39228976.0,
      "step": 19370
    },
    {
      "entropy": 5.890560245513916,
      "epoch": 1.6931748667307525,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004714841737981028,
      "loss": 5.2705,
      "mean_token_accuracy": 0.19913362562656403,
      "num_tokens": 39238921.0,
      "step": 19375
    },
    {
      "entropy": 5.911904335021973,
      "epoch": 1.6936118150834572,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047146898869504717,
      "loss": 5.285,
      "mean_token_accuracy": 0.19366183280944824,
      "num_tokens": 39247978.0,
      "step": 19380
    },
    {
      "entropy": 5.878648996353149,
      "epoch": 1.6940487634361618,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047145379982361336,
      "loss": 5.3466,
      "mean_token_accuracy": 0.19457070976495744,
      "num_tokens": 39257487.0,
      "step": 19385
    },
    {
      "entropy": 5.878313302993774,
      "epoch": 1.6944857117888665,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047143860718409285,
      "loss": 5.3084,
      "mean_token_accuracy": 0.19508877098560334,
      "num_tokens": 39267819.0,
      "step": 19390
    },
    {
      "entropy": 5.9759973049163815,
      "epoch": 1.6949226601415712,
      "grad_norm": 1.125,
      "learning_rate": 0.00047142341077677685,
      "loss": 5.4183,
      "mean_token_accuracy": 0.18929689675569533,
      "num_tokens": 39277673.0,
      "step": 19395
    },
    {
      "entropy": 5.932804250717163,
      "epoch": 1.6953596084942761,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004714082106019569,
      "loss": 5.2645,
      "mean_token_accuracy": 0.2021329805254936,
      "num_tokens": 39287340.0,
      "step": 19400
    },
    {
      "entropy": 5.922144079208374,
      "epoch": 1.6957965568469806,
      "grad_norm": 1.0,
      "learning_rate": 0.0004713930066599246,
      "loss": 5.3914,
      "mean_token_accuracy": 0.19910265803337096,
      "num_tokens": 39298277.0,
      "step": 19405
    },
    {
      "entropy": 5.937264823913575,
      "epoch": 1.6962335051996855,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047137779895097145,
      "loss": 5.36,
      "mean_token_accuracy": 0.191498102247715,
      "num_tokens": 39307996.0,
      "step": 19410
    },
    {
      "entropy": 5.9264723777771,
      "epoch": 1.69667045355239,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047136258747538926,
      "loss": 5.3397,
      "mean_token_accuracy": 0.19877322614192963,
      "num_tokens": 39317327.0,
      "step": 19415
    },
    {
      "entropy": 5.874369287490845,
      "epoch": 1.697107401905095,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004713473722334697,
      "loss": 5.3086,
      "mean_token_accuracy": 0.20526477992534636,
      "num_tokens": 39326728.0,
      "step": 19420
    },
    {
      "entropy": 5.913986873626709,
      "epoch": 1.6975443502577994,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047133215322550464,
      "loss": 5.4633,
      "mean_token_accuracy": 0.18921436071395875,
      "num_tokens": 39336606.0,
      "step": 19425
    },
    {
      "entropy": 5.9624532699584964,
      "epoch": 1.6979812986105043,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004713169304517861,
      "loss": 5.3636,
      "mean_token_accuracy": 0.19012525528669358,
      "num_tokens": 39345622.0,
      "step": 19430
    },
    {
      "entropy": 5.9221885204315186,
      "epoch": 1.698418246963209,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047130170391260597,
      "loss": 5.3453,
      "mean_token_accuracy": 0.19442405700683593,
      "num_tokens": 39356868.0,
      "step": 19435
    },
    {
      "entropy": 5.885597467422485,
      "epoch": 1.6988551953159137,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004712864736082562,
      "loss": 5.2664,
      "mean_token_accuracy": 0.19978632926940917,
      "num_tokens": 39366866.0,
      "step": 19440
    },
    {
      "entropy": 5.914266967773438,
      "epoch": 1.6992921436686184,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047127123953902907,
      "loss": 5.3421,
      "mean_token_accuracy": 0.1925971433520317,
      "num_tokens": 39377303.0,
      "step": 19445
    },
    {
      "entropy": 5.894758415222168,
      "epoch": 1.699729092021323,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004712560017052167,
      "loss": 5.2927,
      "mean_token_accuracy": 0.20235924273729325,
      "num_tokens": 39387191.0,
      "step": 19450
    },
    {
      "entropy": 5.829927539825439,
      "epoch": 1.7001660403740277,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004712407601071114,
      "loss": 5.3138,
      "mean_token_accuracy": 0.20642631202936174,
      "num_tokens": 39397416.0,
      "step": 19455
    },
    {
      "entropy": 5.904181480407715,
      "epoch": 1.7006029887267324,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047122551474500555,
      "loss": 5.3403,
      "mean_token_accuracy": 0.19162139296531677,
      "num_tokens": 39407855.0,
      "step": 19460
    },
    {
      "entropy": 5.939141416549683,
      "epoch": 1.7010399370794373,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047121026561919137,
      "loss": 5.3991,
      "mean_token_accuracy": 0.18919705152511596,
      "num_tokens": 39417735.0,
      "step": 19465
    },
    {
      "entropy": 5.9519411563873295,
      "epoch": 1.7014768854321418,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004711950127299616,
      "loss": 5.3164,
      "mean_token_accuracy": 0.1959323614835739,
      "num_tokens": 39427641.0,
      "step": 19470
    },
    {
      "entropy": 5.888665246963501,
      "epoch": 1.7019138337848467,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004711797560776087,
      "loss": 5.2409,
      "mean_token_accuracy": 0.2104995146393776,
      "num_tokens": 39437372.0,
      "step": 19475
    },
    {
      "entropy": 5.962058782577515,
      "epoch": 1.7023507821375512,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047116449566242516,
      "loss": 5.3466,
      "mean_token_accuracy": 0.20320177525281907,
      "num_tokens": 39447800.0,
      "step": 19480
    },
    {
      "entropy": 5.927330446243286,
      "epoch": 1.7027877304902561,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004711492314847038,
      "loss": 5.3729,
      "mean_token_accuracy": 0.19514647275209426,
      "num_tokens": 39457310.0,
      "step": 19485
    },
    {
      "entropy": 5.891929721832275,
      "epoch": 1.7032246788429608,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047113396354473753,
      "loss": 5.3706,
      "mean_token_accuracy": 0.19220768064260482,
      "num_tokens": 39468579.0,
      "step": 19490
    },
    {
      "entropy": 5.9402440071105955,
      "epoch": 1.7036616271956655,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000471118691842819,
      "loss": 5.2988,
      "mean_token_accuracy": 0.1990757942199707,
      "num_tokens": 39479027.0,
      "step": 19495
    },
    {
      "entropy": 5.861589622497559,
      "epoch": 1.7040985755483702,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047110341637924126,
      "loss": 5.3173,
      "mean_token_accuracy": 0.20968015640974044,
      "num_tokens": 39488651.0,
      "step": 19500
    },
    {
      "entropy": 5.912119817733765,
      "epoch": 1.7045355239010749,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004710881371542972,
      "loss": 5.3757,
      "mean_token_accuracy": 0.1949328824877739,
      "num_tokens": 39498633.0,
      "step": 19505
    },
    {
      "entropy": 5.86829252243042,
      "epoch": 1.7049724722537796,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047107285416827984,
      "loss": 5.2729,
      "mean_token_accuracy": 0.20579459369182587,
      "num_tokens": 39509351.0,
      "step": 19510
    },
    {
      "entropy": 5.876588153839111,
      "epoch": 1.7054094206064843,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004710575674214824,
      "loss": 5.3257,
      "mean_token_accuracy": 0.19901105761528015,
      "num_tokens": 39519821.0,
      "step": 19515
    },
    {
      "entropy": 5.900717258453369,
      "epoch": 1.7058463689591892,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004710422769141982,
      "loss": 5.3404,
      "mean_token_accuracy": 0.19487130790948867,
      "num_tokens": 39530040.0,
      "step": 19520
    },
    {
      "entropy": 5.891128635406494,
      "epoch": 1.7062833173118936,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004710269826467203,
      "loss": 5.3513,
      "mean_token_accuracy": 0.19063884913921356,
      "num_tokens": 39540669.0,
      "step": 19525
    },
    {
      "entropy": 5.966311836242676,
      "epoch": 1.7067202656645986,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047101168461934225,
      "loss": 5.3491,
      "mean_token_accuracy": 0.19380797445774078,
      "num_tokens": 39550651.0,
      "step": 19530
    },
    {
      "entropy": 5.919613218307495,
      "epoch": 1.707157214017303,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004709963828323573,
      "loss": 5.322,
      "mean_token_accuracy": 0.18836230635643006,
      "num_tokens": 39561748.0,
      "step": 19535
    },
    {
      "entropy": 5.928906965255737,
      "epoch": 1.707594162370008,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004709810772860591,
      "loss": 5.2976,
      "mean_token_accuracy": 0.19958372563123702,
      "num_tokens": 39571328.0,
      "step": 19540
    },
    {
      "entropy": 5.82282600402832,
      "epoch": 1.7080311107227124,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000470965767980741,
      "loss": 5.3096,
      "mean_token_accuracy": 0.20180499106645583,
      "num_tokens": 39581931.0,
      "step": 19545
    },
    {
      "entropy": 5.917238712310791,
      "epoch": 1.7084680590754173,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047095045491669696,
      "loss": 5.44,
      "mean_token_accuracy": 0.1888043314218521,
      "num_tokens": 39592818.0,
      "step": 19550
    },
    {
      "entropy": 5.9408180713653564,
      "epoch": 1.708905007428122,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047093513809422043,
      "loss": 5.3439,
      "mean_token_accuracy": 0.1980011582374573,
      "num_tokens": 39602867.0,
      "step": 19555
    },
    {
      "entropy": 5.916329574584961,
      "epoch": 1.7093419557808267,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004709198175136053,
      "loss": 5.3603,
      "mean_token_accuracy": 0.1935013085603714,
      "num_tokens": 39613185.0,
      "step": 19560
    },
    {
      "entropy": 5.8423319339752195,
      "epoch": 1.7097789041335314,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004709044931751454,
      "loss": 5.2997,
      "mean_token_accuracy": 0.20169659554958344,
      "num_tokens": 39623262.0,
      "step": 19565
    },
    {
      "entropy": 5.969347953796387,
      "epoch": 1.710215852486236,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004708891650791347,
      "loss": 5.3397,
      "mean_token_accuracy": 0.19467520117759704,
      "num_tokens": 39633184.0,
      "step": 19570
    },
    {
      "entropy": 5.9565776824951175,
      "epoch": 1.710652800838941,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004708738332258671,
      "loss": 5.3838,
      "mean_token_accuracy": 0.19274740666151047,
      "num_tokens": 39643778.0,
      "step": 19575
    },
    {
      "entropy": 5.902467107772827,
      "epoch": 1.7110897491916455,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004708584976156368,
      "loss": 5.281,
      "mean_token_accuracy": 0.19660976976156236,
      "num_tokens": 39653519.0,
      "step": 19580
    },
    {
      "entropy": 5.8894885063171385,
      "epoch": 1.7115266975443504,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047084315824873793,
      "loss": 5.351,
      "mean_token_accuracy": 0.19538945108652114,
      "num_tokens": 39663824.0,
      "step": 19585
    },
    {
      "entropy": 5.919812536239624,
      "epoch": 1.7119636458970549,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004708278151254646,
      "loss": 5.3668,
      "mean_token_accuracy": 0.19098558723926545,
      "num_tokens": 39673631.0,
      "step": 19590
    },
    {
      "entropy": 6.013787031173706,
      "epoch": 1.7124005942497598,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004708124682461112,
      "loss": 5.416,
      "mean_token_accuracy": 0.18662769347429276,
      "num_tokens": 39683118.0,
      "step": 19595
    },
    {
      "entropy": 5.938241910934448,
      "epoch": 1.7128375426024642,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047079711761097197,
      "loss": 5.3331,
      "mean_token_accuracy": 0.19841467291116716,
      "num_tokens": 39693947.0,
      "step": 19600
    },
    {
      "entropy": 5.943591117858887,
      "epoch": 1.7132744909551691,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004707817632203415,
      "loss": 5.3565,
      "mean_token_accuracy": 0.18912701457738876,
      "num_tokens": 39703874.0,
      "step": 19605
    },
    {
      "entropy": 5.862623929977417,
      "epoch": 1.7137114393078738,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047076640507451425,
      "loss": 5.3314,
      "mean_token_accuracy": 0.19495512694120407,
      "num_tokens": 39714997.0,
      "step": 19610
    },
    {
      "entropy": 5.964521265029907,
      "epoch": 1.7141483876605785,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004707510431737847,
      "loss": 5.3169,
      "mean_token_accuracy": 0.19372852593660356,
      "num_tokens": 39726042.0,
      "step": 19615
    },
    {
      "entropy": 5.964812898635865,
      "epoch": 1.7145853360132832,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004707356775184476,
      "loss": 5.3516,
      "mean_token_accuracy": 0.19369154572486877,
      "num_tokens": 39736733.0,
      "step": 19620
    },
    {
      "entropy": 5.986809206008911,
      "epoch": 1.715022284365988,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004707203081087977,
      "loss": 5.4422,
      "mean_token_accuracy": 0.1856246680021286,
      "num_tokens": 39746811.0,
      "step": 19625
    },
    {
      "entropy": 5.961586999893188,
      "epoch": 1.7154592327186926,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004707049349451297,
      "loss": 5.3674,
      "mean_token_accuracy": 0.19513397961854934,
      "num_tokens": 39757371.0,
      "step": 19630
    },
    {
      "entropy": 5.87864727973938,
      "epoch": 1.7158961810713973,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004706895580277384,
      "loss": 5.3354,
      "mean_token_accuracy": 0.1985072076320648,
      "num_tokens": 39767430.0,
      "step": 19635
    },
    {
      "entropy": 5.920934677124023,
      "epoch": 1.7163331294241022,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047067417735691884,
      "loss": 5.3216,
      "mean_token_accuracy": 0.19903651773929595,
      "num_tokens": 39777270.0,
      "step": 19640
    },
    {
      "entropy": 5.89995174407959,
      "epoch": 1.7167700777768067,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004706587929329662,
      "loss": 5.3332,
      "mean_token_accuracy": 0.19905025511980057,
      "num_tokens": 39787647.0,
      "step": 19645
    },
    {
      "entropy": 5.912028789520264,
      "epoch": 1.7172070261295116,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047064340475617527,
      "loss": 5.385,
      "mean_token_accuracy": 0.19222538024187089,
      "num_tokens": 39797315.0,
      "step": 19650
    },
    {
      "entropy": 5.920211362838745,
      "epoch": 1.717643974482216,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004706280128268413,
      "loss": 5.2678,
      "mean_token_accuracy": 0.19694646149873735,
      "num_tokens": 39806625.0,
      "step": 19655
    },
    {
      "entropy": 5.909769725799561,
      "epoch": 1.718080922834921,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004706126171452596,
      "loss": 5.4011,
      "mean_token_accuracy": 0.1945561945438385,
      "num_tokens": 39818113.0,
      "step": 19660
    },
    {
      "entropy": 5.9059347152709964,
      "epoch": 1.7185178711876257,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004705972177117254,
      "loss": 5.3857,
      "mean_token_accuracy": 0.18980478197336198,
      "num_tokens": 39827897.0,
      "step": 19665
    },
    {
      "entropy": 5.913657903671265,
      "epoch": 1.7189548195403304,
      "grad_norm": 1.125,
      "learning_rate": 0.00047058181452653403,
      "loss": 5.2235,
      "mean_token_accuracy": 0.20581398606300355,
      "num_tokens": 39837209.0,
      "step": 19670
    },
    {
      "entropy": 5.9327037811279295,
      "epoch": 1.719391767893035,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000470566407589981,
      "loss": 5.3332,
      "mean_token_accuracy": 0.19629070907831192,
      "num_tokens": 39848023.0,
      "step": 19675
    },
    {
      "entropy": 5.920609760284424,
      "epoch": 1.7198287162457397,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004705509969023618,
      "loss": 5.391,
      "mean_token_accuracy": 0.19517710506916047,
      "num_tokens": 39857992.0,
      "step": 19680
    },
    {
      "entropy": 5.967924404144287,
      "epoch": 1.7202656645984444,
      "grad_norm": 1.078125,
      "learning_rate": 0.000470535582463972,
      "loss": 5.401,
      "mean_token_accuracy": 0.19185743033885955,
      "num_tokens": 39867927.0,
      "step": 19685
    },
    {
      "entropy": 5.90995683670044,
      "epoch": 1.7207026129511491,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004705201642751073,
      "loss": 5.3128,
      "mean_token_accuracy": 0.2028667762875557,
      "num_tokens": 39877327.0,
      "step": 19690
    },
    {
      "entropy": 5.864987754821778,
      "epoch": 1.721139561303854,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047050474233606334,
      "loss": 5.3428,
      "mean_token_accuracy": 0.19404876232147217,
      "num_tokens": 39887369.0,
      "step": 19695
    },
    {
      "entropy": 5.94201626777649,
      "epoch": 1.7215765096565585,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000470489316647136,
      "loss": 5.3591,
      "mean_token_accuracy": 0.19538068473339082,
      "num_tokens": 39897530.0,
      "step": 19700
    },
    {
      "entropy": 5.9223371028900145,
      "epoch": 1.7220134580092634,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004704738872086212,
      "loss": 5.3454,
      "mean_token_accuracy": 0.19078458696603776,
      "num_tokens": 39907234.0,
      "step": 19705
    },
    {
      "entropy": 5.988466358184814,
      "epoch": 1.722450406361968,
      "grad_norm": 1.125,
      "learning_rate": 0.0004704584540208148,
      "loss": 5.379,
      "mean_token_accuracy": 0.19098335653543472,
      "num_tokens": 39917940.0,
      "step": 19710
    },
    {
      "entropy": 5.8972029209136965,
      "epoch": 1.7228873547146728,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004704430170840127,
      "loss": 5.2894,
      "mean_token_accuracy": 0.20368268936872483,
      "num_tokens": 39927569.0,
      "step": 19715
    },
    {
      "entropy": 5.831471252441406,
      "epoch": 1.7233243030673773,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004704275763985113,
      "loss": 5.3696,
      "mean_token_accuracy": 0.19795337021350862,
      "num_tokens": 39938302.0,
      "step": 19720
    },
    {
      "entropy": 5.9457926750183105,
      "epoch": 1.7237612514200822,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004704121319646065,
      "loss": 5.3115,
      "mean_token_accuracy": 0.20155392587184906,
      "num_tokens": 39948133.0,
      "step": 19725
    },
    {
      "entropy": 5.917707443237305,
      "epoch": 1.7241981997727869,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047039668378259455,
      "loss": 5.2846,
      "mean_token_accuracy": 0.20324569791555405,
      "num_tokens": 39958924.0,
      "step": 19730
    },
    {
      "entropy": 5.9087999820709225,
      "epoch": 1.7246351481254916,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047038123185277187,
      "loss": 5.2623,
      "mean_token_accuracy": 0.1990640491247177,
      "num_tokens": 39968578.0,
      "step": 19735
    },
    {
      "entropy": 5.847929954528809,
      "epoch": 1.7250720964781963,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047036577617543484,
      "loss": 5.2622,
      "mean_token_accuracy": 0.2098663792014122,
      "num_tokens": 39978112.0,
      "step": 19740
    },
    {
      "entropy": 5.897514867782593,
      "epoch": 1.725509044830901,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047035031675087983,
      "loss": 5.3647,
      "mean_token_accuracy": 0.18783574551343918,
      "num_tokens": 39988071.0,
      "step": 19745
    },
    {
      "entropy": 5.985564756393432,
      "epoch": 1.7259459931836056,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004703348535794034,
      "loss": 5.3706,
      "mean_token_accuracy": 0.19210540652275085,
      "num_tokens": 39998112.0,
      "step": 19750
    },
    {
      "entropy": 5.923358631134033,
      "epoch": 1.7263829415363103,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047031938666130206,
      "loss": 5.3322,
      "mean_token_accuracy": 0.1998075544834137,
      "num_tokens": 40008869.0,
      "step": 19755
    },
    {
      "entropy": 5.924812459945679,
      "epoch": 1.7268198898890152,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004703039159968726,
      "loss": 5.347,
      "mean_token_accuracy": 0.19793977737426757,
      "num_tokens": 40018711.0,
      "step": 19760
    },
    {
      "entropy": 5.943233442306519,
      "epoch": 1.7272568382417197,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047028844158641166,
      "loss": 5.359,
      "mean_token_accuracy": 0.19197206050157548,
      "num_tokens": 40029209.0,
      "step": 19765
    },
    {
      "entropy": 5.933167934417725,
      "epoch": 1.7276937865944246,
      "grad_norm": 1.0625,
      "learning_rate": 0.000470272963430216,
      "loss": 5.3278,
      "mean_token_accuracy": 0.20001130849123,
      "num_tokens": 40039071.0,
      "step": 19770
    },
    {
      "entropy": 5.889946031570434,
      "epoch": 1.728130734947129,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004702574815285827,
      "loss": 5.2971,
      "mean_token_accuracy": 0.19736414700746535,
      "num_tokens": 40049765.0,
      "step": 19775
    },
    {
      "entropy": 5.876913690567017,
      "epoch": 1.728567683299834,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004702419958818085,
      "loss": 5.2879,
      "mean_token_accuracy": 0.2007022202014923,
      "num_tokens": 40060020.0,
      "step": 19780
    },
    {
      "entropy": 5.908238554000855,
      "epoch": 1.7290046316525387,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047022650649019063,
      "loss": 5.3138,
      "mean_token_accuracy": 0.1955873742699623,
      "num_tokens": 40070081.0,
      "step": 19785
    },
    {
      "entropy": 5.950274991989136,
      "epoch": 1.7294415800052434,
      "grad_norm": 0.984375,
      "learning_rate": 0.000470211013354026,
      "loss": 5.3815,
      "mean_token_accuracy": 0.19015220403671265,
      "num_tokens": 40080532.0,
      "step": 19790
    },
    {
      "entropy": 5.965419673919678,
      "epoch": 1.729878528357948,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004701955164736118,
      "loss": 5.3373,
      "mean_token_accuracy": 0.19806033223867417,
      "num_tokens": 40091288.0,
      "step": 19795
    },
    {
      "entropy": 5.910866594314575,
      "epoch": 1.7303154767106528,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004701800158492454,
      "loss": 5.3192,
      "mean_token_accuracy": 0.1957208916544914,
      "num_tokens": 40100461.0,
      "step": 19800
    },
    {
      "entropy": 5.883841896057129,
      "epoch": 1.7307524250633575,
      "grad_norm": 1.015625,
      "learning_rate": 0.000470164511481224,
      "loss": 5.2933,
      "mean_token_accuracy": 0.19903205037117006,
      "num_tokens": 40109627.0,
      "step": 19805
    },
    {
      "entropy": 5.872559118270874,
      "epoch": 1.7311893734160622,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000470149003369845,
      "loss": 5.3389,
      "mean_token_accuracy": 0.19951961040496827,
      "num_tokens": 40119183.0,
      "step": 19810
    },
    {
      "entropy": 5.9243165969848635,
      "epoch": 1.731626321768767,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004701334915154058,
      "loss": 5.3505,
      "mean_token_accuracy": 0.19762726426124572,
      "num_tokens": 40129569.0,
      "step": 19815
    },
    {
      "entropy": 5.950009298324585,
      "epoch": 1.7320632701214715,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047011797591820404,
      "loss": 5.2863,
      "mean_token_accuracy": 0.20163607597351074,
      "num_tokens": 40139618.0,
      "step": 19820
    },
    {
      "entropy": 5.856990480422974,
      "epoch": 1.7325002184741765,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047010245657853726,
      "loss": 5.249,
      "mean_token_accuracy": 0.2041134938597679,
      "num_tokens": 40149349.0,
      "step": 19825
    },
    {
      "entropy": 5.972581434249878,
      "epoch": 1.732937166826881,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004700869334967031,
      "loss": 5.3558,
      "mean_token_accuracy": 0.19736634641885759,
      "num_tokens": 40159161.0,
      "step": 19830
    },
    {
      "entropy": 5.9161123752594,
      "epoch": 1.7333741151795858,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047007140667299944,
      "loss": 5.3481,
      "mean_token_accuracy": 0.19506853818893433,
      "num_tokens": 40169006.0,
      "step": 19835
    },
    {
      "entropy": 5.920161533355713,
      "epoch": 1.7338110635322905,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047005587610772384,
      "loss": 5.3058,
      "mean_token_accuracy": 0.19567049145698548,
      "num_tokens": 40179964.0,
      "step": 19840
    },
    {
      "entropy": 5.932272100448609,
      "epoch": 1.7342480118849952,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047004034180117446,
      "loss": 5.3574,
      "mean_token_accuracy": 0.1924600139260292,
      "num_tokens": 40190598.0,
      "step": 19845
    },
    {
      "entropy": 5.93349256515503,
      "epoch": 1.7346849602377,
      "grad_norm": 1.140625,
      "learning_rate": 0.000470024803753649,
      "loss": 5.3743,
      "mean_token_accuracy": 0.19357751905918122,
      "num_tokens": 40199537.0,
      "step": 19850
    },
    {
      "entropy": 5.922520303726197,
      "epoch": 1.7351219085904046,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004700092619654456,
      "loss": 5.318,
      "mean_token_accuracy": 0.19805318862199783,
      "num_tokens": 40209167.0,
      "step": 19855
    },
    {
      "entropy": 5.955082798004151,
      "epoch": 1.7355588569431093,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004699937164368625,
      "loss": 5.3222,
      "mean_token_accuracy": 0.19404130131006242,
      "num_tokens": 40218738.0,
      "step": 19860
    },
    {
      "entropy": 5.904144668579102,
      "epoch": 1.735995805295814,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004699781671681977,
      "loss": 5.308,
      "mean_token_accuracy": 0.20020924657583236,
      "num_tokens": 40227897.0,
      "step": 19865
    },
    {
      "entropy": 5.904859399795532,
      "epoch": 1.736432753648519,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046996261415974946,
      "loss": 5.3514,
      "mean_token_accuracy": 0.1928849220275879,
      "num_tokens": 40238870.0,
      "step": 19870
    },
    {
      "entropy": 5.89803991317749,
      "epoch": 1.7368697020012234,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046994705741181615,
      "loss": 5.2866,
      "mean_token_accuracy": 0.19475654065608977,
      "num_tokens": 40247832.0,
      "step": 19875
    },
    {
      "entropy": 5.9164430618286135,
      "epoch": 1.7373066503539283,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004699314969246961,
      "loss": 5.3132,
      "mean_token_accuracy": 0.19945163428783416,
      "num_tokens": 40258696.0,
      "step": 19880
    },
    {
      "entropy": 5.894814538955688,
      "epoch": 1.7377435987066328,
      "grad_norm": 1.125,
      "learning_rate": 0.00046991593269868776,
      "loss": 5.3027,
      "mean_token_accuracy": 0.2038818523287773,
      "num_tokens": 40267767.0,
      "step": 19885
    },
    {
      "entropy": 5.940172004699707,
      "epoch": 1.7381805470593377,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004699003647340898,
      "loss": 5.3679,
      "mean_token_accuracy": 0.19580054879188538,
      "num_tokens": 40277813.0,
      "step": 19890
    },
    {
      "entropy": 5.912716913223266,
      "epoch": 1.7386174954120421,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046988479303120066,
      "loss": 5.3252,
      "mean_token_accuracy": 0.19926552027463912,
      "num_tokens": 40289031.0,
      "step": 19895
    },
    {
      "entropy": 5.869098854064942,
      "epoch": 1.739054443764747,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046986921759031906,
      "loss": 5.3393,
      "mean_token_accuracy": 0.19585131108760834,
      "num_tokens": 40299428.0,
      "step": 19900
    },
    {
      "entropy": 5.900316429138184,
      "epoch": 1.7394913921174517,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004698536384117438,
      "loss": 5.2923,
      "mean_token_accuracy": 0.19841198474168778,
      "num_tokens": 40308996.0,
      "step": 19905
    },
    {
      "entropy": 5.9653702735900875,
      "epoch": 1.7399283404701564,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046983805549577365,
      "loss": 5.3751,
      "mean_token_accuracy": 0.1922510027885437,
      "num_tokens": 40319753.0,
      "step": 19910
    },
    {
      "entropy": 6.030674409866333,
      "epoch": 1.7403652888228611,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004698224688427074,
      "loss": 5.4783,
      "mean_token_accuracy": 0.18825637251138688,
      "num_tokens": 40330729.0,
      "step": 19915
    },
    {
      "entropy": 6.016564607620239,
      "epoch": 1.7408022371755658,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004698068784528442,
      "loss": 5.3948,
      "mean_token_accuracy": 0.19318338483572006,
      "num_tokens": 40340246.0,
      "step": 19920
    },
    {
      "entropy": 5.956729936599731,
      "epoch": 1.7412391855282705,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000469791284326483,
      "loss": 5.2682,
      "mean_token_accuracy": 0.19927387982606887,
      "num_tokens": 40350099.0,
      "step": 19925
    },
    {
      "entropy": 5.90136137008667,
      "epoch": 1.7416761338809752,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00046977568646392284,
      "loss": 5.3256,
      "mean_token_accuracy": 0.2064494401216507,
      "num_tokens": 40360825.0,
      "step": 19930
    },
    {
      "entropy": 5.964863681793213,
      "epoch": 1.74211308223368,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000469760084865463,
      "loss": 5.4278,
      "mean_token_accuracy": 0.18090111762285233,
      "num_tokens": 40371025.0,
      "step": 19935
    },
    {
      "entropy": 5.966486978530884,
      "epoch": 1.7425500305863846,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004697444795314027,
      "loss": 5.3099,
      "mean_token_accuracy": 0.20541592240333556,
      "num_tokens": 40380338.0,
      "step": 19940
    },
    {
      "entropy": 5.909269666671753,
      "epoch": 1.7429869789390895,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046972887046204114,
      "loss": 5.3469,
      "mean_token_accuracy": 0.19843609184026717,
      "num_tokens": 40391814.0,
      "step": 19945
    },
    {
      "entropy": 5.863748407363891,
      "epoch": 1.743423927291794,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046971325765767793,
      "loss": 5.3474,
      "mean_token_accuracy": 0.19197427183389665,
      "num_tokens": 40402013.0,
      "step": 19950
    },
    {
      "entropy": 5.946173477172851,
      "epoch": 1.7438608756444989,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046969764111861234,
      "loss": 5.321,
      "mean_token_accuracy": 0.1886312335729599,
      "num_tokens": 40412205.0,
      "step": 19955
    },
    {
      "entropy": 5.821734189987183,
      "epoch": 1.7442978239972036,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046968202084514405,
      "loss": 5.2772,
      "mean_token_accuracy": 0.20348888337612153,
      "num_tokens": 40423212.0,
      "step": 19960
    },
    {
      "entropy": 5.9296385765075685,
      "epoch": 1.7447347723499083,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004696663968375725,
      "loss": 5.2818,
      "mean_token_accuracy": 0.19966590106487275,
      "num_tokens": 40433272.0,
      "step": 19965
    },
    {
      "entropy": 5.927946710586548,
      "epoch": 1.745171720702613,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004696507690961975,
      "loss": 5.3983,
      "mean_token_accuracy": 0.19071025401353836,
      "num_tokens": 40443931.0,
      "step": 19970
    },
    {
      "entropy": 5.890453720092774,
      "epoch": 1.7456086690553176,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004696351376213188,
      "loss": 5.3491,
      "mean_token_accuracy": 0.1925002172589302,
      "num_tokens": 40453008.0,
      "step": 19975
    },
    {
      "entropy": 5.907824325561523,
      "epoch": 1.7460456174080223,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000469619502413236,
      "loss": 5.3896,
      "mean_token_accuracy": 0.19855455607175826,
      "num_tokens": 40463081.0,
      "step": 19980
    },
    {
      "entropy": 5.939786052703857,
      "epoch": 1.746482565760727,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004696038634722493,
      "loss": 5.2499,
      "mean_token_accuracy": 0.19647064805030823,
      "num_tokens": 40472390.0,
      "step": 19985
    },
    {
      "entropy": 5.9383259296417235,
      "epoch": 1.746919514113432,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046958822079865855,
      "loss": 5.3766,
      "mean_token_accuracy": 0.19558853507041932,
      "num_tokens": 40483068.0,
      "step": 19990
    },
    {
      "entropy": 5.949021673202514,
      "epoch": 1.7473564624661364,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004695725743927637,
      "loss": 5.3553,
      "mean_token_accuracy": 0.20267508029937745,
      "num_tokens": 40493125.0,
      "step": 19995
    },
    {
      "entropy": 5.9127381324768065,
      "epoch": 1.7477934108188413,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046955692425486496,
      "loss": 5.3641,
      "mean_token_accuracy": 0.19155369699001312,
      "num_tokens": 40502618.0,
      "step": 20000
    },
    {
      "entropy": 5.924920654296875,
      "epoch": 1.7482303591715458,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004695412703852624,
      "loss": 5.3737,
      "mean_token_accuracy": 0.19271163642406464,
      "num_tokens": 40512878.0,
      "step": 20005
    },
    {
      "entropy": 5.912238645553589,
      "epoch": 1.7486673075242507,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004695256127842563,
      "loss": 5.3322,
      "mean_token_accuracy": 0.19886060208082199,
      "num_tokens": 40521983.0,
      "step": 20010
    },
    {
      "entropy": 5.9119062423706055,
      "epoch": 1.7491042558769554,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046950995145214716,
      "loss": 5.3251,
      "mean_token_accuracy": 0.20158322006464005,
      "num_tokens": 40533126.0,
      "step": 20015
    },
    {
      "entropy": 5.875011682510376,
      "epoch": 1.74954120422966,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004694942863892352,
      "loss": 5.3185,
      "mean_token_accuracy": 0.19736259430646896,
      "num_tokens": 40543321.0,
      "step": 20020
    },
    {
      "entropy": 5.895274496078491,
      "epoch": 1.7499781525823648,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004694786175958209,
      "loss": 5.3359,
      "mean_token_accuracy": 0.19467836767435073,
      "num_tokens": 40554208.0,
      "step": 20025
    },
    {
      "entropy": 5.944083642959595,
      "epoch": 1.7504151009350695,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046946294507220477,
      "loss": 5.2994,
      "mean_token_accuracy": 0.19329520910978318,
      "num_tokens": 40563365.0,
      "step": 20030
    },
    {
      "entropy": 5.810305881500244,
      "epoch": 1.7508520492877742,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004694472688186874,
      "loss": 5.2497,
      "mean_token_accuracy": 0.2014814093708992,
      "num_tokens": 40573765.0,
      "step": 20035
    },
    {
      "entropy": 5.863768243789673,
      "epoch": 1.7512889976404789,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004694315888355697,
      "loss": 5.3437,
      "mean_token_accuracy": 0.20015811920166016,
      "num_tokens": 40583786.0,
      "step": 20040
    },
    {
      "entropy": 5.97484073638916,
      "epoch": 1.7517259459931838,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046941590512315215,
      "loss": 5.3556,
      "mean_token_accuracy": 0.1915882259607315,
      "num_tokens": 40593686.0,
      "step": 20045
    },
    {
      "entropy": 5.914437913894654,
      "epoch": 1.7521628943458882,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004694002176817356,
      "loss": 5.3565,
      "mean_token_accuracy": 0.1979151725769043,
      "num_tokens": 40604326.0,
      "step": 20050
    },
    {
      "entropy": 5.961737203598022,
      "epoch": 1.7525998426985931,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004693845265116211,
      "loss": 5.3302,
      "mean_token_accuracy": 0.19153402000665665,
      "num_tokens": 40613683.0,
      "step": 20055
    },
    {
      "entropy": 5.919159889221191,
      "epoch": 1.7530367910512976,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046936883161310944,
      "loss": 5.3593,
      "mean_token_accuracy": 0.19730237424373626,
      "num_tokens": 40623675.0,
      "step": 20060
    },
    {
      "entropy": 5.92514820098877,
      "epoch": 1.7534737394040025,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046935313298650185,
      "loss": 5.3776,
      "mean_token_accuracy": 0.19302769601345063,
      "num_tokens": 40634006.0,
      "step": 20065
    },
    {
      "entropy": 5.873923683166504,
      "epoch": 1.753910687756707,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046933743063209927,
      "loss": 5.2744,
      "mean_token_accuracy": 0.20196341574192048,
      "num_tokens": 40643175.0,
      "step": 20070
    },
    {
      "entropy": 5.9749226570129395,
      "epoch": 1.754347636109412,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046932172455020296,
      "loss": 5.3593,
      "mean_token_accuracy": 0.19196659624576567,
      "num_tokens": 40653447.0,
      "step": 20075
    },
    {
      "entropy": 5.88383994102478,
      "epoch": 1.7547845844621166,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004693060147411141,
      "loss": 5.2891,
      "mean_token_accuracy": 0.19802545011043549,
      "num_tokens": 40663499.0,
      "step": 20080
    },
    {
      "entropy": 5.892583417892456,
      "epoch": 1.7552215328148213,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004692903012051341,
      "loss": 5.3278,
      "mean_token_accuracy": 0.19430385977029802,
      "num_tokens": 40672943.0,
      "step": 20085
    },
    {
      "entropy": 5.93006157875061,
      "epoch": 1.755658481167526,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004692745839425643,
      "loss": 5.3283,
      "mean_token_accuracy": 0.20240439027547835,
      "num_tokens": 40683659.0,
      "step": 20090
    },
    {
      "entropy": 5.989038467407227,
      "epoch": 1.7560954295202307,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004692588629537062,
      "loss": 5.4047,
      "mean_token_accuracy": 0.18983942717313768,
      "num_tokens": 40693829.0,
      "step": 20095
    },
    {
      "entropy": 5.937786722183228,
      "epoch": 1.7565323778729354,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004692431382388613,
      "loss": 5.3659,
      "mean_token_accuracy": 0.19245025068521499,
      "num_tokens": 40704717.0,
      "step": 20100
    },
    {
      "entropy": 5.9455742835998535,
      "epoch": 1.75696932622564,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046922740979833135,
      "loss": 5.3208,
      "mean_token_accuracy": 0.18833200335502626,
      "num_tokens": 40714970.0,
      "step": 20105
    },
    {
      "entropy": 5.951665019989013,
      "epoch": 1.757406274578345,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046921167763241777,
      "loss": 5.3087,
      "mean_token_accuracy": 0.19667809009552,
      "num_tokens": 40723664.0,
      "step": 20110
    },
    {
      "entropy": 5.903662157058716,
      "epoch": 1.7578432229310494,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046919594174142247,
      "loss": 5.31,
      "mean_token_accuracy": 0.19484255313873292,
      "num_tokens": 40733735.0,
      "step": 20115
    },
    {
      "entropy": 5.929986190795899,
      "epoch": 1.7582801712837544,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046918020212564725,
      "loss": 5.3591,
      "mean_token_accuracy": 0.19656122326850892,
      "num_tokens": 40743699.0,
      "step": 20120
    },
    {
      "entropy": 5.922766876220703,
      "epoch": 1.7587171196364588,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000469164458785394,
      "loss": 5.4026,
      "mean_token_accuracy": 0.19362856894731523,
      "num_tokens": 40754553.0,
      "step": 20125
    },
    {
      "entropy": 5.962058353424072,
      "epoch": 1.7591540679891637,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046914871172096473,
      "loss": 5.3751,
      "mean_token_accuracy": 0.18748732954263686,
      "num_tokens": 40764660.0,
      "step": 20130
    },
    {
      "entropy": 5.913122129440308,
      "epoch": 1.7595910163418684,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004691329609326614,
      "loss": 5.2266,
      "mean_token_accuracy": 0.20604646652936937,
      "num_tokens": 40774237.0,
      "step": 20135
    },
    {
      "entropy": 5.888273429870606,
      "epoch": 1.7600279646945731,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004691172064207862,
      "loss": 5.3216,
      "mean_token_accuracy": 0.19565796852111816,
      "num_tokens": 40783608.0,
      "step": 20140
    },
    {
      "entropy": 5.948986005783081,
      "epoch": 1.7604649130472778,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004691014481856412,
      "loss": 5.3638,
      "mean_token_accuracy": 0.18923437744379043,
      "num_tokens": 40793987.0,
      "step": 20145
    },
    {
      "entropy": 5.959045696258545,
      "epoch": 1.7609018613999825,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046908568622752874,
      "loss": 5.3774,
      "mean_token_accuracy": 0.19223299771547317,
      "num_tokens": 40805600.0,
      "step": 20150
    },
    {
      "entropy": 5.9145214557647705,
      "epoch": 1.7613388097526872,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046906992054675115,
      "loss": 5.3524,
      "mean_token_accuracy": 0.19671232253313065,
      "num_tokens": 40815252.0,
      "step": 20155
    },
    {
      "entropy": 5.905719995498657,
      "epoch": 1.7617757581053919,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004690541511436107,
      "loss": 5.3271,
      "mean_token_accuracy": 0.20216090232133865,
      "num_tokens": 40825945.0,
      "step": 20160
    },
    {
      "entropy": 5.976608753204346,
      "epoch": 1.7622127064580968,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004690383780184101,
      "loss": 5.3968,
      "mean_token_accuracy": 0.1895548403263092,
      "num_tokens": 40835811.0,
      "step": 20165
    },
    {
      "entropy": 5.9654594421386715,
      "epoch": 1.7626496548108013,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046902260117145163,
      "loss": 5.378,
      "mean_token_accuracy": 0.18885264843702315,
      "num_tokens": 40845088.0,
      "step": 20170
    },
    {
      "entropy": 5.910682487487793,
      "epoch": 1.7630866031635062,
      "grad_norm": 1.09375,
      "learning_rate": 0.000469006820603038,
      "loss": 5.4027,
      "mean_token_accuracy": 0.193803508579731,
      "num_tokens": 40854703.0,
      "step": 20175
    },
    {
      "entropy": 5.960954236984253,
      "epoch": 1.7635235515162107,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046899103631347196,
      "loss": 5.33,
      "mean_token_accuracy": 0.1951819598674774,
      "num_tokens": 40865056.0,
      "step": 20180
    },
    {
      "entropy": 5.884387969970703,
      "epoch": 1.7639604998689156,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004689752483030561,
      "loss": 5.2439,
      "mean_token_accuracy": 0.20297405868768692,
      "num_tokens": 40874355.0,
      "step": 20185
    },
    {
      "entropy": 5.925824403762817,
      "epoch": 1.7643974482216203,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004689594565720934,
      "loss": 5.3184,
      "mean_token_accuracy": 0.19874671548604966,
      "num_tokens": 40884329.0,
      "step": 20190
    },
    {
      "entropy": 5.921063232421875,
      "epoch": 1.764834396574325,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004689436611208867,
      "loss": 5.3343,
      "mean_token_accuracy": 0.19625988006591796,
      "num_tokens": 40894702.0,
      "step": 20195
    },
    {
      "entropy": 5.914037799835205,
      "epoch": 1.7652713449270296,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046892786194973897,
      "loss": 5.2962,
      "mean_token_accuracy": 0.1912674367427826,
      "num_tokens": 40905467.0,
      "step": 20200
    },
    {
      "entropy": 5.96941032409668,
      "epoch": 1.7657082932797343,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004689120590589532,
      "loss": 5.3706,
      "mean_token_accuracy": 0.19233258664608002,
      "num_tokens": 40915915.0,
      "step": 20205
    },
    {
      "entropy": 5.899391841888428,
      "epoch": 1.766145241632439,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004688962524488325,
      "loss": 5.297,
      "mean_token_accuracy": 0.19911694824695586,
      "num_tokens": 40925941.0,
      "step": 20210
    },
    {
      "entropy": 5.85947618484497,
      "epoch": 1.7665821899851437,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004688804421196801,
      "loss": 5.3533,
      "mean_token_accuracy": 0.20057331025600433,
      "num_tokens": 40935831.0,
      "step": 20215
    },
    {
      "entropy": 5.904662322998047,
      "epoch": 1.7670191383378486,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046886462807179926,
      "loss": 5.3401,
      "mean_token_accuracy": 0.20154190212488174,
      "num_tokens": 40945341.0,
      "step": 20220
    },
    {
      "entropy": 5.949592876434326,
      "epoch": 1.767456086690553,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004688488103054931,
      "loss": 5.3667,
      "mean_token_accuracy": 0.19643764644861222,
      "num_tokens": 40955759.0,
      "step": 20225
    },
    {
      "entropy": 5.913175296783447,
      "epoch": 1.767893035043258,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004688329888210654,
      "loss": 5.3002,
      "mean_token_accuracy": 0.19549245983362198,
      "num_tokens": 40966184.0,
      "step": 20230
    },
    {
      "entropy": 5.923462915420532,
      "epoch": 1.7683299833959625,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004688171636188194,
      "loss": 5.347,
      "mean_token_accuracy": 0.19373768270015718,
      "num_tokens": 40975797.0,
      "step": 20235
    },
    {
      "entropy": 5.953031730651856,
      "epoch": 1.7687669317486674,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004688013346990585,
      "loss": 5.3324,
      "mean_token_accuracy": 0.19574716836214065,
      "num_tokens": 40985830.0,
      "step": 20240
    },
    {
      "entropy": 5.842448806762695,
      "epoch": 1.7692038801013719,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046878550206208656,
      "loss": 5.252,
      "mean_token_accuracy": 0.2051468789577484,
      "num_tokens": 40995692.0,
      "step": 20245
    },
    {
      "entropy": 5.925488805770874,
      "epoch": 1.7696408284540768,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004687696657082071,
      "loss": 5.3918,
      "mean_token_accuracy": 0.19208904802799226,
      "num_tokens": 41005506.0,
      "step": 20250
    },
    {
      "entropy": 5.931990432739258,
      "epoch": 1.7700777768067815,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046875382563772396,
      "loss": 5.26,
      "mean_token_accuracy": 0.19487065225839614,
      "num_tokens": 41014944.0,
      "step": 20255
    },
    {
      "entropy": 5.950332450866699,
      "epoch": 1.7705147251594862,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046873798185094093,
      "loss": 5.4072,
      "mean_token_accuracy": 0.19088623225688933,
      "num_tokens": 41025424.0,
      "step": 20260
    },
    {
      "entropy": 5.838885354995727,
      "epoch": 1.7709516735121908,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004687221343481618,
      "loss": 5.252,
      "mean_token_accuracy": 0.2061323493719101,
      "num_tokens": 41035817.0,
      "step": 20265
    },
    {
      "entropy": 5.859311532974243,
      "epoch": 1.7713886218648955,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046870628312969064,
      "loss": 5.3032,
      "mean_token_accuracy": 0.19870791137218474,
      "num_tokens": 41045730.0,
      "step": 20270
    },
    {
      "entropy": 5.972204637527466,
      "epoch": 1.7718255702176002,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004686904281958315,
      "loss": 5.3935,
      "mean_token_accuracy": 0.1839871421456337,
      "num_tokens": 41057921.0,
      "step": 20275
    },
    {
      "entropy": 5.995052766799927,
      "epoch": 1.772262518570305,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046867456954688847,
      "loss": 5.4034,
      "mean_token_accuracy": 0.19391241818666458,
      "num_tokens": 41068879.0,
      "step": 20280
    },
    {
      "entropy": 5.955120182037353,
      "epoch": 1.7726994669230098,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004686587071831657,
      "loss": 5.3999,
      "mean_token_accuracy": 0.19475507736206055,
      "num_tokens": 41079443.0,
      "step": 20285
    },
    {
      "entropy": 5.929945611953736,
      "epoch": 1.7731364152757143,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004686428411049675,
      "loss": 5.3214,
      "mean_token_accuracy": 0.19846466034650803,
      "num_tokens": 41089932.0,
      "step": 20290
    },
    {
      "entropy": 5.984380435943604,
      "epoch": 1.7735733636284192,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004686269713125981,
      "loss": 5.3668,
      "mean_token_accuracy": 0.19453175067901612,
      "num_tokens": 41100412.0,
      "step": 20295
    },
    {
      "entropy": 5.947173929214477,
      "epoch": 1.7740103119811237,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004686110978063619,
      "loss": 5.3439,
      "mean_token_accuracy": 0.19351837188005447,
      "num_tokens": 41110830.0,
      "step": 20300
    },
    {
      "entropy": 5.941296911239624,
      "epoch": 1.7744472603338286,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004685952205865634,
      "loss": 5.3405,
      "mean_token_accuracy": 0.19279694706201553,
      "num_tokens": 41120662.0,
      "step": 20305
    },
    {
      "entropy": 5.861632537841797,
      "epoch": 1.7748842086865333,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046857933965350714,
      "loss": 5.294,
      "mean_token_accuracy": 0.2060410961508751,
      "num_tokens": 41130523.0,
      "step": 20310
    },
    {
      "entropy": 5.934337377548218,
      "epoch": 1.775321157039238,
      "grad_norm": 1.125,
      "learning_rate": 0.0004685634550074977,
      "loss": 5.399,
      "mean_token_accuracy": 0.19071809351444244,
      "num_tokens": 41141659.0,
      "step": 20315
    },
    {
      "entropy": 5.9836619853973385,
      "epoch": 1.7757581053919427,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004685475666488398,
      "loss": 5.4285,
      "mean_token_accuracy": 0.18248261511325836,
      "num_tokens": 41151038.0,
      "step": 20320
    },
    {
      "entropy": 5.913211059570313,
      "epoch": 1.7761950537446474,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046853167457783815,
      "loss": 5.39,
      "mean_token_accuracy": 0.19372752606868743,
      "num_tokens": 41161888.0,
      "step": 20325
    },
    {
      "entropy": 6.029018068313599,
      "epoch": 1.776632002097352,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004685157787947975,
      "loss": 5.4419,
      "mean_token_accuracy": 0.19002725183963776,
      "num_tokens": 41172722.0,
      "step": 20330
    },
    {
      "entropy": 5.931641674041748,
      "epoch": 1.7770689504500568,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004684998793000229,
      "loss": 5.2814,
      "mean_token_accuracy": 0.1959647059440613,
      "num_tokens": 41181716.0,
      "step": 20335
    },
    {
      "entropy": 5.896604013442993,
      "epoch": 1.7775058988027617,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004684839760938193,
      "loss": 5.3793,
      "mean_token_accuracy": 0.18823183178901673,
      "num_tokens": 41192903.0,
      "step": 20340
    },
    {
      "entropy": 5.911653232574463,
      "epoch": 1.7779428471554661,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046846806917649145,
      "loss": 5.3384,
      "mean_token_accuracy": 0.19115843921899794,
      "num_tokens": 41203283.0,
      "step": 20345
    },
    {
      "entropy": 5.978838157653809,
      "epoch": 1.778379795508171,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004684521585483447,
      "loss": 5.3343,
      "mean_token_accuracy": 0.19642902612686158,
      "num_tokens": 41213615.0,
      "step": 20350
    },
    {
      "entropy": 5.933233737945557,
      "epoch": 1.7788167438608755,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046843624420968435,
      "loss": 5.2809,
      "mean_token_accuracy": 0.20643192529678345,
      "num_tokens": 41222705.0,
      "step": 20355
    },
    {
      "entropy": 5.862203121185303,
      "epoch": 1.7792536922135804,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004684203261608153,
      "loss": 5.2974,
      "mean_token_accuracy": 0.20058666914701462,
      "num_tokens": 41233201.0,
      "step": 20360
    },
    {
      "entropy": 5.963309288024902,
      "epoch": 1.7796906405662851,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004684044044020431,
      "loss": 5.3709,
      "mean_token_accuracy": 0.18946562856435775,
      "num_tokens": 41243088.0,
      "step": 20365
    },
    {
      "entropy": 6.011613512039185,
      "epoch": 1.7801275889189898,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004683884789336731,
      "loss": 5.376,
      "mean_token_accuracy": 0.1887464016675949,
      "num_tokens": 41253009.0,
      "step": 20370
    },
    {
      "entropy": 5.914664840698242,
      "epoch": 1.7805645372716945,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046837254975601073,
      "loss": 5.3844,
      "mean_token_accuracy": 0.19706502109766005,
      "num_tokens": 41262209.0,
      "step": 20375
    },
    {
      "entropy": 5.906603050231934,
      "epoch": 1.7810014856243992,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046835661686936154,
      "loss": 5.337,
      "mean_token_accuracy": 0.19838622361421585,
      "num_tokens": 41272762.0,
      "step": 20380
    },
    {
      "entropy": 5.927085208892822,
      "epoch": 1.7814384339771039,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046834068027403115,
      "loss": 5.416,
      "mean_token_accuracy": 0.19142892211675644,
      "num_tokens": 41284306.0,
      "step": 20385
    },
    {
      "entropy": 5.929792833328247,
      "epoch": 1.7818753823298086,
      "grad_norm": 1.0,
      "learning_rate": 0.00046832473997032514,
      "loss": 5.3712,
      "mean_token_accuracy": 0.19115519970655442,
      "num_tokens": 41295211.0,
      "step": 20390
    },
    {
      "entropy": 5.918662261962891,
      "epoch": 1.7823123306825135,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004683087959585493,
      "loss": 5.2164,
      "mean_token_accuracy": 0.20559938848018647,
      "num_tokens": 41303773.0,
      "step": 20395
    },
    {
      "entropy": 5.878101491928101,
      "epoch": 1.782749279035218,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004682928482390095,
      "loss": 5.3217,
      "mean_token_accuracy": 0.19544345438480376,
      "num_tokens": 41313536.0,
      "step": 20400
    },
    {
      "entropy": 5.959080839157105,
      "epoch": 1.7831862273879229,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046827689681201164,
      "loss": 5.3325,
      "mean_token_accuracy": 0.1937728226184845,
      "num_tokens": 41323831.0,
      "step": 20405
    },
    {
      "entropy": 5.954738473892212,
      "epoch": 1.7836231757406273,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004682609416778615,
      "loss": 5.3532,
      "mean_token_accuracy": 0.1946051150560379,
      "num_tokens": 41334295.0,
      "step": 20410
    },
    {
      "entropy": 5.936319971084595,
      "epoch": 1.7840601240933323,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004682449828368653,
      "loss": 5.3524,
      "mean_token_accuracy": 0.1898101359605789,
      "num_tokens": 41344672.0,
      "step": 20415
    },
    {
      "entropy": 5.966075897216797,
      "epoch": 1.7844970724460367,
      "grad_norm": 1.125,
      "learning_rate": 0.0004682290202893291,
      "loss": 5.3687,
      "mean_token_accuracy": 0.18830661624670028,
      "num_tokens": 41355108.0,
      "step": 20420
    },
    {
      "entropy": 6.024651384353637,
      "epoch": 1.7849340207987416,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046821305403555905,
      "loss": 5.4625,
      "mean_token_accuracy": 0.1862081468105316,
      "num_tokens": 41366277.0,
      "step": 20425
    },
    {
      "entropy": 5.97805814743042,
      "epoch": 1.7853709691514463,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046819708407586135,
      "loss": 5.3331,
      "mean_token_accuracy": 0.1959231123328209,
      "num_tokens": 41374757.0,
      "step": 20430
    },
    {
      "entropy": 5.894075012207031,
      "epoch": 1.785807917504151,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004681811104105423,
      "loss": 5.3123,
      "mean_token_accuracy": 0.20218145996332168,
      "num_tokens": 41384883.0,
      "step": 20435
    },
    {
      "entropy": 5.906810283660889,
      "epoch": 1.7862448658568557,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004681651330399084,
      "loss": 5.3461,
      "mean_token_accuracy": 0.19746147394180297,
      "num_tokens": 41396492.0,
      "step": 20440
    },
    {
      "entropy": 5.90576205253601,
      "epoch": 1.7866818142095604,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000468149151964266,
      "loss": 5.317,
      "mean_token_accuracy": 0.19576033651828767,
      "num_tokens": 41406217.0,
      "step": 20445
    },
    {
      "entropy": 5.913767910003662,
      "epoch": 1.787118762562265,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004681331671839217,
      "loss": 5.2683,
      "mean_token_accuracy": 0.20321421921253205,
      "num_tokens": 41415895.0,
      "step": 20450
    },
    {
      "entropy": 5.960805463790893,
      "epoch": 1.7875557109149698,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046811717869918196,
      "loss": 5.4092,
      "mean_token_accuracy": 0.18762925267219543,
      "num_tokens": 41426021.0,
      "step": 20455
    },
    {
      "entropy": 5.978705024719238,
      "epoch": 1.7879926592676747,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004681011865103536,
      "loss": 5.3105,
      "mean_token_accuracy": 0.19555696547031404,
      "num_tokens": 41434656.0,
      "step": 20460
    },
    {
      "entropy": 5.933466625213623,
      "epoch": 1.7884296076203792,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046808519061774324,
      "loss": 5.3515,
      "mean_token_accuracy": 0.1933719038963318,
      "num_tokens": 41443940.0,
      "step": 20465
    },
    {
      "entropy": 5.927581071853638,
      "epoch": 1.788866555973084,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046806919102165785,
      "loss": 5.4014,
      "mean_token_accuracy": 0.19415586739778518,
      "num_tokens": 41454907.0,
      "step": 20470
    },
    {
      "entropy": 5.931322717666626,
      "epoch": 1.7893035043257886,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004680531877224042,
      "loss": 5.2632,
      "mean_token_accuracy": 0.20196777433156968,
      "num_tokens": 41464248.0,
      "step": 20475
    },
    {
      "entropy": 5.901649713516235,
      "epoch": 1.7897404526784935,
      "grad_norm": 1.125,
      "learning_rate": 0.0004680371807202892,
      "loss": 5.3055,
      "mean_token_accuracy": 0.20639698207378387,
      "num_tokens": 41475192.0,
      "step": 20480
    },
    {
      "entropy": 5.946984004974365,
      "epoch": 1.7901774010311982,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046802117001561996,
      "loss": 5.4042,
      "mean_token_accuracy": 0.18257917761802672,
      "num_tokens": 41484753.0,
      "step": 20485
    },
    {
      "entropy": 5.971106815338135,
      "epoch": 1.7906143493839028,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046800515560870346,
      "loss": 5.4044,
      "mean_token_accuracy": 0.18612490296363832,
      "num_tokens": 41495885.0,
      "step": 20490
    },
    {
      "entropy": 6.000931358337402,
      "epoch": 1.7910512977366075,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046798913749984694,
      "loss": 5.3526,
      "mean_token_accuracy": 0.1991802468895912,
      "num_tokens": 41506330.0,
      "step": 20495
    },
    {
      "entropy": 5.923953056335449,
      "epoch": 1.7914882460893122,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004679731156893577,
      "loss": 5.3631,
      "mean_token_accuracy": 0.19112589359283447,
      "num_tokens": 41515850.0,
      "step": 20500
    },
    {
      "entropy": 5.901385021209717,
      "epoch": 1.791925194442017,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000467957090177543,
      "loss": 5.3397,
      "mean_token_accuracy": 0.19585326611995696,
      "num_tokens": 41525910.0,
      "step": 20505
    },
    {
      "entropy": 5.926617813110352,
      "epoch": 1.7923621427947216,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046794106096471015,
      "loss": 5.2659,
      "mean_token_accuracy": 0.19943616837263106,
      "num_tokens": 41536417.0,
      "step": 20510
    },
    {
      "entropy": 5.922287797927856,
      "epoch": 1.7927990911474265,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004679250280511667,
      "loss": 5.3401,
      "mean_token_accuracy": 0.19556146264076232,
      "num_tokens": 41547215.0,
      "step": 20515
    },
    {
      "entropy": 5.966777753829956,
      "epoch": 1.793236039500131,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004679089914372201,
      "loss": 5.3717,
      "mean_token_accuracy": 0.1923124983906746,
      "num_tokens": 41556036.0,
      "step": 20520
    },
    {
      "entropy": 5.930334901809692,
      "epoch": 1.793672987852836,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046789295112317796,
      "loss": 5.3733,
      "mean_token_accuracy": 0.1912417009472847,
      "num_tokens": 41565534.0,
      "step": 20525
    },
    {
      "entropy": 5.907385778427124,
      "epoch": 1.7941099362055404,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004678769071093479,
      "loss": 5.314,
      "mean_token_accuracy": 0.19555848240852355,
      "num_tokens": 41576150.0,
      "step": 20530
    },
    {
      "entropy": 5.891354274749756,
      "epoch": 1.7945468845582453,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046786085939603785,
      "loss": 5.3174,
      "mean_token_accuracy": 0.20140010565519334,
      "num_tokens": 41585248.0,
      "step": 20535
    },
    {
      "entropy": 5.913039112091065,
      "epoch": 1.7949838329109498,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004678448079835552,
      "loss": 5.3638,
      "mean_token_accuracy": 0.19171476513147354,
      "num_tokens": 41595215.0,
      "step": 20540
    },
    {
      "entropy": 5.953266048431397,
      "epoch": 1.7954207812636547,
      "grad_norm": 1.125,
      "learning_rate": 0.0004678287528722083,
      "loss": 5.4082,
      "mean_token_accuracy": 0.19307370930910112,
      "num_tokens": 41605268.0,
      "step": 20545
    },
    {
      "entropy": 5.896877670288086,
      "epoch": 1.7958577296163594,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004678126940623048,
      "loss": 5.2533,
      "mean_token_accuracy": 0.19982493817806243,
      "num_tokens": 41614554.0,
      "step": 20550
    },
    {
      "entropy": 5.899889421463013,
      "epoch": 1.796294677969064,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046779663155415287,
      "loss": 5.3702,
      "mean_token_accuracy": 0.19761285036802292,
      "num_tokens": 41624839.0,
      "step": 20555
    },
    {
      "entropy": 5.82927451133728,
      "epoch": 1.7967316263217687,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004677805653480604,
      "loss": 5.2336,
      "mean_token_accuracy": 0.21084865182638168,
      "num_tokens": 41634577.0,
      "step": 20560
    },
    {
      "entropy": 5.913674211502075,
      "epoch": 1.7971685746744734,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046776449544433575,
      "loss": 5.3776,
      "mean_token_accuracy": 0.187717804312706,
      "num_tokens": 41645213.0,
      "step": 20565
    },
    {
      "entropy": 5.894586896896362,
      "epoch": 1.7976055230271784,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000467748421843287,
      "loss": 5.2602,
      "mean_token_accuracy": 0.20337387919425964,
      "num_tokens": 41655160.0,
      "step": 20570
    },
    {
      "entropy": 5.925815010070801,
      "epoch": 1.7980424713798828,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046773234454522265,
      "loss": 5.3102,
      "mean_token_accuracy": 0.19598203897476196,
      "num_tokens": 41665861.0,
      "step": 20575
    },
    {
      "entropy": 5.935397481918335,
      "epoch": 1.7984794197325877,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046771626355045085,
      "loss": 5.3565,
      "mean_token_accuracy": 0.1955948770046234,
      "num_tokens": 41676699.0,
      "step": 20580
    },
    {
      "entropy": 5.941192531585694,
      "epoch": 1.7989163680852922,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004677001788592801,
      "loss": 5.3366,
      "mean_token_accuracy": 0.20105181783437728,
      "num_tokens": 41688162.0,
      "step": 20585
    },
    {
      "entropy": 5.963338613510132,
      "epoch": 1.7993533164379971,
      "grad_norm": 1.171875,
      "learning_rate": 0.000467684090472019,
      "loss": 5.3687,
      "mean_token_accuracy": 0.19627980440855025,
      "num_tokens": 41698088.0,
      "step": 20590
    },
    {
      "entropy": 5.927029991149903,
      "epoch": 1.7997902647907016,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004676679983889761,
      "loss": 5.3,
      "mean_token_accuracy": 0.20788575261831282,
      "num_tokens": 41707446.0,
      "step": 20595
    },
    {
      "entropy": 5.927051448822022,
      "epoch": 1.8002272131434065,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046765190261046,
      "loss": 5.3967,
      "mean_token_accuracy": 0.19062488973140718,
      "num_tokens": 41716388.0,
      "step": 20600
    },
    {
      "entropy": 5.951300477981567,
      "epoch": 1.8006641614961112,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004676358031367795,
      "loss": 5.3953,
      "mean_token_accuracy": 0.19230061620473862,
      "num_tokens": 41726328.0,
      "step": 20605
    },
    {
      "entropy": 5.958097314834594,
      "epoch": 1.8011011098488159,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046761969996824326,
      "loss": 5.3079,
      "mean_token_accuracy": 0.19323919862508773,
      "num_tokens": 41736128.0,
      "step": 20610
    },
    {
      "entropy": 5.88479962348938,
      "epoch": 1.8015380582015206,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004676035931051604,
      "loss": 5.2959,
      "mean_token_accuracy": 0.1964612439274788,
      "num_tokens": 41746316.0,
      "step": 20615
    },
    {
      "entropy": 5.885156679153442,
      "epoch": 1.8019750065542253,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046758748254783957,
      "loss": 5.2578,
      "mean_token_accuracy": 0.20799477845430375,
      "num_tokens": 41756015.0,
      "step": 20620
    },
    {
      "entropy": 5.9326146125793455,
      "epoch": 1.80241195490693,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046757136829658996,
      "loss": 5.2783,
      "mean_token_accuracy": 0.2035985141992569,
      "num_tokens": 41766360.0,
      "step": 20625
    },
    {
      "entropy": 5.974231338500976,
      "epoch": 1.8028489032596346,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046755525035172056,
      "loss": 5.4009,
      "mean_token_accuracy": 0.19356951415538787,
      "num_tokens": 41776640.0,
      "step": 20630
    },
    {
      "entropy": 5.9204905986785885,
      "epoch": 1.8032858516123396,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004675391287135406,
      "loss": 5.3881,
      "mean_token_accuracy": 0.19882839173078537,
      "num_tokens": 41787329.0,
      "step": 20635
    },
    {
      "entropy": 5.936491298675537,
      "epoch": 1.803722799965044,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046752300338235924,
      "loss": 5.377,
      "mean_token_accuracy": 0.18841838985681533,
      "num_tokens": 41797890.0,
      "step": 20640
    },
    {
      "entropy": 5.930166339874267,
      "epoch": 1.804159748317749,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004675068743584858,
      "loss": 5.3803,
      "mean_token_accuracy": 0.18791066706180573,
      "num_tokens": 41808845.0,
      "step": 20645
    },
    {
      "entropy": 5.928231334686279,
      "epoch": 1.8045966966704534,
      "grad_norm": 1.125,
      "learning_rate": 0.0004674907416422297,
      "loss": 5.3553,
      "mean_token_accuracy": 0.19804320931434632,
      "num_tokens": 41819499.0,
      "step": 20650
    },
    {
      "entropy": 5.959622573852539,
      "epoch": 1.8050336450231583,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004674746052339002,
      "loss": 5.4,
      "mean_token_accuracy": 0.18445470929145813,
      "num_tokens": 41827998.0,
      "step": 20655
    },
    {
      "entropy": 5.966298341751099,
      "epoch": 1.805470593375863,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046745846513380704,
      "loss": 5.3813,
      "mean_token_accuracy": 0.1934305340051651,
      "num_tokens": 41839610.0,
      "step": 20660
    },
    {
      "entropy": 5.9602500915527346,
      "epoch": 1.8059075417285677,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046744232134225954,
      "loss": 5.3051,
      "mean_token_accuracy": 0.19973748326301574,
      "num_tokens": 41850111.0,
      "step": 20665
    },
    {
      "entropy": 5.8999258518219,
      "epoch": 1.8063444900812724,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004674261738595676,
      "loss": 5.3417,
      "mean_token_accuracy": 0.1907190576195717,
      "num_tokens": 41859843.0,
      "step": 20670
    },
    {
      "entropy": 5.936347770690918,
      "epoch": 1.806781438433977,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046741002268604073,
      "loss": 5.3379,
      "mean_token_accuracy": 0.1963289588689804,
      "num_tokens": 41868864.0,
      "step": 20675
    },
    {
      "entropy": 5.936072254180909,
      "epoch": 1.8072183867866818,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046739386782198883,
      "loss": 5.3478,
      "mean_token_accuracy": 0.1959891900420189,
      "num_tokens": 41878274.0,
      "step": 20680
    },
    {
      "entropy": 5.904694557189941,
      "epoch": 1.8076553351393865,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046737770926772175,
      "loss": 5.3232,
      "mean_token_accuracy": 0.1914282873272896,
      "num_tokens": 41888005.0,
      "step": 20685
    },
    {
      "entropy": 5.990843296051025,
      "epoch": 1.8080922834920914,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004673615470235494,
      "loss": 5.444,
      "mean_token_accuracy": 0.18665173500776291,
      "num_tokens": 41897415.0,
      "step": 20690
    },
    {
      "entropy": 5.940118598937988,
      "epoch": 1.8085292318447959,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004673453810897817,
      "loss": 5.3151,
      "mean_token_accuracy": 0.19964141249656678,
      "num_tokens": 41907021.0,
      "step": 20695
    },
    {
      "entropy": 5.924611186981201,
      "epoch": 1.8089661801975008,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004673292114667289,
      "loss": 5.4037,
      "mean_token_accuracy": 0.18832848370075225,
      "num_tokens": 41918358.0,
      "step": 20700
    },
    {
      "entropy": 6.007497549057007,
      "epoch": 1.8094031285502052,
      "grad_norm": 1.125,
      "learning_rate": 0.00046731303815470104,
      "loss": 5.396,
      "mean_token_accuracy": 0.18466378897428512,
      "num_tokens": 41929311.0,
      "step": 20705
    },
    {
      "entropy": 6.00486798286438,
      "epoch": 1.8098400769029102,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046729686115400826,
      "loss": 5.356,
      "mean_token_accuracy": 0.19420384913682937,
      "num_tokens": 41939623.0,
      "step": 20710
    },
    {
      "entropy": 5.913223695755005,
      "epoch": 1.8102770252556146,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004672806804649611,
      "loss": 5.3357,
      "mean_token_accuracy": 0.1947373330593109,
      "num_tokens": 41949650.0,
      "step": 20715
    },
    {
      "entropy": 5.91931562423706,
      "epoch": 1.8107139736083195,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046726449608786957,
      "loss": 5.3095,
      "mean_token_accuracy": 0.20309367626905442,
      "num_tokens": 41960262.0,
      "step": 20720
    },
    {
      "entropy": 5.966704654693603,
      "epoch": 1.8111509219610242,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004672483080230443,
      "loss": 5.3539,
      "mean_token_accuracy": 0.18918833136558533,
      "num_tokens": 41969874.0,
      "step": 20725
    },
    {
      "entropy": 5.954252529144287,
      "epoch": 1.811587870313729,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046723211627079577,
      "loss": 5.3394,
      "mean_token_accuracy": 0.19364485144615173,
      "num_tokens": 41978810.0,
      "step": 20730
    },
    {
      "entropy": 5.895397186279297,
      "epoch": 1.8120248186664336,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004672159208314344,
      "loss": 5.2911,
      "mean_token_accuracy": 0.19692443460226058,
      "num_tokens": 41988785.0,
      "step": 20735
    },
    {
      "entropy": 5.890549659729004,
      "epoch": 1.8124617670191383,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004671997217052711,
      "loss": 5.3643,
      "mean_token_accuracy": 0.20018889009952545,
      "num_tokens": 41998698.0,
      "step": 20740
    },
    {
      "entropy": 5.947597312927246,
      "epoch": 1.812898715371843,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004671835188926164,
      "loss": 5.3108,
      "mean_token_accuracy": 0.20749209076166153,
      "num_tokens": 42009082.0,
      "step": 20745
    },
    {
      "entropy": 6.002422189712524,
      "epoch": 1.8133356637245477,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004671673123937811,
      "loss": 5.3381,
      "mean_token_accuracy": 0.19272368401288986,
      "num_tokens": 42019960.0,
      "step": 20750
    },
    {
      "entropy": 5.880385541915894,
      "epoch": 1.8137726120772526,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000467151102209076,
      "loss": 5.2669,
      "mean_token_accuracy": 0.19643403440713883,
      "num_tokens": 42028799.0,
      "step": 20755
    },
    {
      "entropy": 5.87654480934143,
      "epoch": 1.814209560429957,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046713488833881214,
      "loss": 5.3419,
      "mean_token_accuracy": 0.19009020328521728,
      "num_tokens": 42038967.0,
      "step": 20760
    },
    {
      "entropy": 5.918560266494751,
      "epoch": 1.814646508782662,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046711867078330047,
      "loss": 5.3009,
      "mean_token_accuracy": 0.19600185602903367,
      "num_tokens": 42049982.0,
      "step": 20765
    },
    {
      "entropy": 5.934220886230468,
      "epoch": 1.8150834571353665,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046710244954285184,
      "loss": 5.4153,
      "mean_token_accuracy": 0.19311527013778687,
      "num_tokens": 42059835.0,
      "step": 20770
    },
    {
      "entropy": 5.9067902088165285,
      "epoch": 1.8155204054880714,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046708622461777774,
      "loss": 5.2437,
      "mean_token_accuracy": 0.20367998778820037,
      "num_tokens": 42069339.0,
      "step": 20775
    },
    {
      "entropy": 5.89956226348877,
      "epoch": 1.815957353840776,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004670699960083892,
      "loss": 5.2592,
      "mean_token_accuracy": 0.1951482206583023,
      "num_tokens": 42080618.0,
      "step": 20780
    },
    {
      "entropy": 5.983381700515747,
      "epoch": 1.8163943021934807,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004670537637149974,
      "loss": 5.4186,
      "mean_token_accuracy": 0.18638374507427216,
      "num_tokens": 42091458.0,
      "step": 20785
    },
    {
      "entropy": 5.944495344161988,
      "epoch": 1.8168312505461854,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046703752773791387,
      "loss": 5.4015,
      "mean_token_accuracy": 0.19249362647533416,
      "num_tokens": 42102515.0,
      "step": 20790
    },
    {
      "entropy": 5.934634160995484,
      "epoch": 1.8172681988988901,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004670212880774499,
      "loss": 5.3816,
      "mean_token_accuracy": 0.18766273111104964,
      "num_tokens": 42113853.0,
      "step": 20795
    },
    {
      "entropy": 5.969643926620483,
      "epoch": 1.8177051472515948,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000467005044733917,
      "loss": 5.3887,
      "mean_token_accuracy": 0.1888352930545807,
      "num_tokens": 42123101.0,
      "step": 20800
    },
    {
      "entropy": 5.973903656005859,
      "epoch": 1.8181420956042995,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004669887977076267,
      "loss": 5.3558,
      "mean_token_accuracy": 0.1977720856666565,
      "num_tokens": 42133051.0,
      "step": 20805
    },
    {
      "entropy": 5.933545637130737,
      "epoch": 1.8185790439570044,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046697254699889064,
      "loss": 5.3348,
      "mean_token_accuracy": 0.19556159824132918,
      "num_tokens": 42142705.0,
      "step": 20810
    },
    {
      "entropy": 5.882030963897705,
      "epoch": 1.819015992309709,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046695629260802064,
      "loss": 5.3259,
      "mean_token_accuracy": 0.19572998881340026,
      "num_tokens": 42152399.0,
      "step": 20815
    },
    {
      "entropy": 5.919317007064819,
      "epoch": 1.8194529406624138,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004669400345353283,
      "loss": 5.351,
      "mean_token_accuracy": 0.19746672213077546,
      "num_tokens": 42162651.0,
      "step": 20820
    },
    {
      "entropy": 5.886063528060913,
      "epoch": 1.8198898890151183,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004669237727811255,
      "loss": 5.241,
      "mean_token_accuracy": 0.1996626079082489,
      "num_tokens": 42172595.0,
      "step": 20825
    },
    {
      "entropy": 5.914587068557739,
      "epoch": 1.8203268373678232,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004669075073457242,
      "loss": 5.3349,
      "mean_token_accuracy": 0.19702446162700654,
      "num_tokens": 42182631.0,
      "step": 20830
    },
    {
      "entropy": 5.925761747360229,
      "epoch": 1.8207637857205279,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046689123822943643,
      "loss": 5.3292,
      "mean_token_accuracy": 0.19222403019666673,
      "num_tokens": 42192743.0,
      "step": 20835
    },
    {
      "entropy": 5.935552310943604,
      "epoch": 1.8212007340732326,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004668749654325741,
      "loss": 5.3229,
      "mean_token_accuracy": 0.19752402901649474,
      "num_tokens": 42203090.0,
      "step": 20840
    },
    {
      "entropy": 5.951908302307129,
      "epoch": 1.8216376824259373,
      "grad_norm": 1.125,
      "learning_rate": 0.0004668586889554493,
      "loss": 5.405,
      "mean_token_accuracy": 0.19034480452537536,
      "num_tokens": 42213341.0,
      "step": 20845
    },
    {
      "entropy": 5.948003530502319,
      "epoch": 1.822074630778642,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004668424087983744,
      "loss": 5.4031,
      "mean_token_accuracy": 0.1946099281311035,
      "num_tokens": 42222862.0,
      "step": 20850
    },
    {
      "entropy": 5.930844354629516,
      "epoch": 1.8225115791313466,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046682612496166156,
      "loss": 5.3989,
      "mean_token_accuracy": 0.19022073745727539,
      "num_tokens": 42234045.0,
      "step": 20855
    },
    {
      "entropy": 5.990992212295533,
      "epoch": 1.8229485274840513,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046680983744562315,
      "loss": 5.3265,
      "mean_token_accuracy": 0.1933588221669197,
      "num_tokens": 42244093.0,
      "step": 20860
    },
    {
      "entropy": 5.911411952972412,
      "epoch": 1.8233854758367563,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004667935462505716,
      "loss": 5.3798,
      "mean_token_accuracy": 0.19526235461235047,
      "num_tokens": 42254865.0,
      "step": 20865
    },
    {
      "entropy": 5.9063252925872805,
      "epoch": 1.8238224241894607,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046677725137681924,
      "loss": 5.2926,
      "mean_token_accuracy": 0.2003451853990555,
      "num_tokens": 42264565.0,
      "step": 20870
    },
    {
      "entropy": 5.983190727233887,
      "epoch": 1.8242593725421656,
      "grad_norm": 1.125,
      "learning_rate": 0.00046676095282467873,
      "loss": 5.3804,
      "mean_token_accuracy": 0.1888802319765091,
      "num_tokens": 42274289.0,
      "step": 20875
    },
    {
      "entropy": 5.966687250137329,
      "epoch": 1.82469632089487,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004667446505944627,
      "loss": 5.3461,
      "mean_token_accuracy": 0.1932797208428383,
      "num_tokens": 42283855.0,
      "step": 20880
    },
    {
      "entropy": 5.883030128479004,
      "epoch": 1.825133269247575,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046672834468648375,
      "loss": 5.3601,
      "mean_token_accuracy": 0.20229913741350175,
      "num_tokens": 42294871.0,
      "step": 20885
    },
    {
      "entropy": 5.947096538543701,
      "epoch": 1.8255702176002795,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004667120351010547,
      "loss": 5.32,
      "mean_token_accuracy": 0.19752324372529984,
      "num_tokens": 42305315.0,
      "step": 20890
    },
    {
      "entropy": 5.915869617462159,
      "epoch": 1.8260071659529844,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046669572183848833,
      "loss": 5.2836,
      "mean_token_accuracy": 0.20396759659051894,
      "num_tokens": 42316893.0,
      "step": 20895
    },
    {
      "entropy": 5.88280234336853,
      "epoch": 1.826444114305689,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046667940489909766,
      "loss": 5.2794,
      "mean_token_accuracy": 0.19809929728507997,
      "num_tokens": 42326784.0,
      "step": 20900
    },
    {
      "entropy": 5.926232814788818,
      "epoch": 1.8268810626583938,
      "grad_norm": 1.125,
      "learning_rate": 0.00046666308428319547,
      "loss": 5.2763,
      "mean_token_accuracy": 0.2025792494416237,
      "num_tokens": 42336283.0,
      "step": 20905
    },
    {
      "entropy": 5.902861642837524,
      "epoch": 1.8273180110110985,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046664675999109494,
      "loss": 5.3102,
      "mean_token_accuracy": 0.20163041055202485,
      "num_tokens": 42346933.0,
      "step": 20910
    },
    {
      "entropy": 5.907564115524292,
      "epoch": 1.8277549593638032,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004666304320231091,
      "loss": 5.3887,
      "mean_token_accuracy": 0.1952345997095108,
      "num_tokens": 42357616.0,
      "step": 20915
    },
    {
      "entropy": 5.9767005443573,
      "epoch": 1.8281919077165079,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004666141003795512,
      "loss": 5.3418,
      "mean_token_accuracy": 0.19971691220998763,
      "num_tokens": 42367566.0,
      "step": 20920
    },
    {
      "entropy": 5.980360412597657,
      "epoch": 1.8286288560692125,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046659776506073446,
      "loss": 5.3204,
      "mean_token_accuracy": 0.20076457262039185,
      "num_tokens": 42377412.0,
      "step": 20925
    },
    {
      "entropy": 5.974689674377442,
      "epoch": 1.8290658044219175,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046658142606697216,
      "loss": 5.309,
      "mean_token_accuracy": 0.19862452298402786,
      "num_tokens": 42387952.0,
      "step": 20930
    },
    {
      "entropy": 5.932408046722412,
      "epoch": 1.829502752774622,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004665650833985777,
      "loss": 5.308,
      "mean_token_accuracy": 0.19058774709701537,
      "num_tokens": 42398488.0,
      "step": 20935
    },
    {
      "entropy": 5.911058235168457,
      "epoch": 1.8299397011273268,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046654873705586464,
      "loss": 5.3113,
      "mean_token_accuracy": 0.19930914491415025,
      "num_tokens": 42409137.0,
      "step": 20940
    },
    {
      "entropy": 6.007994031906128,
      "epoch": 1.8303766494800313,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004665323870391464,
      "loss": 5.4201,
      "mean_token_accuracy": 0.1868569403886795,
      "num_tokens": 42419867.0,
      "step": 20945
    },
    {
      "entropy": 5.954726505279541,
      "epoch": 1.8308135978327362,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004665160333487365,
      "loss": 5.3494,
      "mean_token_accuracy": 0.1952267363667488,
      "num_tokens": 42429590.0,
      "step": 20950
    },
    {
      "entropy": 5.9517840385437015,
      "epoch": 1.831250546185441,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004664996759849489,
      "loss": 5.3222,
      "mean_token_accuracy": 0.20029521435499192,
      "num_tokens": 42439777.0,
      "step": 20955
    },
    {
      "entropy": 5.917471504211425,
      "epoch": 1.8316874945381456,
      "grad_norm": 1.125,
      "learning_rate": 0.0004664833149480971,
      "loss": 5.3465,
      "mean_token_accuracy": 0.19896059781312941,
      "num_tokens": 42449843.0,
      "step": 20960
    },
    {
      "entropy": 5.852067232131958,
      "epoch": 1.8321244428908503,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046646695023849506,
      "loss": 5.2128,
      "mean_token_accuracy": 0.20487748384475707,
      "num_tokens": 42459056.0,
      "step": 20965
    },
    {
      "entropy": 5.877831315994262,
      "epoch": 1.832561391243555,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046645058185645644,
      "loss": 5.2535,
      "mean_token_accuracy": 0.19697408080101014,
      "num_tokens": 42468946.0,
      "step": 20970
    },
    {
      "entropy": 5.916455936431885,
      "epoch": 1.8329983395962597,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004664342098022954,
      "loss": 5.3303,
      "mean_token_accuracy": 0.19828195869922638,
      "num_tokens": 42478894.0,
      "step": 20975
    },
    {
      "entropy": 5.934530353546142,
      "epoch": 1.8334352879489644,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004664178340763259,
      "loss": 5.3286,
      "mean_token_accuracy": 0.20290619134902954,
      "num_tokens": 42490124.0,
      "step": 20980
    },
    {
      "entropy": 5.926880931854248,
      "epoch": 1.8338722363016693,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004664014546788621,
      "loss": 5.3648,
      "mean_token_accuracy": 0.19886808544397355,
      "num_tokens": 42499554.0,
      "step": 20985
    },
    {
      "entropy": 5.876231575012207,
      "epoch": 1.8343091846543738,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046638507161021807,
      "loss": 5.2129,
      "mean_token_accuracy": 0.20397316217422484,
      "num_tokens": 42510364.0,
      "step": 20990
    },
    {
      "entropy": 5.979144287109375,
      "epoch": 1.8347461330070787,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046636868487070804,
      "loss": 5.4072,
      "mean_token_accuracy": 0.18734091222286225,
      "num_tokens": 42521671.0,
      "step": 20995
    },
    {
      "entropy": 5.94301929473877,
      "epoch": 1.8351830813597831,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046635229446064647,
      "loss": 5.2944,
      "mean_token_accuracy": 0.19969464391469954,
      "num_tokens": 42532784.0,
      "step": 21000
    },
    {
      "epoch": 1.8351830813597831,
      "eval_entropy": 5.677052411915598,
      "eval_loss": 5.363137722015381,
      "eval_mean_token_accuracy": 0.2040186183382593,
      "eval_num_tokens": 42532784.0,
      "eval_runtime": 21.5568,
      "eval_samples_per_second": 1717.044,
      "eval_steps_per_second": 214.642,
      "step": 21000
    },
    {
      "entropy": 5.827615356445312,
      "epoch": 1.835620029712488,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004663359003803475,
      "loss": 5.2464,
      "mean_token_accuracy": 0.20711819231510162,
      "num_tokens": 42542502.0,
      "step": 21005
    },
    {
      "entropy": 5.892910575866699,
      "epoch": 1.8360569780651927,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004663195026301258,
      "loss": 5.2732,
      "mean_token_accuracy": 0.19816680997610092,
      "num_tokens": 42551837.0,
      "step": 21010
    },
    {
      "entropy": 5.977866458892822,
      "epoch": 1.8364939264178974,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046630310121029573,
      "loss": 5.3172,
      "mean_token_accuracy": 0.1943538159132004,
      "num_tokens": 42560537.0,
      "step": 21015
    },
    {
      "entropy": 5.940708303451538,
      "epoch": 1.8369308747706021,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046628669612117194,
      "loss": 5.3443,
      "mean_token_accuracy": 0.1930681511759758,
      "num_tokens": 42570386.0,
      "step": 21020
    },
    {
      "entropy": 5.919587755203247,
      "epoch": 1.8373678231233068,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046627028736306907,
      "loss": 5.3937,
      "mean_token_accuracy": 0.18806538581848145,
      "num_tokens": 42579860.0,
      "step": 21025
    },
    {
      "entropy": 5.991345167160034,
      "epoch": 1.8378047714760115,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046625387493630187,
      "loss": 5.3675,
      "mean_token_accuracy": 0.194059357047081,
      "num_tokens": 42590320.0,
      "step": 21030
    },
    {
      "entropy": 5.986344051361084,
      "epoch": 1.8382417198287162,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004662374588411852,
      "loss": 5.31,
      "mean_token_accuracy": 0.19763300865888594,
      "num_tokens": 42600331.0,
      "step": 21035
    },
    {
      "entropy": 5.893977308273316,
      "epoch": 1.8386786681814211,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046622103907803384,
      "loss": 5.3468,
      "mean_token_accuracy": 0.1962536409497261,
      "num_tokens": 42610860.0,
      "step": 21040
    },
    {
      "entropy": 5.938946390151978,
      "epoch": 1.8391156165341256,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046620461564716265,
      "loss": 5.3801,
      "mean_token_accuracy": 0.18710486590862274,
      "num_tokens": 42620305.0,
      "step": 21045
    },
    {
      "entropy": 5.9251714706420895,
      "epoch": 1.8395525648868305,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004661881885488869,
      "loss": 5.2049,
      "mean_token_accuracy": 0.21125410497188568,
      "num_tokens": 42629667.0,
      "step": 21050
    },
    {
      "entropy": 5.920599031448364,
      "epoch": 1.839989513239535,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004661717577835214,
      "loss": 5.3826,
      "mean_token_accuracy": 0.19067495465278625,
      "num_tokens": 42640116.0,
      "step": 21055
    },
    {
      "entropy": 5.91111273765564,
      "epoch": 1.8404264615922399,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004661553233513815,
      "loss": 5.2997,
      "mean_token_accuracy": 0.202205391228199,
      "num_tokens": 42650014.0,
      "step": 21060
    },
    {
      "entropy": 5.97265214920044,
      "epoch": 1.8408634099449444,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004661388852527822,
      "loss": 5.3934,
      "mean_token_accuracy": 0.18628674149513244,
      "num_tokens": 42661280.0,
      "step": 21065
    },
    {
      "entropy": 5.949058866500854,
      "epoch": 1.8413003582976493,
      "grad_norm": 1.0,
      "learning_rate": 0.00046612244348803906,
      "loss": 5.3928,
      "mean_token_accuracy": 0.19517471492290497,
      "num_tokens": 42671530.0,
      "step": 21070
    },
    {
      "entropy": 6.002533721923828,
      "epoch": 1.841737306650354,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004661059980574672,
      "loss": 5.3719,
      "mean_token_accuracy": 0.19360223710536956,
      "num_tokens": 42682738.0,
      "step": 21075
    },
    {
      "entropy": 5.906646680831909,
      "epoch": 1.8421742550030586,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004660895489613822,
      "loss": 5.352,
      "mean_token_accuracy": 0.19611034393310547,
      "num_tokens": 42694271.0,
      "step": 21080
    },
    {
      "entropy": 5.931552791595459,
      "epoch": 1.8426112033557633,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004660730962000996,
      "loss": 5.4181,
      "mean_token_accuracy": 0.19074644148349762,
      "num_tokens": 42704896.0,
      "step": 21085
    },
    {
      "entropy": 5.964112615585327,
      "epoch": 1.843048151708468,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004660566397739348,
      "loss": 5.2884,
      "mean_token_accuracy": 0.19581288546323777,
      "num_tokens": 42715506.0,
      "step": 21090
    },
    {
      "entropy": 5.990620708465576,
      "epoch": 1.8434851000611727,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046604017968320357,
      "loss": 5.3338,
      "mean_token_accuracy": 0.2024087205529213,
      "num_tokens": 42725428.0,
      "step": 21095
    },
    {
      "entropy": 5.899059629440307,
      "epoch": 1.8439220484138774,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004660237159282216,
      "loss": 5.3094,
      "mean_token_accuracy": 0.19873516708612443,
      "num_tokens": 42735105.0,
      "step": 21100
    },
    {
      "entropy": 5.877509450912475,
      "epoch": 1.8443589967665823,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046600724850930466,
      "loss": 5.2689,
      "mean_token_accuracy": 0.19708696752786636,
      "num_tokens": 42744191.0,
      "step": 21105
    },
    {
      "entropy": 5.975206422805786,
      "epoch": 1.8447959451192868,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004659907774267686,
      "loss": 5.4353,
      "mean_token_accuracy": 0.19464376121759414,
      "num_tokens": 42755416.0,
      "step": 21110
    },
    {
      "entropy": 5.980559015274048,
      "epoch": 1.8452328934719917,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046597430268092933,
      "loss": 5.2526,
      "mean_token_accuracy": 0.2008285865187645,
      "num_tokens": 42764974.0,
      "step": 21115
    },
    {
      "entropy": 5.905614519119263,
      "epoch": 1.8456698418246962,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046595782427210285,
      "loss": 5.2437,
      "mean_token_accuracy": 0.20124009400606155,
      "num_tokens": 42775080.0,
      "step": 21120
    },
    {
      "entropy": 5.843774080276489,
      "epoch": 1.846106790177401,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004659413422006053,
      "loss": 5.3165,
      "mean_token_accuracy": 0.19781703501939774,
      "num_tokens": 42784333.0,
      "step": 21125
    },
    {
      "entropy": 5.983361339569091,
      "epoch": 1.8465437385301058,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004659248564667527,
      "loss": 5.3408,
      "mean_token_accuracy": 0.19453159123659133,
      "num_tokens": 42792912.0,
      "step": 21130
    },
    {
      "entropy": 5.974687194824218,
      "epoch": 1.8469806868828105,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046590836707086135,
      "loss": 5.3688,
      "mean_token_accuracy": 0.19319060444831848,
      "num_tokens": 42803812.0,
      "step": 21135
    },
    {
      "entropy": 5.978670120239258,
      "epoch": 1.8474176352355152,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046589187401324753,
      "loss": 5.3956,
      "mean_token_accuracy": 0.19163028299808502,
      "num_tokens": 42815172.0,
      "step": 21140
    },
    {
      "entropy": 5.988814830780029,
      "epoch": 1.8478545835882199,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004658753772942274,
      "loss": 5.3008,
      "mean_token_accuracy": 0.19447272568941115,
      "num_tokens": 42824958.0,
      "step": 21145
    },
    {
      "entropy": 5.913511943817139,
      "epoch": 1.8482915319409245,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046585887691411764,
      "loss": 5.3744,
      "mean_token_accuracy": 0.1926693618297577,
      "num_tokens": 42834940.0,
      "step": 21150
    },
    {
      "entropy": 5.898491716384887,
      "epoch": 1.8487284802936292,
      "grad_norm": 1.0,
      "learning_rate": 0.0004658423728732346,
      "loss": 5.2604,
      "mean_token_accuracy": 0.2051940992474556,
      "num_tokens": 42845102.0,
      "step": 21155
    },
    {
      "entropy": 5.879827404022217,
      "epoch": 1.8491654286463342,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004658258651718948,
      "loss": 5.2798,
      "mean_token_accuracy": 0.20444120615720748,
      "num_tokens": 42853876.0,
      "step": 21160
    },
    {
      "entropy": 5.966984510421753,
      "epoch": 1.8496023769990386,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046580935381041496,
      "loss": 5.4405,
      "mean_token_accuracy": 0.18197776675224303,
      "num_tokens": 42863592.0,
      "step": 21165
    },
    {
      "entropy": 5.938733100891113,
      "epoch": 1.8500393253517435,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004657928387891117,
      "loss": 5.274,
      "mean_token_accuracy": 0.20574983805418015,
      "num_tokens": 42873478.0,
      "step": 21170
    },
    {
      "entropy": 5.9648510932922365,
      "epoch": 1.850476273704448,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046577632010830184,
      "loss": 5.2716,
      "mean_token_accuracy": 0.20267782658338546,
      "num_tokens": 42882591.0,
      "step": 21175
    },
    {
      "entropy": 5.847501802444458,
      "epoch": 1.850913222057153,
      "grad_norm": 1.125,
      "learning_rate": 0.0004657597977683022,
      "loss": 5.2214,
      "mean_token_accuracy": 0.20613889545202255,
      "num_tokens": 42892391.0,
      "step": 21180
    },
    {
      "entropy": 5.945397996902466,
      "epoch": 1.8513501704098576,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004657432717694297,
      "loss": 5.3212,
      "mean_token_accuracy": 0.19690980166196823,
      "num_tokens": 42902839.0,
      "step": 21185
    },
    {
      "entropy": 5.969271039962768,
      "epoch": 1.8517871187625623,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004657267421120012,
      "loss": 5.4155,
      "mean_token_accuracy": 0.18952429294586182,
      "num_tokens": 42912404.0,
      "step": 21190
    },
    {
      "entropy": 5.930845880508423,
      "epoch": 1.852224067115267,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004657102087963339,
      "loss": 5.3116,
      "mean_token_accuracy": 0.20272448658943176,
      "num_tokens": 42922939.0,
      "step": 21195
    },
    {
      "entropy": 5.890776634216309,
      "epoch": 1.8526610154679717,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046569367182274494,
      "loss": 5.3146,
      "mean_token_accuracy": 0.2009749412536621,
      "num_tokens": 42932095.0,
      "step": 21200
    },
    {
      "entropy": 5.957844209671021,
      "epoch": 1.8530979638206764,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004656771311915513,
      "loss": 5.3622,
      "mean_token_accuracy": 0.1947920560836792,
      "num_tokens": 42941910.0,
      "step": 21205
    },
    {
      "entropy": 5.984005355834961,
      "epoch": 1.853534912173381,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004656605869030704,
      "loss": 5.3263,
      "mean_token_accuracy": 0.19860579669475556,
      "num_tokens": 42952869.0,
      "step": 21210
    },
    {
      "entropy": 5.994774723052979,
      "epoch": 1.853971860526086,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046564403895761966,
      "loss": 5.4814,
      "mean_token_accuracy": 0.18472112566232682,
      "num_tokens": 42963300.0,
      "step": 21215
    },
    {
      "entropy": 5.954620742797852,
      "epoch": 1.8544088088787904,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004656274873555162,
      "loss": 5.2869,
      "mean_token_accuracy": 0.19838813692331314,
      "num_tokens": 42973505.0,
      "step": 21220
    },
    {
      "entropy": 5.949646949768066,
      "epoch": 1.8548457572314954,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004656109320970778,
      "loss": 5.3109,
      "mean_token_accuracy": 0.20087701380252837,
      "num_tokens": 42982813.0,
      "step": 21225
    },
    {
      "entropy": 5.885499572753906,
      "epoch": 1.8552827055841998,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004655943731826217,
      "loss": 5.3473,
      "mean_token_accuracy": 0.19702766835689545,
      "num_tokens": 42991792.0,
      "step": 21230
    },
    {
      "entropy": 5.843118190765381,
      "epoch": 1.8557196539369047,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004655778106124658,
      "loss": 5.2836,
      "mean_token_accuracy": 0.2065907284617424,
      "num_tokens": 43002191.0,
      "step": 21235
    },
    {
      "entropy": 5.933025693893432,
      "epoch": 1.8561566022896092,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004655612443869275,
      "loss": 5.287,
      "mean_token_accuracy": 0.20427079647779464,
      "num_tokens": 43013771.0,
      "step": 21240
    },
    {
      "entropy": 5.94587254524231,
      "epoch": 1.8565935506423141,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004655446745063247,
      "loss": 5.3804,
      "mean_token_accuracy": 0.18936565816402434,
      "num_tokens": 43023982.0,
      "step": 21245
    },
    {
      "entropy": 5.941466045379639,
      "epoch": 1.8570304989950188,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004655281009709753,
      "loss": 5.3953,
      "mean_token_accuracy": 0.18793707191944123,
      "num_tokens": 43034827.0,
      "step": 21250
    },
    {
      "entropy": 5.938536024093628,
      "epoch": 1.8574674473477235,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046551152378119696,
      "loss": 5.2741,
      "mean_token_accuracy": 0.20239160656929017,
      "num_tokens": 43044811.0,
      "step": 21255
    },
    {
      "entropy": 5.954619407653809,
      "epoch": 1.8579043957004282,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004654949429373078,
      "loss": 5.3329,
      "mean_token_accuracy": 0.19534917771816254,
      "num_tokens": 43054513.0,
      "step": 21260
    },
    {
      "entropy": 5.9240141868591305,
      "epoch": 1.858341344053133,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004654783584396258,
      "loss": 5.3451,
      "mean_token_accuracy": 0.1960986077785492,
      "num_tokens": 43064356.0,
      "step": 21265
    },
    {
      "entropy": 5.895987176895142,
      "epoch": 1.8587782924058376,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046546177028846915,
      "loss": 5.211,
      "mean_token_accuracy": 0.1998668998479843,
      "num_tokens": 43074663.0,
      "step": 21270
    },
    {
      "entropy": 5.936753463745117,
      "epoch": 1.8592152407585423,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004654451784841559,
      "loss": 5.3232,
      "mean_token_accuracy": 0.19607627093791963,
      "num_tokens": 43084542.0,
      "step": 21275
    },
    {
      "entropy": 5.85318021774292,
      "epoch": 1.8596521891112472,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046542858302700443,
      "loss": 5.1573,
      "mean_token_accuracy": 0.21706153750419616,
      "num_tokens": 43093864.0,
      "step": 21280
    },
    {
      "entropy": 5.868777561187744,
      "epoch": 1.8600891374639517,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046541198391733286,
      "loss": 5.2776,
      "mean_token_accuracy": 0.20117453634738922,
      "num_tokens": 43104541.0,
      "step": 21285
    },
    {
      "entropy": 5.919512462615967,
      "epoch": 1.8605260858166566,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004653953811554596,
      "loss": 5.3163,
      "mean_token_accuracy": 0.19514066278934478,
      "num_tokens": 43114378.0,
      "step": 21290
    },
    {
      "entropy": 6.008157682418823,
      "epoch": 1.860963034169361,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004653787747417033,
      "loss": 5.4821,
      "mean_token_accuracy": 0.1884505569934845,
      "num_tokens": 43125503.0,
      "step": 21295
    },
    {
      "entropy": 5.944977569580078,
      "epoch": 1.861399982522066,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046536216467638227,
      "loss": 5.2783,
      "mean_token_accuracy": 0.2091786414384842,
      "num_tokens": 43135034.0,
      "step": 21300
    },
    {
      "entropy": 5.922474431991577,
      "epoch": 1.8618369308747706,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004653455509598152,
      "loss": 5.3566,
      "mean_token_accuracy": 0.19675617665052414,
      "num_tokens": 43145359.0,
      "step": 21305
    },
    {
      "entropy": 5.881095218658447,
      "epoch": 1.8622738792274753,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004653289335923206,
      "loss": 5.2382,
      "mean_token_accuracy": 0.2055964782834053,
      "num_tokens": 43154995.0,
      "step": 21310
    },
    {
      "entropy": 5.873111009597778,
      "epoch": 1.86271082758018,
      "grad_norm": 1.125,
      "learning_rate": 0.0004653123125742175,
      "loss": 5.3232,
      "mean_token_accuracy": 0.19914344996213912,
      "num_tokens": 43164703.0,
      "step": 21315
    },
    {
      "entropy": 5.937816190719604,
      "epoch": 1.8631477759328847,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004652956879058245,
      "loss": 5.349,
      "mean_token_accuracy": 0.19687008410692214,
      "num_tokens": 43174193.0,
      "step": 21320
    },
    {
      "entropy": 5.928623056411743,
      "epoch": 1.8635847242855894,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046527905958746043,
      "loss": 5.366,
      "mean_token_accuracy": 0.20345349758863449,
      "num_tokens": 43184236.0,
      "step": 21325
    },
    {
      "entropy": 5.943773794174194,
      "epoch": 1.864021672638294,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046526242761944434,
      "loss": 5.3313,
      "mean_token_accuracy": 0.19828449189662933,
      "num_tokens": 43194421.0,
      "step": 21330
    },
    {
      "entropy": 5.922401857376099,
      "epoch": 1.864458620990999,
      "grad_norm": 1.125,
      "learning_rate": 0.00046524579200209513,
      "loss": 5.3664,
      "mean_token_accuracy": 0.20446736067533494,
      "num_tokens": 43205200.0,
      "step": 21335
    },
    {
      "entropy": 5.890233325958252,
      "epoch": 1.8648955693437035,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046522915273573194,
      "loss": 5.2972,
      "mean_token_accuracy": 0.20125289559364318,
      "num_tokens": 43215415.0,
      "step": 21340
    },
    {
      "entropy": 5.937063694000244,
      "epoch": 1.8653325176964084,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004652125098206739,
      "loss": 5.328,
      "mean_token_accuracy": 0.1990373969078064,
      "num_tokens": 43225648.0,
      "step": 21345
    },
    {
      "entropy": 5.970781660079956,
      "epoch": 1.8657694660491129,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046519586325724033,
      "loss": 5.3129,
      "mean_token_accuracy": 0.1940717428922653,
      "num_tokens": 43236358.0,
      "step": 21350
    },
    {
      "entropy": 5.993614625930786,
      "epoch": 1.8662064144018178,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046517921304575045,
      "loss": 5.3569,
      "mean_token_accuracy": 0.193955135345459,
      "num_tokens": 43247022.0,
      "step": 21355
    },
    {
      "entropy": 5.94713191986084,
      "epoch": 1.8666433627545223,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046516255918652356,
      "loss": 5.4298,
      "mean_token_accuracy": 0.19158256500959397,
      "num_tokens": 43258249.0,
      "step": 21360
    },
    {
      "entropy": 5.934100198745727,
      "epoch": 1.8670803111072272,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004651459016798791,
      "loss": 5.377,
      "mean_token_accuracy": 0.18905359506607056,
      "num_tokens": 43268997.0,
      "step": 21365
    },
    {
      "entropy": 5.930557632446289,
      "epoch": 1.8675172594599319,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004651292405261368,
      "loss": 5.3056,
      "mean_token_accuracy": 0.19773261845111847,
      "num_tokens": 43279165.0,
      "step": 21370
    },
    {
      "entropy": 5.908797883987427,
      "epoch": 1.8679542078126365,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046511257572561584,
      "loss": 5.3408,
      "mean_token_accuracy": 0.1998481944203377,
      "num_tokens": 43289778.0,
      "step": 21375
    },
    {
      "entropy": 5.881054925918579,
      "epoch": 1.8683911561653412,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004650959072786362,
      "loss": 5.2582,
      "mean_token_accuracy": 0.20054346472024917,
      "num_tokens": 43299865.0,
      "step": 21380
    },
    {
      "entropy": 5.934758853912354,
      "epoch": 1.868828104518046,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004650792351855173,
      "loss": 5.2962,
      "mean_token_accuracy": 0.19900370836257936,
      "num_tokens": 43309224.0,
      "step": 21385
    },
    {
      "entropy": 5.931779289245606,
      "epoch": 1.8692650528707508,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004650625594465792,
      "loss": 5.2426,
      "mean_token_accuracy": 0.19307797104120256,
      "num_tokens": 43320001.0,
      "step": 21390
    },
    {
      "entropy": 5.882133197784424,
      "epoch": 1.8697020012234553,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004650458800621415,
      "loss": 5.2664,
      "mean_token_accuracy": 0.19956571906805037,
      "num_tokens": 43329790.0,
      "step": 21395
    },
    {
      "entropy": 5.942141532897949,
      "epoch": 1.8701389495761602,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004650291970325244,
      "loss": 5.3838,
      "mean_token_accuracy": 0.19426932632923127,
      "num_tokens": 43340394.0,
      "step": 21400
    },
    {
      "entropy": 5.940850830078125,
      "epoch": 1.8705758979288647,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004650125103580476,
      "loss": 5.3826,
      "mean_token_accuracy": 0.1922239825129509,
      "num_tokens": 43351404.0,
      "step": 21405
    },
    {
      "entropy": 5.825051498413086,
      "epoch": 1.8710128462815696,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004649958200390313,
      "loss": 5.1788,
      "mean_token_accuracy": 0.21406774669885636,
      "num_tokens": 43361017.0,
      "step": 21410
    },
    {
      "entropy": 5.944527864456177,
      "epoch": 1.871449794634274,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004649791260757956,
      "loss": 5.3885,
      "mean_token_accuracy": 0.18815624415874482,
      "num_tokens": 43371680.0,
      "step": 21415
    },
    {
      "entropy": 5.9590959548950195,
      "epoch": 1.871886742986979,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004649624284686608,
      "loss": 5.3743,
      "mean_token_accuracy": 0.19624909907579421,
      "num_tokens": 43381908.0,
      "step": 21420
    },
    {
      "entropy": 5.981729984283447,
      "epoch": 1.8723236913396837,
      "grad_norm": 0.984375,
      "learning_rate": 0.000464945727217947,
      "loss": 5.4323,
      "mean_token_accuracy": 0.18663828670978547,
      "num_tokens": 43393234.0,
      "step": 21425
    },
    {
      "entropy": 6.002905035018921,
      "epoch": 1.8727606396923884,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046492902232397464,
      "loss": 5.4083,
      "mean_token_accuracy": 0.1924288511276245,
      "num_tokens": 43404065.0,
      "step": 21430
    },
    {
      "entropy": 5.918922710418701,
      "epoch": 1.873197588045093,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046491231378706415,
      "loss": 5.3538,
      "mean_token_accuracy": 0.1906503215432167,
      "num_tokens": 43414417.0,
      "step": 21435
    },
    {
      "entropy": 5.939477396011353,
      "epoch": 1.8736345363977978,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000464895601607536,
      "loss": 5.3287,
      "mean_token_accuracy": 0.19364565163850783,
      "num_tokens": 43424265.0,
      "step": 21440
    },
    {
      "entropy": 5.8904393196105955,
      "epoch": 1.8740714847505024,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004648788857857106,
      "loss": 5.2523,
      "mean_token_accuracy": 0.2018768906593323,
      "num_tokens": 43434699.0,
      "step": 21445
    },
    {
      "entropy": 5.954385900497437,
      "epoch": 1.8745084331032071,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046486216632190874,
      "loss": 5.3691,
      "mean_token_accuracy": 0.19419123977422714,
      "num_tokens": 43444113.0,
      "step": 21450
    },
    {
      "entropy": 5.903964281082153,
      "epoch": 1.874945381455912,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046484544321645103,
      "loss": 5.3057,
      "mean_token_accuracy": 0.1957397475838661,
      "num_tokens": 43454256.0,
      "step": 21455
    },
    {
      "entropy": 5.952550268173217,
      "epoch": 1.8753823298086165,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004648287164696583,
      "loss": 5.3031,
      "mean_token_accuracy": 0.19583788365125657,
      "num_tokens": 43463182.0,
      "step": 21460
    },
    {
      "entropy": 5.9132664680480955,
      "epoch": 1.8758192781613214,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004648119860818512,
      "loss": 5.2844,
      "mean_token_accuracy": 0.19898941814899446,
      "num_tokens": 43473326.0,
      "step": 21465
    },
    {
      "entropy": 5.922366714477539,
      "epoch": 1.876256226514026,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046479525205335084,
      "loss": 5.3559,
      "mean_token_accuracy": 0.19493156522512436,
      "num_tokens": 43483210.0,
      "step": 21470
    },
    {
      "entropy": 5.948488998413086,
      "epoch": 1.8766931748667308,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004647785143844781,
      "loss": 5.2988,
      "mean_token_accuracy": 0.20058835595846175,
      "num_tokens": 43493834.0,
      "step": 21475
    },
    {
      "entropy": 5.917633008956909,
      "epoch": 1.8771301232194355,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046476177307555394,
      "loss": 5.3143,
      "mean_token_accuracy": 0.1953283354640007,
      "num_tokens": 43504522.0,
      "step": 21480
    },
    {
      "entropy": 5.962610054016113,
      "epoch": 1.8775670715721402,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046474502812689954,
      "loss": 5.4125,
      "mean_token_accuracy": 0.19478332251310349,
      "num_tokens": 43514950.0,
      "step": 21485
    },
    {
      "entropy": 5.937058782577514,
      "epoch": 1.878004019924845,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004647282795388361,
      "loss": 5.3388,
      "mean_token_accuracy": 0.1986715629696846,
      "num_tokens": 43526463.0,
      "step": 21490
    },
    {
      "entropy": 5.902377557754517,
      "epoch": 1.8784409682775496,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004647115273116849,
      "loss": 5.3368,
      "mean_token_accuracy": 0.19612595289945603,
      "num_tokens": 43536104.0,
      "step": 21495
    },
    {
      "entropy": 6.010428762435913,
      "epoch": 1.8788779166302543,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046469477144576716,
      "loss": 5.4517,
      "mean_token_accuracy": 0.18943212777376175,
      "num_tokens": 43547863.0,
      "step": 21500
    },
    {
      "entropy": 5.996421241760254,
      "epoch": 1.879314864982959,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004646780119414043,
      "loss": 5.3217,
      "mean_token_accuracy": 0.19182891100645066,
      "num_tokens": 43558024.0,
      "step": 21505
    },
    {
      "entropy": 5.974741840362549,
      "epoch": 1.8797518133356639,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004646612487989178,
      "loss": 5.3452,
      "mean_token_accuracy": 0.2009940966963768,
      "num_tokens": 43567464.0,
      "step": 21510
    },
    {
      "entropy": 5.878597354888916,
      "epoch": 1.8801887616883683,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004646444820186291,
      "loss": 5.2606,
      "mean_token_accuracy": 0.2036131128668785,
      "num_tokens": 43576826.0,
      "step": 21515
    },
    {
      "entropy": 5.870602321624756,
      "epoch": 1.8806257100410733,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046462771160085995,
      "loss": 5.2687,
      "mean_token_accuracy": 0.20124755799770355,
      "num_tokens": 43586106.0,
      "step": 21520
    },
    {
      "entropy": 5.947538089752197,
      "epoch": 1.8810626583937777,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046461093754593185,
      "loss": 5.3962,
      "mean_token_accuracy": 0.1871456876397133,
      "num_tokens": 43597077.0,
      "step": 21525
    },
    {
      "entropy": 5.979398488998413,
      "epoch": 1.8814996067464826,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004645941598541667,
      "loss": 5.3906,
      "mean_token_accuracy": 0.1868507519364357,
      "num_tokens": 43608773.0,
      "step": 21530
    },
    {
      "entropy": 6.055860137939453,
      "epoch": 1.8819365550991871,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004645773785258862,
      "loss": 5.3453,
      "mean_token_accuracy": 0.20012060552835464,
      "num_tokens": 43618017.0,
      "step": 21535
    },
    {
      "entropy": 5.966092491149903,
      "epoch": 1.882373503451892,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004645605935614122,
      "loss": 5.3451,
      "mean_token_accuracy": 0.19194610118865968,
      "num_tokens": 43627865.0,
      "step": 21540
    },
    {
      "entropy": 5.96460862159729,
      "epoch": 1.8828104518045967,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046454380496106665,
      "loss": 5.3518,
      "mean_token_accuracy": 0.19304685443639755,
      "num_tokens": 43637772.0,
      "step": 21545
    },
    {
      "entropy": 5.908488130569458,
      "epoch": 1.8832474001573014,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046452701272517167,
      "loss": 5.2861,
      "mean_token_accuracy": 0.19981013238430023,
      "num_tokens": 43648242.0,
      "step": 21550
    },
    {
      "entropy": 5.910612297058106,
      "epoch": 1.883684348510006,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046451021685404927,
      "loss": 5.3844,
      "mean_token_accuracy": 0.19788715243339539,
      "num_tokens": 43658298.0,
      "step": 21555
    },
    {
      "entropy": 5.9401620388031,
      "epoch": 1.8841212968627108,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046449341734802155,
      "loss": 5.3284,
      "mean_token_accuracy": 0.1935879111289978,
      "num_tokens": 43667792.0,
      "step": 21560
    },
    {
      "entropy": 5.912754726409912,
      "epoch": 1.8845582452154155,
      "grad_norm": 1.0,
      "learning_rate": 0.0004644766142074108,
      "loss": 5.244,
      "mean_token_accuracy": 0.21089005917310716,
      "num_tokens": 43678571.0,
      "step": 21565
    },
    {
      "entropy": 5.947126674652099,
      "epoch": 1.8849951935681202,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046445980743253936,
      "loss": 5.385,
      "mean_token_accuracy": 0.1924894392490387,
      "num_tokens": 43690851.0,
      "step": 21570
    },
    {
      "entropy": 5.913163757324218,
      "epoch": 1.885432141920825,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046444299702372955,
      "loss": 5.3124,
      "mean_token_accuracy": 0.19926949292421342,
      "num_tokens": 43700960.0,
      "step": 21575
    },
    {
      "entropy": 5.98815689086914,
      "epoch": 1.8858690902735296,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004644261829813038,
      "loss": 5.2961,
      "mean_token_accuracy": 0.2006883904337883,
      "num_tokens": 43710466.0,
      "step": 21580
    },
    {
      "entropy": 5.954820585250855,
      "epoch": 1.8863060386262345,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046440936530558455,
      "loss": 5.3627,
      "mean_token_accuracy": 0.1953015446662903,
      "num_tokens": 43720261.0,
      "step": 21585
    },
    {
      "entropy": 5.939949226379395,
      "epoch": 1.886742986978939,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004643925439968944,
      "loss": 5.3579,
      "mean_token_accuracy": 0.19306943118572234,
      "num_tokens": 43731045.0,
      "step": 21590
    },
    {
      "entropy": 5.998937177658081,
      "epoch": 1.8871799353316439,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046437571905555604,
      "loss": 5.3614,
      "mean_token_accuracy": 0.19398723691701888,
      "num_tokens": 43741873.0,
      "step": 21595
    },
    {
      "entropy": 5.94652829170227,
      "epoch": 1.8876168836843485,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046435889048189216,
      "loss": 5.3661,
      "mean_token_accuracy": 0.19809812307357788,
      "num_tokens": 43751473.0,
      "step": 21600
    },
    {
      "entropy": 5.934212493896484,
      "epoch": 1.8880538320370532,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046434205827622555,
      "loss": 5.3669,
      "mean_token_accuracy": 0.1967080146074295,
      "num_tokens": 43762517.0,
      "step": 21605
    },
    {
      "entropy": 6.011364889144898,
      "epoch": 1.888490780389758,
      "grad_norm": 1.125,
      "learning_rate": 0.0004643252224388791,
      "loss": 5.3905,
      "mean_token_accuracy": 0.19826458245515824,
      "num_tokens": 43772491.0,
      "step": 21610
    },
    {
      "entropy": 5.960420894622803,
      "epoch": 1.8889277287424626,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046430838297017556,
      "loss": 5.3672,
      "mean_token_accuracy": 0.1913052260875702,
      "num_tokens": 43781646.0,
      "step": 21615
    },
    {
      "entropy": 5.919161701202393,
      "epoch": 1.8893646770951673,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046429153987043815,
      "loss": 5.2519,
      "mean_token_accuracy": 0.206016443669796,
      "num_tokens": 43792020.0,
      "step": 21620
    },
    {
      "entropy": 5.91430287361145,
      "epoch": 1.889801625447872,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004642746931399898,
      "loss": 5.2625,
      "mean_token_accuracy": 0.20038066953420638,
      "num_tokens": 43801638.0,
      "step": 21625
    },
    {
      "entropy": 5.976167488098144,
      "epoch": 1.890238573800577,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046425784277915363,
      "loss": 5.4037,
      "mean_token_accuracy": 0.18483459800481797,
      "num_tokens": 43811475.0,
      "step": 21630
    },
    {
      "entropy": 5.976805639266968,
      "epoch": 1.8906755221532814,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004642409887882528,
      "loss": 5.3157,
      "mean_token_accuracy": 0.19509542882442474,
      "num_tokens": 43820811.0,
      "step": 21635
    },
    {
      "entropy": 5.999322319030762,
      "epoch": 1.8911124705059863,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004642241311676107,
      "loss": 5.3912,
      "mean_token_accuracy": 0.18454687148332596,
      "num_tokens": 43830804.0,
      "step": 21640
    },
    {
      "entropy": 5.851601839065552,
      "epoch": 1.8915494188586908,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004642072699175506,
      "loss": 5.2023,
      "mean_token_accuracy": 0.21547115296125413,
      "num_tokens": 43840156.0,
      "step": 21645
    },
    {
      "entropy": 5.943044853210449,
      "epoch": 1.8919863672113957,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000464190405038396,
      "loss": 5.3116,
      "mean_token_accuracy": 0.19481014609336852,
      "num_tokens": 43849622.0,
      "step": 21650
    },
    {
      "entropy": 5.982365989685059,
      "epoch": 1.8924233155641004,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004641735365304703,
      "loss": 5.3856,
      "mean_token_accuracy": 0.18889328092336655,
      "num_tokens": 43860184.0,
      "step": 21655
    },
    {
      "entropy": 5.84652271270752,
      "epoch": 1.892860263916805,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046415666439409694,
      "loss": 5.2208,
      "mean_token_accuracy": 0.21641599237918854,
      "num_tokens": 43870157.0,
      "step": 21660
    },
    {
      "entropy": 5.924202966690063,
      "epoch": 1.8932972122695098,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046413978862959964,
      "loss": 5.2998,
      "mean_token_accuracy": 0.1968616470694542,
      "num_tokens": 43879867.0,
      "step": 21665
    },
    {
      "entropy": 5.860871410369873,
      "epoch": 1.8937341606222144,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004641229092373022,
      "loss": 5.326,
      "mean_token_accuracy": 0.1985461965203285,
      "num_tokens": 43890314.0,
      "step": 21670
    },
    {
      "entropy": 5.990491771697998,
      "epoch": 1.8941711089749191,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004641060262175282,
      "loss": 5.3443,
      "mean_token_accuracy": 0.1971310168504715,
      "num_tokens": 43901075.0,
      "step": 21675
    },
    {
      "entropy": 5.964300060272217,
      "epoch": 1.8946080573276238,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046408913957060144,
      "loss": 5.3734,
      "mean_token_accuracy": 0.1927620694041252,
      "num_tokens": 43911777.0,
      "step": 21680
    },
    {
      "entropy": 5.923954057693481,
      "epoch": 1.8950450056803287,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046407224929684597,
      "loss": 5.289,
      "mean_token_accuracy": 0.2016446128487587,
      "num_tokens": 43921901.0,
      "step": 21685
    },
    {
      "entropy": 5.9573173999786375,
      "epoch": 1.8954819540330332,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004640553553965857,
      "loss": 5.3762,
      "mean_token_accuracy": 0.1946873188018799,
      "num_tokens": 43931142.0,
      "step": 21690
    },
    {
      "entropy": 5.95158371925354,
      "epoch": 1.8959189023857381,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004640384578701447,
      "loss": 5.294,
      "mean_token_accuracy": 0.20429461151361467,
      "num_tokens": 43940642.0,
      "step": 21695
    },
    {
      "entropy": 5.99565863609314,
      "epoch": 1.8963558507384426,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046402155671784696,
      "loss": 5.3706,
      "mean_token_accuracy": 0.18796294927597046,
      "num_tokens": 43950808.0,
      "step": 21700
    },
    {
      "entropy": 6.0462713718414305,
      "epoch": 1.8967927990911475,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046400465194001674,
      "loss": 5.4325,
      "mean_token_accuracy": 0.18194012343883514,
      "num_tokens": 43961349.0,
      "step": 21705
    },
    {
      "entropy": 5.961616802215576,
      "epoch": 1.897229747443852,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046398774353697826,
      "loss": 5.3047,
      "mean_token_accuracy": 0.19766228944063186,
      "num_tokens": 43970072.0,
      "step": 21710
    },
    {
      "entropy": 5.961359262466431,
      "epoch": 1.897666695796557,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004639708315090558,
      "loss": 5.3737,
      "mean_token_accuracy": 0.19853961020708083,
      "num_tokens": 43980076.0,
      "step": 21715
    },
    {
      "entropy": 5.882131481170655,
      "epoch": 1.8981036441492616,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004639539158565738,
      "loss": 5.185,
      "mean_token_accuracy": 0.22039134204387664,
      "num_tokens": 43989876.0,
      "step": 21720
    },
    {
      "entropy": 5.961143875122071,
      "epoch": 1.8985405925019663,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004639369965798567,
      "loss": 5.3748,
      "mean_token_accuracy": 0.18701271265745162,
      "num_tokens": 44000920.0,
      "step": 21725
    },
    {
      "entropy": 5.88689398765564,
      "epoch": 1.898977540854671,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046392007367922894,
      "loss": 5.2233,
      "mean_token_accuracy": 0.2015180006623268,
      "num_tokens": 44011539.0,
      "step": 21730
    },
    {
      "entropy": 5.947545909881592,
      "epoch": 1.8994144892073757,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004639031471550152,
      "loss": 5.3276,
      "mean_token_accuracy": 0.19608149230480193,
      "num_tokens": 44022624.0,
      "step": 21735
    },
    {
      "entropy": 5.92074065208435,
      "epoch": 1.8998514375600803,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046388621700754014,
      "loss": 5.288,
      "mean_token_accuracy": 0.19925405979156494,
      "num_tokens": 44033105.0,
      "step": 21740
    },
    {
      "entropy": 5.958618545532227,
      "epoch": 1.900288385912785,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046386928323712847,
      "loss": 5.298,
      "mean_token_accuracy": 0.19227776527404786,
      "num_tokens": 44042805.0,
      "step": 21745
    },
    {
      "entropy": 5.897384548187256,
      "epoch": 1.90072533426549,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000463852345844105,
      "loss": 5.2523,
      "mean_token_accuracy": 0.20703170895576478,
      "num_tokens": 44052471.0,
      "step": 21750
    },
    {
      "entropy": 5.9032227993011475,
      "epoch": 1.9011622826181944,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004638354048287945,
      "loss": 5.3887,
      "mean_token_accuracy": 0.19168189615011216,
      "num_tokens": 44063358.0,
      "step": 21755
    },
    {
      "entropy": 5.926678800582886,
      "epoch": 1.9015992309708993,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046381846019152194,
      "loss": 5.3576,
      "mean_token_accuracy": 0.19664463847875596,
      "num_tokens": 44073598.0,
      "step": 21760
    },
    {
      "entropy": 5.987596893310547,
      "epoch": 1.9020361793236038,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046380151193261247,
      "loss": 5.3306,
      "mean_token_accuracy": 0.19762443155050277,
      "num_tokens": 44084752.0,
      "step": 21765
    },
    {
      "entropy": 5.913965702056885,
      "epoch": 1.9024731276763087,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004637845600523911,
      "loss": 5.2884,
      "mean_token_accuracy": 0.20246027261018754,
      "num_tokens": 44094857.0,
      "step": 21770
    },
    {
      "entropy": 5.982855606079101,
      "epoch": 1.9029100760290134,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004637676045511829,
      "loss": 5.3665,
      "mean_token_accuracy": 0.19244514405727386,
      "num_tokens": 44105700.0,
      "step": 21775
    },
    {
      "entropy": 5.925276756286621,
      "epoch": 1.903347024381718,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004637506454293131,
      "loss": 5.2999,
      "mean_token_accuracy": 0.20221123695373536,
      "num_tokens": 44116588.0,
      "step": 21780
    },
    {
      "entropy": 5.941097688674927,
      "epoch": 1.9037839727344228,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004637336826871071,
      "loss": 5.3601,
      "mean_token_accuracy": 0.19974849969148636,
      "num_tokens": 44126948.0,
      "step": 21785
    },
    {
      "entropy": 5.960813522338867,
      "epoch": 1.9042209210871275,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046371671632489017,
      "loss": 5.4277,
      "mean_token_accuracy": 0.1894749879837036,
      "num_tokens": 44136239.0,
      "step": 21790
    },
    {
      "entropy": 5.8970935344696045,
      "epoch": 1.9046578694398322,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046369974634298766,
      "loss": 5.2212,
      "mean_token_accuracy": 0.2067580983042717,
      "num_tokens": 44146314.0,
      "step": 21795
    },
    {
      "entropy": 5.925810289382935,
      "epoch": 1.9050948177925369,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004636827727417252,
      "loss": 5.3561,
      "mean_token_accuracy": 0.20278082489967347,
      "num_tokens": 44155767.0,
      "step": 21800
    },
    {
      "entropy": 5.966243839263916,
      "epoch": 1.9055317661452418,
      "grad_norm": 1.125,
      "learning_rate": 0.00046366579552142823,
      "loss": 5.3556,
      "mean_token_accuracy": 0.1976285845041275,
      "num_tokens": 44165830.0,
      "step": 21805
    },
    {
      "entropy": 5.938599586486816,
      "epoch": 1.9059687144979462,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004636488146824225,
      "loss": 5.2358,
      "mean_token_accuracy": 0.20594149976968765,
      "num_tokens": 44175526.0,
      "step": 21810
    },
    {
      "entropy": 5.969092750549317,
      "epoch": 1.9064056628506512,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004636318302250336,
      "loss": 5.3692,
      "mean_token_accuracy": 0.19839905202388763,
      "num_tokens": 44185950.0,
      "step": 21815
    },
    {
      "entropy": 5.944149923324585,
      "epoch": 1.9068426112033556,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004636148421495874,
      "loss": 5.2881,
      "mean_token_accuracy": 0.19986458271741867,
      "num_tokens": 44196518.0,
      "step": 21820
    },
    {
      "entropy": 5.9532393455505375,
      "epoch": 1.9072795595560605,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046359785045640967,
      "loss": 5.395,
      "mean_token_accuracy": 0.1968684896826744,
      "num_tokens": 44207709.0,
      "step": 21825
    },
    {
      "entropy": 5.988089466094971,
      "epoch": 1.9077165079087652,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004635808551458264,
      "loss": 5.3682,
      "mean_token_accuracy": 0.18828809410333633,
      "num_tokens": 44218311.0,
      "step": 21830
    },
    {
      "entropy": 5.977954721450805,
      "epoch": 1.90815345626147,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046356385621816343,
      "loss": 5.3183,
      "mean_token_accuracy": 0.19434832483530046,
      "num_tokens": 44229059.0,
      "step": 21835
    },
    {
      "entropy": 5.952482795715332,
      "epoch": 1.9085904046141746,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004635468536737469,
      "loss": 5.3233,
      "mean_token_accuracy": 0.19442462772130967,
      "num_tokens": 44239254.0,
      "step": 21840
    },
    {
      "entropy": 5.979149293899536,
      "epoch": 1.9090273529668793,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046352984751290297,
      "loss": 5.2229,
      "mean_token_accuracy": 0.20315756201744078,
      "num_tokens": 44248465.0,
      "step": 21845
    },
    {
      "entropy": 5.9396350383758545,
      "epoch": 1.909464301319584,
      "grad_norm": 1.0,
      "learning_rate": 0.00046351283773595777,
      "loss": 5.3247,
      "mean_token_accuracy": 0.1949627786874771,
      "num_tokens": 44259121.0,
      "step": 21850
    },
    {
      "entropy": 5.964802026748657,
      "epoch": 1.9099012496722887,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046349582434323757,
      "loss": 5.3074,
      "mean_token_accuracy": 0.20483284145593644,
      "num_tokens": 44269291.0,
      "step": 21855
    },
    {
      "entropy": 5.88116569519043,
      "epoch": 1.9103381980249936,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046347880733506867,
      "loss": 5.3104,
      "mean_token_accuracy": 0.19410927444696427,
      "num_tokens": 44280985.0,
      "step": 21860
    },
    {
      "entropy": 5.911561918258667,
      "epoch": 1.910775146377698,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046346178671177746,
      "loss": 5.3061,
      "mean_token_accuracy": 0.2005904421210289,
      "num_tokens": 44290370.0,
      "step": 21865
    },
    {
      "entropy": 5.889088201522827,
      "epoch": 1.911212094730403,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046344476247369044,
      "loss": 5.3018,
      "mean_token_accuracy": 0.19509860277175903,
      "num_tokens": 44300670.0,
      "step": 21870
    },
    {
      "entropy": 5.96503005027771,
      "epoch": 1.9116490430831075,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046342773462113414,
      "loss": 5.3421,
      "mean_token_accuracy": 0.19909617751836778,
      "num_tokens": 44311578.0,
      "step": 21875
    },
    {
      "entropy": 5.911732006072998,
      "epoch": 1.9120859914358124,
      "grad_norm": 1.125,
      "learning_rate": 0.00046341070315443516,
      "loss": 5.2633,
      "mean_token_accuracy": 0.20184085965156556,
      "num_tokens": 44321692.0,
      "step": 21880
    },
    {
      "entropy": 5.857010364532471,
      "epoch": 1.9125229397885168,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046339366807392017,
      "loss": 5.3016,
      "mean_token_accuracy": 0.20018395483493806,
      "num_tokens": 44331468.0,
      "step": 21885
    },
    {
      "entropy": 5.938638353347779,
      "epoch": 1.9129598881412218,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046337662937991585,
      "loss": 5.3314,
      "mean_token_accuracy": 0.19425952881574632,
      "num_tokens": 44341491.0,
      "step": 21890
    },
    {
      "entropy": 5.96850552558899,
      "epoch": 1.9133968364939264,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004633595870727492,
      "loss": 5.3991,
      "mean_token_accuracy": 0.19119295179843904,
      "num_tokens": 44351568.0,
      "step": 21895
    },
    {
      "entropy": 6.011800956726074,
      "epoch": 1.9138337848466311,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046334254115274685,
      "loss": 5.3964,
      "mean_token_accuracy": 0.19100967645645142,
      "num_tokens": 44362009.0,
      "step": 21900
    },
    {
      "entropy": 5.9154994010925295,
      "epoch": 1.9142707331993358,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004633254916202359,
      "loss": 5.2696,
      "mean_token_accuracy": 0.20538127571344375,
      "num_tokens": 44372671.0,
      "step": 21905
    },
    {
      "entropy": 5.884560918807983,
      "epoch": 1.9147076815520405,
      "grad_norm": 1.125,
      "learning_rate": 0.0004633084384755434,
      "loss": 5.2812,
      "mean_token_accuracy": 0.19949712604284286,
      "num_tokens": 44382658.0,
      "step": 21910
    },
    {
      "entropy": 5.931003522872925,
      "epoch": 1.9151446299047452,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004632913817189963,
      "loss": 5.3017,
      "mean_token_accuracy": 0.2026939332485199,
      "num_tokens": 44392090.0,
      "step": 21915
    },
    {
      "entropy": 5.96098108291626,
      "epoch": 1.91558157825745,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004632743213509219,
      "loss": 5.3745,
      "mean_token_accuracy": 0.1910662218928337,
      "num_tokens": 44402896.0,
      "step": 21920
    },
    {
      "entropy": 5.988338041305542,
      "epoch": 1.9160185266101548,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046325725737164734,
      "loss": 5.3362,
      "mean_token_accuracy": 0.19669898599386215,
      "num_tokens": 44413700.0,
      "step": 21925
    },
    {
      "entropy": 5.988709163665772,
      "epoch": 1.9164554749628593,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004632401897814999,
      "loss": 5.2451,
      "mean_token_accuracy": 0.20614902973175048,
      "num_tokens": 44423442.0,
      "step": 21930
    },
    {
      "entropy": 5.8576476097106935,
      "epoch": 1.9168924233155642,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004632231185808071,
      "loss": 5.2924,
      "mean_token_accuracy": 0.20079896748065948,
      "num_tokens": 44434197.0,
      "step": 21935
    },
    {
      "entropy": 5.877588558197021,
      "epoch": 1.9173293716682687,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046320604376989604,
      "loss": 5.2897,
      "mean_token_accuracy": 0.20349168926477432,
      "num_tokens": 44444483.0,
      "step": 21940
    },
    {
      "entropy": 5.938827610015869,
      "epoch": 1.9177663200209736,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004631889653490946,
      "loss": 5.3055,
      "mean_token_accuracy": 0.20383722633123397,
      "num_tokens": 44456303.0,
      "step": 21945
    },
    {
      "entropy": 5.937610244750976,
      "epoch": 1.9182032683736783,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046317188331873024,
      "loss": 5.3709,
      "mean_token_accuracy": 0.20012621730566024,
      "num_tokens": 44467314.0,
      "step": 21950
    },
    {
      "entropy": 5.979625415802002,
      "epoch": 1.918640216726383,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004631547976791305,
      "loss": 5.3325,
      "mean_token_accuracy": 0.2003057762980461,
      "num_tokens": 44476466.0,
      "step": 21955
    },
    {
      "entropy": 5.942109107971191,
      "epoch": 1.9190771650790877,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004631377084306231,
      "loss": 5.3263,
      "mean_token_accuracy": 0.1944798246026039,
      "num_tokens": 44486247.0,
      "step": 21960
    },
    {
      "entropy": 5.918974876403809,
      "epoch": 1.9195141134317923,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046312061557353585,
      "loss": 5.3077,
      "mean_token_accuracy": 0.20047129392623902,
      "num_tokens": 44497166.0,
      "step": 21965
    },
    {
      "entropy": 5.999119901657105,
      "epoch": 1.919951061784497,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004631035191081967,
      "loss": 5.3395,
      "mean_token_accuracy": 0.1921832635998726,
      "num_tokens": 44507677.0,
      "step": 21970
    },
    {
      "entropy": 5.901353168487549,
      "epoch": 1.9203880101372017,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004630864190349334,
      "loss": 5.2926,
      "mean_token_accuracy": 0.20851749181747437,
      "num_tokens": 44517392.0,
      "step": 21975
    },
    {
      "entropy": 5.971704912185669,
      "epoch": 1.9208249584899066,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046306931535407416,
      "loss": 5.3244,
      "mean_token_accuracy": 0.19857660979032515,
      "num_tokens": 44528914.0,
      "step": 21980
    },
    {
      "entropy": 6.008872556686401,
      "epoch": 1.9212619068426111,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046305220806594683,
      "loss": 5.4581,
      "mean_token_accuracy": 0.18548050820827483,
      "num_tokens": 44539419.0,
      "step": 21985
    },
    {
      "entropy": 5.928853559494018,
      "epoch": 1.921698855195316,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004630350971708797,
      "loss": 5.2847,
      "mean_token_accuracy": 0.20330277234315872,
      "num_tokens": 44548704.0,
      "step": 21990
    },
    {
      "entropy": 5.907692909240723,
      "epoch": 1.9221358035480205,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046301798266920076,
      "loss": 5.3069,
      "mean_token_accuracy": 0.20045274645090103,
      "num_tokens": 44559110.0,
      "step": 21995
    },
    {
      "entropy": 5.995951080322266,
      "epoch": 1.9225727519007254,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004630008645612384,
      "loss": 5.3768,
      "mean_token_accuracy": 0.1905987501144409,
      "num_tokens": 44570414.0,
      "step": 22000
    },
    {
      "entropy": 6.01231837272644,
      "epoch": 1.92300970025343,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046298374284732096,
      "loss": 5.4545,
      "mean_token_accuracy": 0.19143782556056976,
      "num_tokens": 44580194.0,
      "step": 22005
    },
    {
      "entropy": 5.9455677509307865,
      "epoch": 1.9234466486061348,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004629666175277768,
      "loss": 5.3167,
      "mean_token_accuracy": 0.1990347146987915,
      "num_tokens": 44590174.0,
      "step": 22010
    },
    {
      "entropy": 5.9913990020751955,
      "epoch": 1.9238835969588395,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004629494886029345,
      "loss": 5.4031,
      "mean_token_accuracy": 0.19456478655338288,
      "num_tokens": 44600940.0,
      "step": 22015
    },
    {
      "entropy": 5.98202600479126,
      "epoch": 1.9243205453115442,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004629323560731225,
      "loss": 5.3787,
      "mean_token_accuracy": 0.19481016397476197,
      "num_tokens": 44610852.0,
      "step": 22020
    },
    {
      "entropy": 6.000680732727051,
      "epoch": 1.9247574936642489,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046291521993866943,
      "loss": 5.3942,
      "mean_token_accuracy": 0.19859048277139663,
      "num_tokens": 44621479.0,
      "step": 22025
    },
    {
      "entropy": 5.9784716129302975,
      "epoch": 1.9251944420169536,
      "grad_norm": 1.125,
      "learning_rate": 0.000462898080199904,
      "loss": 5.3336,
      "mean_token_accuracy": 0.19811972677707673,
      "num_tokens": 44631838.0,
      "step": 22030
    },
    {
      "entropy": 5.9259813785552975,
      "epoch": 1.9256313903696585,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004628809368571549,
      "loss": 5.3519,
      "mean_token_accuracy": 0.19119983166456223,
      "num_tokens": 44643103.0,
      "step": 22035
    },
    {
      "entropy": 6.019061660766601,
      "epoch": 1.926068338722363,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046286378991075095,
      "loss": 5.474,
      "mean_token_accuracy": 0.18991241306066514,
      "num_tokens": 44653659.0,
      "step": 22040
    },
    {
      "entropy": 6.012672710418701,
      "epoch": 1.9265052870750679,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004628466393610211,
      "loss": 5.3414,
      "mean_token_accuracy": 0.1957229733467102,
      "num_tokens": 44665064.0,
      "step": 22045
    },
    {
      "entropy": 5.969572639465332,
      "epoch": 1.9269422354277723,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046282948520829433,
      "loss": 5.2912,
      "mean_token_accuracy": 0.20084867626428604,
      "num_tokens": 44674256.0,
      "step": 22050
    },
    {
      "entropy": 5.9004417896270756,
      "epoch": 1.9273791837804772,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004628123274528996,
      "loss": 5.2774,
      "mean_token_accuracy": 0.19998051226139069,
      "num_tokens": 44683715.0,
      "step": 22055
    },
    {
      "entropy": 5.992755889892578,
      "epoch": 1.9278161321331817,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046279516609516594,
      "loss": 5.4665,
      "mean_token_accuracy": 0.18748639822006224,
      "num_tokens": 44694637.0,
      "step": 22060
    },
    {
      "entropy": 5.991275310516357,
      "epoch": 1.9282530804858866,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004627780011354226,
      "loss": 5.3138,
      "mean_token_accuracy": 0.20153194814920425,
      "num_tokens": 44705307.0,
      "step": 22065
    },
    {
      "entropy": 5.929367971420288,
      "epoch": 1.9286900288385913,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004627608325739989,
      "loss": 5.3562,
      "mean_token_accuracy": 0.1989276796579361,
      "num_tokens": 44715293.0,
      "step": 22070
    },
    {
      "entropy": 5.979690456390381,
      "epoch": 1.929126977191296,
      "grad_norm": 1.109375,
      "learning_rate": 0.000462743660411224,
      "loss": 5.4146,
      "mean_token_accuracy": 0.18428247570991516,
      "num_tokens": 44726406.0,
      "step": 22075
    },
    {
      "entropy": 6.009899663925171,
      "epoch": 1.9295639255440007,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046272648464742733,
      "loss": 5.3537,
      "mean_token_accuracy": 0.19465059638023377,
      "num_tokens": 44737166.0,
      "step": 22080
    },
    {
      "entropy": 5.91895112991333,
      "epoch": 1.9300008738967054,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046270930528293826,
      "loss": 5.2532,
      "mean_token_accuracy": 0.2024080738425255,
      "num_tokens": 44746693.0,
      "step": 22085
    },
    {
      "entropy": 5.930724430084228,
      "epoch": 1.93043782224941,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004626921223180864,
      "loss": 5.2831,
      "mean_token_accuracy": 0.2028308629989624,
      "num_tokens": 44757499.0,
      "step": 22090
    },
    {
      "entropy": 5.914637422561645,
      "epoch": 1.9308747706021148,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046267493575320127,
      "loss": 5.3339,
      "mean_token_accuracy": 0.19746533632278443,
      "num_tokens": 44768112.0,
      "step": 22095
    },
    {
      "entropy": 5.920431137084961,
      "epoch": 1.9313117189548197,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004626577455886125,
      "loss": 5.2678,
      "mean_token_accuracy": 0.19773029088973998,
      "num_tokens": 44778317.0,
      "step": 22100
    },
    {
      "entropy": 5.965678310394287,
      "epoch": 1.9317486673075241,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004626405518246499,
      "loss": 5.41,
      "mean_token_accuracy": 0.1913309007883072,
      "num_tokens": 44790483.0,
      "step": 22105
    },
    {
      "entropy": 5.936016368865967,
      "epoch": 1.932185615660229,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046262335446164315,
      "loss": 5.2804,
      "mean_token_accuracy": 0.20394100844860077,
      "num_tokens": 44799891.0,
      "step": 22110
    },
    {
      "entropy": 5.946571731567383,
      "epoch": 1.9326225640129335,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004626061534999222,
      "loss": 5.3722,
      "mean_token_accuracy": 0.19222929030656816,
      "num_tokens": 44810728.0,
      "step": 22115
    },
    {
      "entropy": 5.9826170921325685,
      "epoch": 1.9330595123656384,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046258894893981686,
      "loss": 5.3812,
      "mean_token_accuracy": 0.19810927659273148,
      "num_tokens": 44820597.0,
      "step": 22120
    },
    {
      "entropy": 5.951891899108887,
      "epoch": 1.9334964607183431,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004625717407816572,
      "loss": 5.3339,
      "mean_token_accuracy": 0.19846695065498351,
      "num_tokens": 44832570.0,
      "step": 22125
    },
    {
      "entropy": 5.931836175918579,
      "epoch": 1.9339334090710478,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004625545290257733,
      "loss": 5.3713,
      "mean_token_accuracy": 0.2020449459552765,
      "num_tokens": 44841936.0,
      "step": 22130
    },
    {
      "entropy": 5.871368360519409,
      "epoch": 1.9343703574237525,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046253731367249527,
      "loss": 5.3278,
      "mean_token_accuracy": 0.2023551806807518,
      "num_tokens": 44852329.0,
      "step": 22135
    },
    {
      "entropy": 5.956622695922851,
      "epoch": 1.9348073057764572,
      "grad_norm": 1.125,
      "learning_rate": 0.0004625200947221532,
      "loss": 5.3413,
      "mean_token_accuracy": 0.19340042173862457,
      "num_tokens": 44862183.0,
      "step": 22140
    },
    {
      "entropy": 5.937865924835205,
      "epoch": 1.935244254129162,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004625028721750776,
      "loss": 5.3591,
      "mean_token_accuracy": 0.19138141125440597,
      "num_tokens": 44872712.0,
      "step": 22145
    },
    {
      "entropy": 5.925636339187622,
      "epoch": 1.9356812024818666,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046248564603159855,
      "loss": 5.3537,
      "mean_token_accuracy": 0.1967277020215988,
      "num_tokens": 44884423.0,
      "step": 22150
    },
    {
      "entropy": 6.00656213760376,
      "epoch": 1.9361181508345715,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004624684162920466,
      "loss": 5.3427,
      "mean_token_accuracy": 0.18976776003837587,
      "num_tokens": 44894556.0,
      "step": 22155
    },
    {
      "entropy": 5.907860946655274,
      "epoch": 1.936555099187276,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046245118295675213,
      "loss": 5.2661,
      "mean_token_accuracy": 0.1998444303870201,
      "num_tokens": 44905208.0,
      "step": 22160
    },
    {
      "entropy": 5.901018047332764,
      "epoch": 1.9369920475399809,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046243394602604585,
      "loss": 5.2897,
      "mean_token_accuracy": 0.1971872478723526,
      "num_tokens": 44915674.0,
      "step": 22165
    },
    {
      "entropy": 5.93257474899292,
      "epoch": 1.9374289958926854,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004624167055002582,
      "loss": 5.3358,
      "mean_token_accuracy": 0.19576922804117203,
      "num_tokens": 44926382.0,
      "step": 22170
    },
    {
      "entropy": 5.972134971618653,
      "epoch": 1.9378659442453903,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046239946137971997,
      "loss": 5.2805,
      "mean_token_accuracy": 0.20526351183652877,
      "num_tokens": 44935963.0,
      "step": 22175
    },
    {
      "entropy": 6.002081251144409,
      "epoch": 1.9383028925980947,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046238221366476183,
      "loss": 5.3943,
      "mean_token_accuracy": 0.18827416747808456,
      "num_tokens": 44946312.0,
      "step": 22180
    },
    {
      "entropy": 5.859976434707642,
      "epoch": 1.9387398409507997,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046236496235571475,
      "loss": 5.299,
      "mean_token_accuracy": 0.19964483380317688,
      "num_tokens": 44956213.0,
      "step": 22185
    },
    {
      "entropy": 5.979175138473511,
      "epoch": 1.9391767893035043,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004623477074529094,
      "loss": 5.4553,
      "mean_token_accuracy": 0.1794587790966034,
      "num_tokens": 44966575.0,
      "step": 22190
    },
    {
      "entropy": 6.018356704711914,
      "epoch": 1.939613737656209,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004623304489566769,
      "loss": 5.3624,
      "mean_token_accuracy": 0.19672729074954987,
      "num_tokens": 44976353.0,
      "step": 22195
    },
    {
      "entropy": 5.933996200561523,
      "epoch": 1.9400506860089137,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046231318686734826,
      "loss": 5.3051,
      "mean_token_accuracy": 0.19607460200786592,
      "num_tokens": 44987046.0,
      "step": 22200
    },
    {
      "entropy": 6.007632875442505,
      "epoch": 1.9404876343616184,
      "grad_norm": 1.125,
      "learning_rate": 0.0004622959211852545,
      "loss": 5.2664,
      "mean_token_accuracy": 0.19968221336603165,
      "num_tokens": 44996606.0,
      "step": 22205
    },
    {
      "entropy": 6.007046890258789,
      "epoch": 1.9409245827143233,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046227865191072684,
      "loss": 5.3815,
      "mean_token_accuracy": 0.19524853825569152,
      "num_tokens": 45006895.0,
      "step": 22210
    },
    {
      "entropy": 5.899101495742798,
      "epoch": 1.9413615310670278,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004622613790440965,
      "loss": 5.2733,
      "mean_token_accuracy": 0.20103184431791304,
      "num_tokens": 45017075.0,
      "step": 22215
    },
    {
      "entropy": 5.972812509536743,
      "epoch": 1.9417984794197327,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004622441025856948,
      "loss": 5.3197,
      "mean_token_accuracy": 0.19744050800800322,
      "num_tokens": 45027331.0,
      "step": 22220
    },
    {
      "entropy": 5.950911045074463,
      "epoch": 1.9422354277724372,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004622268225358531,
      "loss": 5.3131,
      "mean_token_accuracy": 0.18940070867538453,
      "num_tokens": 45037623.0,
      "step": 22225
    },
    {
      "entropy": 5.982494211196899,
      "epoch": 1.942672376125142,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046220953889490287,
      "loss": 5.3842,
      "mean_token_accuracy": 0.19039728343486786,
      "num_tokens": 45048191.0,
      "step": 22230
    },
    {
      "entropy": 5.995823955535888,
      "epoch": 1.9431093244778466,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046219225166317554,
      "loss": 5.3704,
      "mean_token_accuracy": 0.19119347482919694,
      "num_tokens": 45057906.0,
      "step": 22235
    },
    {
      "entropy": 5.950352525711059,
      "epoch": 1.9435462728305515,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046217496084100267,
      "loss": 5.3682,
      "mean_token_accuracy": 0.20848988741636276,
      "num_tokens": 45067842.0,
      "step": 22240
    },
    {
      "entropy": 5.961729526519775,
      "epoch": 1.9439832211832562,
      "grad_norm": 1.125,
      "learning_rate": 0.00046215766642871607,
      "loss": 5.3037,
      "mean_token_accuracy": 0.19456334710121154,
      "num_tokens": 45077388.0,
      "step": 22245
    },
    {
      "entropy": 5.858703851699829,
      "epoch": 1.9444201695359609,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046214036842664736,
      "loss": 5.2194,
      "mean_token_accuracy": 0.20310035049915315,
      "num_tokens": 45086001.0,
      "step": 22250
    },
    {
      "entropy": 5.916921377182007,
      "epoch": 1.9448571178886656,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004621230668351283,
      "loss": 5.2854,
      "mean_token_accuracy": 0.20058358907699586,
      "num_tokens": 45096203.0,
      "step": 22255
    },
    {
      "entropy": 5.926350212097168,
      "epoch": 1.9452940662413702,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046210576165449076,
      "loss": 5.2756,
      "mean_token_accuracy": 0.2062096044421196,
      "num_tokens": 45104997.0,
      "step": 22260
    },
    {
      "entropy": 5.967508745193482,
      "epoch": 1.945731014594075,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046208845288506667,
      "loss": 5.3208,
      "mean_token_accuracy": 0.19998329877853394,
      "num_tokens": 45114150.0,
      "step": 22265
    },
    {
      "entropy": 5.9130446434021,
      "epoch": 1.9461679629467796,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000462071140527188,
      "loss": 5.3295,
      "mean_token_accuracy": 0.2091475784778595,
      "num_tokens": 45124668.0,
      "step": 22270
    },
    {
      "entropy": 5.930591440200805,
      "epoch": 1.9466049112994845,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004620538245811869,
      "loss": 5.336,
      "mean_token_accuracy": 0.19515450149774552,
      "num_tokens": 45135164.0,
      "step": 22275
    },
    {
      "entropy": 5.9092964172363285,
      "epoch": 1.947041859652189,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046203650504739543,
      "loss": 5.1944,
      "mean_token_accuracy": 0.2114340752363205,
      "num_tokens": 45144453.0,
      "step": 22280
    },
    {
      "entropy": 5.94138536453247,
      "epoch": 1.947478808004894,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046201918192614573,
      "loss": 5.3543,
      "mean_token_accuracy": 0.19234440773725509,
      "num_tokens": 45154557.0,
      "step": 22285
    },
    {
      "entropy": 5.955989694595337,
      "epoch": 1.9479157563575984,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004620018552177701,
      "loss": 5.3194,
      "mean_token_accuracy": 0.20048916786909105,
      "num_tokens": 45164938.0,
      "step": 22290
    },
    {
      "entropy": 6.02447476387024,
      "epoch": 1.9483527047103033,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046198452492260096,
      "loss": 5.448,
      "mean_token_accuracy": 0.1830485552549362,
      "num_tokens": 45175823.0,
      "step": 22295
    },
    {
      "entropy": 5.89483118057251,
      "epoch": 1.948789653063008,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004619671910409706,
      "loss": 5.253,
      "mean_token_accuracy": 0.21318984031677246,
      "num_tokens": 45185589.0,
      "step": 22300
    },
    {
      "entropy": 5.928168296813965,
      "epoch": 1.9492266014157127,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004619498535732116,
      "loss": 5.3887,
      "mean_token_accuracy": 0.19839892238378526,
      "num_tokens": 45195654.0,
      "step": 22305
    },
    {
      "entropy": 6.000731039047241,
      "epoch": 1.9496635497684174,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046193251251965646,
      "loss": 5.4094,
      "mean_token_accuracy": 0.1899242252111435,
      "num_tokens": 45205824.0,
      "step": 22310
    },
    {
      "entropy": 5.986055707931518,
      "epoch": 1.950100498121122,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004619151678806377,
      "loss": 5.3487,
      "mean_token_accuracy": 0.20105663388967515,
      "num_tokens": 45215228.0,
      "step": 22315
    },
    {
      "entropy": 5.920574522018432,
      "epoch": 1.9505374464738268,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004618978196564882,
      "loss": 5.242,
      "mean_token_accuracy": 0.20173185169696808,
      "num_tokens": 45225043.0,
      "step": 22320
    },
    {
      "entropy": 5.922919988632202,
      "epoch": 1.9509743948265315,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004618804678475405,
      "loss": 5.3697,
      "mean_token_accuracy": 0.19076576828956604,
      "num_tokens": 45234837.0,
      "step": 22325
    },
    {
      "entropy": 5.955141496658325,
      "epoch": 1.9514113431792364,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046186311245412746,
      "loss": 5.2437,
      "mean_token_accuracy": 0.20007688403129578,
      "num_tokens": 45243540.0,
      "step": 22330
    },
    {
      "entropy": 5.938255643844604,
      "epoch": 1.9518482915319408,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004618457534765821,
      "loss": 5.2928,
      "mean_token_accuracy": 0.19837885200977326,
      "num_tokens": 45253053.0,
      "step": 22335
    },
    {
      "entropy": 5.854796266555786,
      "epoch": 1.9522852398846458,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004618283909152372,
      "loss": 5.2614,
      "mean_token_accuracy": 0.19869113266468047,
      "num_tokens": 45262973.0,
      "step": 22340
    },
    {
      "entropy": 5.898161554336548,
      "epoch": 1.9527221882373502,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004618110247704259,
      "loss": 5.2587,
      "mean_token_accuracy": 0.2060949683189392,
      "num_tokens": 45272951.0,
      "step": 22345
    },
    {
      "entropy": 6.0011045932769775,
      "epoch": 1.9531591365900551,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046179365504248125,
      "loss": 5.3757,
      "mean_token_accuracy": 0.19728566706180573,
      "num_tokens": 45283596.0,
      "step": 22350
    },
    {
      "entropy": 5.92164797782898,
      "epoch": 1.9535960849427596,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046177628173173647,
      "loss": 5.1938,
      "mean_token_accuracy": 0.20823382139205932,
      "num_tokens": 45294712.0,
      "step": 22355
    },
    {
      "entropy": 5.900326108932495,
      "epoch": 1.9540330332954645,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004617589048385247,
      "loss": 5.32,
      "mean_token_accuracy": 0.19442375749349594,
      "num_tokens": 45305681.0,
      "step": 22360
    },
    {
      "entropy": 5.944669723510742,
      "epoch": 1.9544699816481692,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004617415243631793,
      "loss": 5.2872,
      "mean_token_accuracy": 0.2040884777903557,
      "num_tokens": 45315332.0,
      "step": 22365
    },
    {
      "entropy": 5.840245246887207,
      "epoch": 1.954906930000874,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046172414030603356,
      "loss": 5.2579,
      "mean_token_accuracy": 0.20047329515218734,
      "num_tokens": 45324526.0,
      "step": 22370
    },
    {
      "entropy": 5.933843183517456,
      "epoch": 1.9553438783535786,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000461706752667421,
      "loss": 5.3102,
      "mean_token_accuracy": 0.1986495226621628,
      "num_tokens": 45333989.0,
      "step": 22375
    },
    {
      "entropy": 5.961101388931274,
      "epoch": 1.9557808267062833,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004616893614476751,
      "loss": 5.2717,
      "mean_token_accuracy": 0.1967935487627983,
      "num_tokens": 45344545.0,
      "step": 22380
    },
    {
      "entropy": 5.904003429412842,
      "epoch": 1.956217775058988,
      "grad_norm": 1.125,
      "learning_rate": 0.00046167196664712945,
      "loss": 5.3168,
      "mean_token_accuracy": 0.18780230432748796,
      "num_tokens": 45354125.0,
      "step": 22385
    },
    {
      "entropy": 5.965499830245972,
      "epoch": 1.9566547234116927,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046165456826611763,
      "loss": 5.3818,
      "mean_token_accuracy": 0.18991815000772477,
      "num_tokens": 45364577.0,
      "step": 22390
    },
    {
      "entropy": 5.9928879737854,
      "epoch": 1.9570916717643976,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046163716630497336,
      "loss": 5.3439,
      "mean_token_accuracy": 0.2036022126674652,
      "num_tokens": 45374655.0,
      "step": 22395
    },
    {
      "entropy": 5.968493175506592,
      "epoch": 1.957528620117102,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046161976076403047,
      "loss": 5.3303,
      "mean_token_accuracy": 0.19273240864276886,
      "num_tokens": 45384383.0,
      "step": 22400
    },
    {
      "entropy": 5.945876598358154,
      "epoch": 1.957965568469807,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004616023516436228,
      "loss": 5.3203,
      "mean_token_accuracy": 0.19804483354091645,
      "num_tokens": 45394645.0,
      "step": 22405
    },
    {
      "entropy": 5.91227912902832,
      "epoch": 1.9584025168225114,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046158493894408423,
      "loss": 5.2519,
      "mean_token_accuracy": 0.20853750705718993,
      "num_tokens": 45404661.0,
      "step": 22410
    },
    {
      "entropy": 5.982299375534057,
      "epoch": 1.9588394651752163,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004615675226657487,
      "loss": 5.4017,
      "mean_token_accuracy": 0.19422069042921067,
      "num_tokens": 45413915.0,
      "step": 22415
    },
    {
      "entropy": 6.005562973022461,
      "epoch": 1.959276413527921,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004615501028089504,
      "loss": 5.3022,
      "mean_token_accuracy": 0.19417103081941606,
      "num_tokens": 45424019.0,
      "step": 22420
    },
    {
      "entropy": 5.9471910953521725,
      "epoch": 1.9597133618806257,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004615326793740233,
      "loss": 5.2891,
      "mean_token_accuracy": 0.19591305106878282,
      "num_tokens": 45435087.0,
      "step": 22425
    },
    {
      "entropy": 5.968887710571289,
      "epoch": 1.9601503102333304,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046151525236130174,
      "loss": 5.3878,
      "mean_token_accuracy": 0.19168675392866136,
      "num_tokens": 45446363.0,
      "step": 22430
    },
    {
      "entropy": 5.962321710586548,
      "epoch": 1.960587258586035,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004614978217711199,
      "loss": 5.3431,
      "mean_token_accuracy": 0.19515056014060975,
      "num_tokens": 45456268.0,
      "step": 22435
    },
    {
      "entropy": 5.9155768871307375,
      "epoch": 1.9610242069387398,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046148038760381217,
      "loss": 5.2868,
      "mean_token_accuracy": 0.20327192693948745,
      "num_tokens": 45466335.0,
      "step": 22440
    },
    {
      "entropy": 5.96085753440857,
      "epoch": 1.9614611552914445,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004614629498597128,
      "loss": 5.3158,
      "mean_token_accuracy": 0.19278111904859543,
      "num_tokens": 45476128.0,
      "step": 22445
    },
    {
      "entropy": 5.929940938949585,
      "epoch": 1.9618981036441494,
      "grad_norm": 1.125,
      "learning_rate": 0.0004614455085391564,
      "loss": 5.2652,
      "mean_token_accuracy": 0.19832374453544616,
      "num_tokens": 45485986.0,
      "step": 22450
    },
    {
      "entropy": 5.9200445175170895,
      "epoch": 1.9623350519968539,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004614280636424774,
      "loss": 5.3162,
      "mean_token_accuracy": 0.1955774739384651,
      "num_tokens": 45497270.0,
      "step": 22455
    },
    {
      "entropy": 5.908200311660766,
      "epoch": 1.9627720003495588,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004614106151700105,
      "loss": 5.3263,
      "mean_token_accuracy": 0.20173178017139434,
      "num_tokens": 45507415.0,
      "step": 22460
    },
    {
      "entropy": 5.918690395355225,
      "epoch": 1.9632089487022633,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004613931631220903,
      "loss": 5.3403,
      "mean_token_accuracy": 0.1939642071723938,
      "num_tokens": 45518025.0,
      "step": 22465
    },
    {
      "entropy": 5.963577222824097,
      "epoch": 1.9636458970549682,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046137570749905157,
      "loss": 5.3684,
      "mean_token_accuracy": 0.19614943861961365,
      "num_tokens": 45528834.0,
      "step": 22470
    },
    {
      "entropy": 6.000086975097656,
      "epoch": 1.9640828454076729,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004613582483012291,
      "loss": 5.2951,
      "mean_token_accuracy": 0.2006361335515976,
      "num_tokens": 45538543.0,
      "step": 22475
    },
    {
      "entropy": 5.968358135223388,
      "epoch": 1.9645197937603776,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004613407855289578,
      "loss": 5.3948,
      "mean_token_accuracy": 0.19184963256120682,
      "num_tokens": 45549298.0,
      "step": 22480
    },
    {
      "entropy": 5.957469320297241,
      "epoch": 1.9649567421130822,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004613233191825726,
      "loss": 5.3549,
      "mean_token_accuracy": 0.19683887511491777,
      "num_tokens": 45560064.0,
      "step": 22485
    },
    {
      "entropy": 5.972312879562378,
      "epoch": 1.965393690465787,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004613058492624084,
      "loss": 5.2283,
      "mean_token_accuracy": 0.20504531264305115,
      "num_tokens": 45569780.0,
      "step": 22490
    },
    {
      "entropy": 5.866374349594116,
      "epoch": 1.9658306388184916,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004612883757688005,
      "loss": 5.2429,
      "mean_token_accuracy": 0.20507175475358963,
      "num_tokens": 45580256.0,
      "step": 22495
    },
    {
      "entropy": 5.936431741714477,
      "epoch": 1.9662675871711963,
      "grad_norm": 1.125,
      "learning_rate": 0.00046127089870208395,
      "loss": 5.3343,
      "mean_token_accuracy": 0.19529329985380173,
      "num_tokens": 45590615.0,
      "step": 22500
    },
    {
      "entropy": 5.9344476699829105,
      "epoch": 1.9667045355239012,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004612534180625939,
      "loss": 5.2718,
      "mean_token_accuracy": 0.19677692800760269,
      "num_tokens": 45600215.0,
      "step": 22505
    },
    {
      "entropy": 5.896133613586426,
      "epoch": 1.9671414838766057,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046123593385066574,
      "loss": 5.3135,
      "mean_token_accuracy": 0.2031473323702812,
      "num_tokens": 45610484.0,
      "step": 22510
    },
    {
      "entropy": 5.947034454345703,
      "epoch": 1.9675784322293106,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046121844606663475,
      "loss": 5.342,
      "mean_token_accuracy": 0.19280937761068345,
      "num_tokens": 45620528.0,
      "step": 22515
    },
    {
      "entropy": 5.975707244873047,
      "epoch": 1.968015380582015,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004612009547108364,
      "loss": 5.2758,
      "mean_token_accuracy": 0.19899215698242187,
      "num_tokens": 45629985.0,
      "step": 22520
    },
    {
      "entropy": 5.968478775024414,
      "epoch": 1.96845232893472,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046118345978360615,
      "loss": 5.3515,
      "mean_token_accuracy": 0.19020472168922425,
      "num_tokens": 45639750.0,
      "step": 22525
    },
    {
      "entropy": 5.877640724182129,
      "epoch": 1.9688892772874245,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004611659612852796,
      "loss": 5.2928,
      "mean_token_accuracy": 0.20082616060972214,
      "num_tokens": 45650779.0,
      "step": 22530
    },
    {
      "entropy": 5.945648813247681,
      "epoch": 1.9693262256401294,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046114845921619234,
      "loss": 5.3035,
      "mean_token_accuracy": 0.19554663449525833,
      "num_tokens": 45660363.0,
      "step": 22535
    },
    {
      "entropy": 5.930953931808472,
      "epoch": 1.969763173992834,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004611309535766801,
      "loss": 5.3594,
      "mean_token_accuracy": 0.2020055055618286,
      "num_tokens": 45670474.0,
      "step": 22540
    },
    {
      "entropy": 5.879658937454224,
      "epoch": 1.9702001223455388,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004611134443670787,
      "loss": 5.2392,
      "mean_token_accuracy": 0.20460392534732819,
      "num_tokens": 45680157.0,
      "step": 22545
    },
    {
      "entropy": 5.9363562107086185,
      "epoch": 1.9706370706982435,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004610959315877239,
      "loss": 5.3124,
      "mean_token_accuracy": 0.2029518723487854,
      "num_tokens": 45689340.0,
      "step": 22550
    },
    {
      "entropy": 5.985838937759399,
      "epoch": 1.9710740190509481,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046107841523895153,
      "loss": 5.3333,
      "mean_token_accuracy": 0.19430961608886718,
      "num_tokens": 45699453.0,
      "step": 22555
    },
    {
      "entropy": 5.892444133758545,
      "epoch": 1.9715109674036528,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046106089532109776,
      "loss": 5.3053,
      "mean_token_accuracy": 0.19827359914779663,
      "num_tokens": 45711513.0,
      "step": 22560
    },
    {
      "entropy": 5.931056642532349,
      "epoch": 1.9719479157563575,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004610433718344985,
      "loss": 5.3788,
      "mean_token_accuracy": 0.191931913793087,
      "num_tokens": 45722939.0,
      "step": 22565
    },
    {
      "entropy": 5.943284177780152,
      "epoch": 1.9723848641090624,
      "grad_norm": 2.15625,
      "learning_rate": 0.00046102584477948985,
      "loss": 5.3267,
      "mean_token_accuracy": 0.2013324409723282,
      "num_tokens": 45733099.0,
      "step": 22570
    },
    {
      "entropy": 5.926343679428101,
      "epoch": 1.972821812461767,
      "grad_norm": 1.125,
      "learning_rate": 0.0004610083141564079,
      "loss": 5.3708,
      "mean_token_accuracy": 0.20267719626426697,
      "num_tokens": 45743161.0,
      "step": 22575
    },
    {
      "entropy": 5.994105529785156,
      "epoch": 1.9732587608144718,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046099077996558923,
      "loss": 5.4191,
      "mean_token_accuracy": 0.18801861107349396,
      "num_tokens": 45752903.0,
      "step": 22580
    },
    {
      "entropy": 5.91042857170105,
      "epoch": 1.9736957091671763,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046097324220736993,
      "loss": 5.2427,
      "mean_token_accuracy": 0.2028243735432625,
      "num_tokens": 45763360.0,
      "step": 22585
    },
    {
      "entropy": 5.869699573516845,
      "epoch": 1.9741326575198812,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004609557008820863,
      "loss": 5.1969,
      "mean_token_accuracy": 0.21416477262973785,
      "num_tokens": 45773025.0,
      "step": 22590
    },
    {
      "entropy": 5.862930870056152,
      "epoch": 1.974569605872586,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004609381559900749,
      "loss": 5.2241,
      "mean_token_accuracy": 0.20709581524133683,
      "num_tokens": 45782844.0,
      "step": 22595
    },
    {
      "entropy": 5.8945270538330075,
      "epoch": 1.9750065542252906,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046092060753167224,
      "loss": 5.3394,
      "mean_token_accuracy": 0.19691142737865447,
      "num_tokens": 45793058.0,
      "step": 22600
    },
    {
      "entropy": 5.992781448364258,
      "epoch": 1.9754435025779953,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004609030555072149,
      "loss": 5.3225,
      "mean_token_accuracy": 0.19635048508644104,
      "num_tokens": 45802865.0,
      "step": 22605
    },
    {
      "entropy": 5.917921257019043,
      "epoch": 1.9758804509307,
      "grad_norm": 1.125,
      "learning_rate": 0.00046088549991703953,
      "loss": 5.2697,
      "mean_token_accuracy": 0.20597121864557266,
      "num_tokens": 45812981.0,
      "step": 22610
    },
    {
      "entropy": 5.912788009643554,
      "epoch": 1.9763173992834047,
      "grad_norm": 1.078125,
      "learning_rate": 0.000460867940761483,
      "loss": 5.2288,
      "mean_token_accuracy": 0.20473844856023787,
      "num_tokens": 45822477.0,
      "step": 22615
    },
    {
      "entropy": 5.920732879638672,
      "epoch": 1.9767543476361094,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004608503780408818,
      "loss": 5.3995,
      "mean_token_accuracy": 0.1890745222568512,
      "num_tokens": 45833642.0,
      "step": 22620
    },
    {
      "entropy": 5.974919176101684,
      "epoch": 1.9771912959888143,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004608328117555731,
      "loss": 5.312,
      "mean_token_accuracy": 0.19831318855285646,
      "num_tokens": 45843079.0,
      "step": 22625
    },
    {
      "entropy": 5.971398448944091,
      "epoch": 1.9776282443415187,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004608152419058936,
      "loss": 5.3376,
      "mean_token_accuracy": 0.19696791023015975,
      "num_tokens": 45852987.0,
      "step": 22630
    },
    {
      "entropy": 5.935401201248169,
      "epoch": 1.9780651926942237,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046079766849218043,
      "loss": 5.3138,
      "mean_token_accuracy": 0.19433944970369338,
      "num_tokens": 45863148.0,
      "step": 22635
    },
    {
      "entropy": 5.88784008026123,
      "epoch": 1.9785021410469281,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046078009151477063,
      "loss": 5.3061,
      "mean_token_accuracy": 0.20150874853134154,
      "num_tokens": 45873430.0,
      "step": 22640
    },
    {
      "entropy": 6.0152300834655765,
      "epoch": 1.978939089399633,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046076251097400126,
      "loss": 5.3169,
      "mean_token_accuracy": 0.19624081999063492,
      "num_tokens": 45883169.0,
      "step": 22645
    },
    {
      "entropy": 6.001855564117432,
      "epoch": 1.9793760377523377,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046074492687020967,
      "loss": 5.3543,
      "mean_token_accuracy": 0.1950790524482727,
      "num_tokens": 45893948.0,
      "step": 22650
    },
    {
      "entropy": 5.852096605300903,
      "epoch": 1.9798129861050424,
      "grad_norm": 1.0,
      "learning_rate": 0.000460727339203733,
      "loss": 5.3428,
      "mean_token_accuracy": 0.20142484903335572,
      "num_tokens": 45904887.0,
      "step": 22655
    },
    {
      "entropy": 5.971080541610718,
      "epoch": 1.980249934457747,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046070974797490875,
      "loss": 5.3302,
      "mean_token_accuracy": 0.1900048166513443,
      "num_tokens": 45914845.0,
      "step": 22660
    },
    {
      "entropy": 5.979868316650391,
      "epoch": 1.9806868828104518,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004606921531840741,
      "loss": 5.3282,
      "mean_token_accuracy": 0.20405498296022415,
      "num_tokens": 45925104.0,
      "step": 22665
    },
    {
      "entropy": 5.925825595855713,
      "epoch": 1.9811238311631565,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004606745548315667,
      "loss": 5.2613,
      "mean_token_accuracy": 0.19624462872743606,
      "num_tokens": 45934934.0,
      "step": 22670
    },
    {
      "entropy": 5.900884103775025,
      "epoch": 1.9815607795158612,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004606569529177241,
      "loss": 5.2928,
      "mean_token_accuracy": 0.20851506143808365,
      "num_tokens": 45945449.0,
      "step": 22675
    },
    {
      "entropy": 5.9703507900238035,
      "epoch": 1.981997727868566,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004606393474428838,
      "loss": 5.3585,
      "mean_token_accuracy": 0.18884068578481675,
      "num_tokens": 45954897.0,
      "step": 22680
    },
    {
      "entropy": 5.963646650314331,
      "epoch": 1.9824346762212706,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046062173840738356,
      "loss": 5.2753,
      "mean_token_accuracy": 0.20681801736354827,
      "num_tokens": 45965085.0,
      "step": 22685
    },
    {
      "entropy": 5.906875896453857,
      "epoch": 1.9828716245739755,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004606041258115611,
      "loss": 5.3321,
      "mean_token_accuracy": 0.20582203567028046,
      "num_tokens": 45974610.0,
      "step": 22690
    },
    {
      "entropy": 5.943181800842285,
      "epoch": 1.98330857292668,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004605865096557542,
      "loss": 5.2908,
      "mean_token_accuracy": 0.1957750767469406,
      "num_tokens": 45984922.0,
      "step": 22695
    },
    {
      "entropy": 5.926219463348389,
      "epoch": 1.9837455212793849,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004605688899403009,
      "loss": 5.2871,
      "mean_token_accuracy": 0.19992489665746688,
      "num_tokens": 45994591.0,
      "step": 22700
    },
    {
      "entropy": 5.962165689468383,
      "epoch": 1.9841824696320893,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046055126666553906,
      "loss": 5.4013,
      "mean_token_accuracy": 0.18926476389169694,
      "num_tokens": 46005291.0,
      "step": 22705
    },
    {
      "entropy": 5.92244439125061,
      "epoch": 1.9846194179847942,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046053363983180657,
      "loss": 5.2827,
      "mean_token_accuracy": 0.19765469431877136,
      "num_tokens": 46014970.0,
      "step": 22710
    },
    {
      "entropy": 5.982669258117676,
      "epoch": 1.985056366337499,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046051600943944175,
      "loss": 5.3699,
      "mean_token_accuracy": 0.19717931300401687,
      "num_tokens": 46025042.0,
      "step": 22715
    },
    {
      "entropy": 6.003895378112793,
      "epoch": 1.9854933146902036,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046049837548878264,
      "loss": 5.3985,
      "mean_token_accuracy": 0.18773545175790787,
      "num_tokens": 46035440.0,
      "step": 22720
    },
    {
      "entropy": 5.904344129562378,
      "epoch": 1.9859302630429083,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046048073798016746,
      "loss": 5.3165,
      "mean_token_accuracy": 0.20608727186918258,
      "num_tokens": 46045344.0,
      "step": 22725
    },
    {
      "entropy": 5.973990345001221,
      "epoch": 1.986367211395613,
      "grad_norm": 1.125,
      "learning_rate": 0.00046046309691393455,
      "loss": 5.3905,
      "mean_token_accuracy": 0.18932429552078248,
      "num_tokens": 46055054.0,
      "step": 22730
    },
    {
      "entropy": 6.021656608581543,
      "epoch": 1.9868041597483177,
      "grad_norm": 1.125,
      "learning_rate": 0.00046044545229042226,
      "loss": 5.3373,
      "mean_token_accuracy": 0.19156975001096727,
      "num_tokens": 46063835.0,
      "step": 22735
    },
    {
      "entropy": 5.971038341522217,
      "epoch": 1.9872411081010224,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046042780410996906,
      "loss": 5.3638,
      "mean_token_accuracy": 0.19346692860126496,
      "num_tokens": 46074256.0,
      "step": 22740
    },
    {
      "entropy": 6.0093122005462645,
      "epoch": 1.9876780564537273,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046041015237291346,
      "loss": 5.4258,
      "mean_token_accuracy": 0.19663484692573546,
      "num_tokens": 46085563.0,
      "step": 22745
    },
    {
      "entropy": 5.978322982788086,
      "epoch": 1.9881150048064318,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046039249707959386,
      "loss": 5.3319,
      "mean_token_accuracy": 0.20178573578596115,
      "num_tokens": 46094956.0,
      "step": 22750
    },
    {
      "entropy": 5.9421131134033205,
      "epoch": 1.9885519531591367,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046037483823034906,
      "loss": 5.336,
      "mean_token_accuracy": 0.20240222066640853,
      "num_tokens": 46105292.0,
      "step": 22755
    },
    {
      "entropy": 5.98019700050354,
      "epoch": 1.9889889015118412,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004603571758255177,
      "loss": 5.497,
      "mean_token_accuracy": 0.1827549919486046,
      "num_tokens": 46117254.0,
      "step": 22760
    },
    {
      "entropy": 5.978189611434937,
      "epoch": 1.989425849864546,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046033950986543865,
      "loss": 5.384,
      "mean_token_accuracy": 0.1936711147427559,
      "num_tokens": 46127846.0,
      "step": 22765
    },
    {
      "entropy": 5.925633192062378,
      "epoch": 1.9898627982172508,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046032184035045066,
      "loss": 5.3909,
      "mean_token_accuracy": 0.19387353360652923,
      "num_tokens": 46138476.0,
      "step": 22770
    },
    {
      "entropy": 5.935365390777588,
      "epoch": 1.9902997465699555,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046030416728089267,
      "loss": 5.2783,
      "mean_token_accuracy": 0.20018245428800582,
      "num_tokens": 46147789.0,
      "step": 22775
    },
    {
      "entropy": 5.99284291267395,
      "epoch": 1.9907366949226601,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046028649065710367,
      "loss": 5.3904,
      "mean_token_accuracy": 0.19452288150787353,
      "num_tokens": 46158484.0,
      "step": 22780
    },
    {
      "entropy": 5.995391798019409,
      "epoch": 1.9911736432753648,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046026881047942273,
      "loss": 5.355,
      "mean_token_accuracy": 0.1962140366435051,
      "num_tokens": 46168062.0,
      "step": 22785
    },
    {
      "entropy": 5.980566072463989,
      "epoch": 1.9916105916280695,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046025112674818883,
      "loss": 5.3085,
      "mean_token_accuracy": 0.1977388456463814,
      "num_tokens": 46178710.0,
      "step": 22790
    },
    {
      "entropy": 5.937132453918457,
      "epoch": 1.9920475399807742,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046023343946374127,
      "loss": 5.3626,
      "mean_token_accuracy": 0.18529942482709885,
      "num_tokens": 46188728.0,
      "step": 22795
    },
    {
      "entropy": 5.908357572555542,
      "epoch": 1.9924844883334791,
      "grad_norm": 1.0,
      "learning_rate": 0.00046021574862641925,
      "loss": 5.3307,
      "mean_token_accuracy": 0.20100367516279222,
      "num_tokens": 46199716.0,
      "step": 22800
    },
    {
      "entropy": 6.016906785964966,
      "epoch": 1.9929214366861836,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046019805423656217,
      "loss": 5.3978,
      "mean_token_accuracy": 0.19356553107500077,
      "num_tokens": 46210885.0,
      "step": 22805
    },
    {
      "entropy": 5.924517011642456,
      "epoch": 1.9933583850388885,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004601803562945094,
      "loss": 5.3339,
      "mean_token_accuracy": 0.20170504450798035,
      "num_tokens": 46220879.0,
      "step": 22810
    },
    {
      "entropy": 5.911938428878784,
      "epoch": 1.993795333391593,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004601626548006003,
      "loss": 5.3674,
      "mean_token_accuracy": 0.19193432182073594,
      "num_tokens": 46230420.0,
      "step": 22815
    },
    {
      "entropy": 5.9071142196655275,
      "epoch": 1.994232281744298,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004601449497551745,
      "loss": 5.3054,
      "mean_token_accuracy": 0.1992228850722313,
      "num_tokens": 46240548.0,
      "step": 22820
    },
    {
      "entropy": 5.976111650466919,
      "epoch": 1.9946692300970026,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046012724115857153,
      "loss": 5.3086,
      "mean_token_accuracy": 0.19886040836572647,
      "num_tokens": 46250484.0,
      "step": 22825
    },
    {
      "entropy": 6.0041931629180905,
      "epoch": 1.9951061784497073,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046010952901113107,
      "loss": 5.3707,
      "mean_token_accuracy": 0.1903765767812729,
      "num_tokens": 46261184.0,
      "step": 22830
    },
    {
      "entropy": 5.888191032409668,
      "epoch": 1.995543126802412,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004600918133131928,
      "loss": 5.3048,
      "mean_token_accuracy": 0.20787732750177385,
      "num_tokens": 46271301.0,
      "step": 22835
    },
    {
      "entropy": 5.934223747253418,
      "epoch": 1.9959800751551167,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046007409406509663,
      "loss": 5.3623,
      "mean_token_accuracy": 0.18873073905706406,
      "num_tokens": 46281201.0,
      "step": 22840
    },
    {
      "entropy": 5.996232986450195,
      "epoch": 1.9964170235078214,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004600563712671823,
      "loss": 5.3754,
      "mean_token_accuracy": 0.1965030923485756,
      "num_tokens": 46291969.0,
      "step": 22845
    },
    {
      "entropy": 5.850360536575318,
      "epoch": 1.996853971860526,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046003864491978984,
      "loss": 5.1659,
      "mean_token_accuracy": 0.213406765460968,
      "num_tokens": 46301308.0,
      "step": 22850
    },
    {
      "entropy": 5.938668441772461,
      "epoch": 1.997290920213231,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046002091502325926,
      "loss": 5.3274,
      "mean_token_accuracy": 0.19772329032421113,
      "num_tokens": 46311564.0,
      "step": 22855
    },
    {
      "entropy": 5.9407013893127445,
      "epoch": 1.9977278685659354,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004600031815779306,
      "loss": 5.2873,
      "mean_token_accuracy": 0.20775675773620605,
      "num_tokens": 46321843.0,
      "step": 22860
    },
    {
      "entropy": 5.901514673233033,
      "epoch": 1.9981648169186403,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045998544458414385,
      "loss": 5.2701,
      "mean_token_accuracy": 0.19860581904649735,
      "num_tokens": 46331274.0,
      "step": 22865
    },
    {
      "entropy": 5.991129541397095,
      "epoch": 1.9986017652713448,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045996770404223937,
      "loss": 5.3933,
      "mean_token_accuracy": 0.19243981689214706,
      "num_tokens": 46341385.0,
      "step": 22870
    },
    {
      "entropy": 5.935572004318237,
      "epoch": 1.9990387136240497,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045994995995255746,
      "loss": 5.338,
      "mean_token_accuracy": 0.1991019919514656,
      "num_tokens": 46350385.0,
      "step": 22875
    },
    {
      "entropy": 5.871907472610474,
      "epoch": 1.9994756619767542,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045993221231543837,
      "loss": 5.227,
      "mean_token_accuracy": 0.20150377750396728,
      "num_tokens": 46361111.0,
      "step": 22880
    },
    {
      "entropy": 5.939625692367554,
      "epoch": 1.999912610329459,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004599144611312226,
      "loss": 5.383,
      "mean_token_accuracy": 0.19066282659769057,
      "num_tokens": 46371316.0,
      "step": 22885
    },
    {
      "entropy": 5.96906681060791,
      "epoch": 2.0003495586821636,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045989670640025047,
      "loss": 5.3079,
      "mean_token_accuracy": 0.19670072495937346,
      "num_tokens": 46381151.0,
      "step": 22890
    },
    {
      "entropy": 5.980037450790405,
      "epoch": 2.0007865070348685,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045987894812286276,
      "loss": 5.2501,
      "mean_token_accuracy": 0.20123582035303117,
      "num_tokens": 46391420.0,
      "step": 22895
    },
    {
      "entropy": 5.912394189834595,
      "epoch": 2.0012234553875734,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045986118629939984,
      "loss": 5.2293,
      "mean_token_accuracy": 0.20134335458278657,
      "num_tokens": 46401575.0,
      "step": 22900
    },
    {
      "entropy": 5.9070374965667725,
      "epoch": 2.001660403740278,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004598434209302025,
      "loss": 5.2189,
      "mean_token_accuracy": 0.20587948709726334,
      "num_tokens": 46411275.0,
      "step": 22905
    },
    {
      "entropy": 5.937626552581787,
      "epoch": 2.002097352092983,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004598256520156116,
      "loss": 5.2765,
      "mean_token_accuracy": 0.20534908175468444,
      "num_tokens": 46421179.0,
      "step": 22910
    },
    {
      "entropy": 5.945215082168579,
      "epoch": 2.0025343004456873,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004598078795559678,
      "loss": 5.308,
      "mean_token_accuracy": 0.2014886260032654,
      "num_tokens": 46431865.0,
      "step": 22915
    },
    {
      "entropy": 5.979660892486573,
      "epoch": 2.002971248798392,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045979010355161196,
      "loss": 5.3794,
      "mean_token_accuracy": 0.1911380410194397,
      "num_tokens": 46443398.0,
      "step": 22920
    },
    {
      "entropy": 6.028286790847778,
      "epoch": 2.0034081971510966,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004597723240028852,
      "loss": 5.3272,
      "mean_token_accuracy": 0.20097130835056304,
      "num_tokens": 46454818.0,
      "step": 22925
    },
    {
      "entropy": 5.989093160629272,
      "epoch": 2.0038451455038016,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004597545409101283,
      "loss": 5.3153,
      "mean_token_accuracy": 0.19843115508556367,
      "num_tokens": 46465231.0,
      "step": 22930
    },
    {
      "entropy": 5.970626449584961,
      "epoch": 2.004282093856506,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004597367542736826,
      "loss": 5.2934,
      "mean_token_accuracy": 0.20383795499801635,
      "num_tokens": 46476323.0,
      "step": 22935
    },
    {
      "entropy": 5.927985095977784,
      "epoch": 2.004719042209211,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004597189640938892,
      "loss": 5.2603,
      "mean_token_accuracy": 0.19947341829538345,
      "num_tokens": 46486573.0,
      "step": 22940
    },
    {
      "entropy": 5.890193223953247,
      "epoch": 2.0051559905619154,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004597011703710892,
      "loss": 5.24,
      "mean_token_accuracy": 0.20356834679841995,
      "num_tokens": 46497779.0,
      "step": 22945
    },
    {
      "entropy": 5.919165992736817,
      "epoch": 2.0055929389146203,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000459683373105624,
      "loss": 5.3042,
      "mean_token_accuracy": 0.1962435930967331,
      "num_tokens": 46509074.0,
      "step": 22950
    },
    {
      "entropy": 5.903332853317261,
      "epoch": 2.006029887267325,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045966557229783493,
      "loss": 5.1411,
      "mean_token_accuracy": 0.21184581965208055,
      "num_tokens": 46519292.0,
      "step": 22955
    },
    {
      "entropy": 5.879557418823242,
      "epoch": 2.0064668356200297,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004596477679480634,
      "loss": 5.2331,
      "mean_token_accuracy": 0.20712139904499055,
      "num_tokens": 46529689.0,
      "step": 22960
    },
    {
      "entropy": 5.950014591217041,
      "epoch": 2.0069037839727346,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045962996005665096,
      "loss": 5.2898,
      "mean_token_accuracy": 0.19847131371498108,
      "num_tokens": 46538616.0,
      "step": 22965
    },
    {
      "entropy": 5.916062450408935,
      "epoch": 2.007340732325439,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045961214862393914,
      "loss": 5.1617,
      "mean_token_accuracy": 0.20842178612947465,
      "num_tokens": 46548372.0,
      "step": 22970
    },
    {
      "entropy": 5.9323203563690186,
      "epoch": 2.007777680678144,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004595943336502695,
      "loss": 5.2866,
      "mean_token_accuracy": 0.19028770327568054,
      "num_tokens": 46558991.0,
      "step": 22975
    },
    {
      "entropy": 5.9218589782714846,
      "epoch": 2.0082146290308485,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004595765151359839,
      "loss": 5.2441,
      "mean_token_accuracy": 0.20590828061103822,
      "num_tokens": 46569531.0,
      "step": 22980
    },
    {
      "entropy": 6.0035206317901615,
      "epoch": 2.0086515773835534,
      "grad_norm": 1.15625,
      "learning_rate": 0.000459558693081424,
      "loss": 5.2765,
      "mean_token_accuracy": 0.2044375330209732,
      "num_tokens": 46579774.0,
      "step": 22985
    },
    {
      "entropy": 5.974398565292359,
      "epoch": 2.009088525736258,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004595408674869317,
      "loss": 5.2414,
      "mean_token_accuracy": 0.20069456994533538,
      "num_tokens": 46589810.0,
      "step": 22990
    },
    {
      "entropy": 5.876377105712891,
      "epoch": 2.0095254740889628,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004595230383528488,
      "loss": 5.2248,
      "mean_token_accuracy": 0.20657685548067092,
      "num_tokens": 46599347.0,
      "step": 22995
    },
    {
      "entropy": 6.025344896316528,
      "epoch": 2.0099624224416672,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045950520567951734,
      "loss": 5.3108,
      "mean_token_accuracy": 0.20313772410154343,
      "num_tokens": 46609926.0,
      "step": 23000
    },
    {
      "entropy": 5.935063600540161,
      "epoch": 2.010399370794372,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004594873694672794,
      "loss": 5.2359,
      "mean_token_accuracy": 0.20633210241794586,
      "num_tokens": 46620736.0,
      "step": 23005
    },
    {
      "entropy": 5.9493035793304445,
      "epoch": 2.0108363191470766,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045946952971647696,
      "loss": 5.2343,
      "mean_token_accuracy": 0.1981365218758583,
      "num_tokens": 46630713.0,
      "step": 23010
    },
    {
      "entropy": 5.894041585922241,
      "epoch": 2.0112732674997815,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004594516864274524,
      "loss": 5.2603,
      "mean_token_accuracy": 0.20303348004817962,
      "num_tokens": 46641197.0,
      "step": 23015
    },
    {
      "entropy": 5.837397480010987,
      "epoch": 2.0117102158524864,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045943383960054775,
      "loss": 5.1119,
      "mean_token_accuracy": 0.2276800513267517,
      "num_tokens": 46651325.0,
      "step": 23020
    },
    {
      "entropy": 5.898046350479126,
      "epoch": 2.012147164205191,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045941598923610543,
      "loss": 5.273,
      "mean_token_accuracy": 0.20232945680618286,
      "num_tokens": 46662100.0,
      "step": 23025
    },
    {
      "entropy": 5.975554513931274,
      "epoch": 2.012584112557896,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045939813533446783,
      "loss": 5.2613,
      "mean_token_accuracy": 0.20753877013921737,
      "num_tokens": 46671460.0,
      "step": 23030
    },
    {
      "entropy": 5.995179843902588,
      "epoch": 2.0130210609106003,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004593802778959773,
      "loss": 5.3156,
      "mean_token_accuracy": 0.19482629597187043,
      "num_tokens": 46681835.0,
      "step": 23035
    },
    {
      "entropy": 5.855102777481079,
      "epoch": 2.013458009263305,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004593624169209765,
      "loss": 5.2376,
      "mean_token_accuracy": 0.20548096001148225,
      "num_tokens": 46691976.0,
      "step": 23040
    },
    {
      "entropy": 5.886093521118164,
      "epoch": 2.0138949576160097,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045934455240980793,
      "loss": 5.2236,
      "mean_token_accuracy": 0.208451147377491,
      "num_tokens": 46702589.0,
      "step": 23045
    },
    {
      "entropy": 5.983085870742798,
      "epoch": 2.0143319059687146,
      "grad_norm": 1.25,
      "learning_rate": 0.0004593266843628143,
      "loss": 5.3407,
      "mean_token_accuracy": 0.1956626668572426,
      "num_tokens": 46712167.0,
      "step": 23050
    },
    {
      "entropy": 5.933544683456421,
      "epoch": 2.014768854321419,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004593088127803382,
      "loss": 5.2615,
      "mean_token_accuracy": 0.20265561044216157,
      "num_tokens": 46722684.0,
      "step": 23055
    },
    {
      "entropy": 5.925379896163941,
      "epoch": 2.015205802674124,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045929093766272253,
      "loss": 5.2671,
      "mean_token_accuracy": 0.20387496799230576,
      "num_tokens": 46733856.0,
      "step": 23060
    },
    {
      "entropy": 6.0015819549560545,
      "epoch": 2.0156427510268284,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004592730590103101,
      "loss": 5.3399,
      "mean_token_accuracy": 0.19630374610424042,
      "num_tokens": 46744210.0,
      "step": 23065
    },
    {
      "entropy": 6.019638204574585,
      "epoch": 2.0160796993795334,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004592551768234438,
      "loss": 5.226,
      "mean_token_accuracy": 0.2043958619236946,
      "num_tokens": 46754611.0,
      "step": 23070
    },
    {
      "entropy": 5.994345426559448,
      "epoch": 2.0165166477322383,
      "grad_norm": 1.125,
      "learning_rate": 0.0004592372911024667,
      "loss": 5.3361,
      "mean_token_accuracy": 0.1932745799422264,
      "num_tokens": 46765647.0,
      "step": 23075
    },
    {
      "entropy": 5.877986049652099,
      "epoch": 2.0169535960849427,
      "grad_norm": 1.125,
      "learning_rate": 0.00045921940184772184,
      "loss": 5.3109,
      "mean_token_accuracy": 0.19731749445199967,
      "num_tokens": 46776052.0,
      "step": 23080
    },
    {
      "entropy": 5.94441499710083,
      "epoch": 2.0173905444376476,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004592015090595523,
      "loss": 5.2561,
      "mean_token_accuracy": 0.2017272189259529,
      "num_tokens": 46786255.0,
      "step": 23085
    },
    {
      "entropy": 5.93231315612793,
      "epoch": 2.017827492790352,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004591836127383013,
      "loss": 5.2737,
      "mean_token_accuracy": 0.20341991484165192,
      "num_tokens": 46795958.0,
      "step": 23090
    },
    {
      "entropy": 5.911931276321411,
      "epoch": 2.018264441143057,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045916571288431206,
      "loss": 5.2798,
      "mean_token_accuracy": 0.1959069028496742,
      "num_tokens": 46806508.0,
      "step": 23095
    },
    {
      "entropy": 5.955900192260742,
      "epoch": 2.0187013894957615,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045914780949792797,
      "loss": 5.34,
      "mean_token_accuracy": 0.1953381985425949,
      "num_tokens": 46816868.0,
      "step": 23100
    },
    {
      "entropy": 5.9968102931976315,
      "epoch": 2.0191383378484664,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004591299025794924,
      "loss": 5.2626,
      "mean_token_accuracy": 0.20435596853494645,
      "num_tokens": 46827591.0,
      "step": 23105
    },
    {
      "entropy": 5.911914539337158,
      "epoch": 2.019575286201171,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004591119921293488,
      "loss": 5.22,
      "mean_token_accuracy": 0.20777339041233062,
      "num_tokens": 46837394.0,
      "step": 23110
    },
    {
      "entropy": 5.909527683258057,
      "epoch": 2.020012234553876,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045909407814784074,
      "loss": 5.2289,
      "mean_token_accuracy": 0.2071230098605156,
      "num_tokens": 46847881.0,
      "step": 23115
    },
    {
      "entropy": 5.928555583953857,
      "epoch": 2.0204491829065803,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045907616063531185,
      "loss": 5.2579,
      "mean_token_accuracy": 0.19881992638111115,
      "num_tokens": 46857728.0,
      "step": 23120
    },
    {
      "entropy": 6.006346559524536,
      "epoch": 2.020886131259285,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004590582395921057,
      "loss": 5.3292,
      "mean_token_accuracy": 0.1984352648258209,
      "num_tokens": 46866991.0,
      "step": 23125
    },
    {
      "entropy": 5.997034645080566,
      "epoch": 2.0213230796119896,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000459040315018566,
      "loss": 5.3499,
      "mean_token_accuracy": 0.19674808531999588,
      "num_tokens": 46876253.0,
      "step": 23130
    },
    {
      "entropy": 5.9598307609558105,
      "epoch": 2.0217600279646946,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045902238691503675,
      "loss": 5.2482,
      "mean_token_accuracy": 0.20039243549108504,
      "num_tokens": 46887309.0,
      "step": 23135
    },
    {
      "entropy": 5.965746879577637,
      "epoch": 2.0221969763173995,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045900445528186163,
      "loss": 5.321,
      "mean_token_accuracy": 0.20226101726293563,
      "num_tokens": 46896722.0,
      "step": 23140
    },
    {
      "entropy": 5.908303594589233,
      "epoch": 2.022633924670104,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045898652011938456,
      "loss": 5.2964,
      "mean_token_accuracy": 0.19833996891975403,
      "num_tokens": 46907519.0,
      "step": 23145
    },
    {
      "entropy": 5.927731513977051,
      "epoch": 2.023070873022809,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045896858142794976,
      "loss": 5.2429,
      "mean_token_accuracy": 0.200887131690979,
      "num_tokens": 46917387.0,
      "step": 23150
    },
    {
      "entropy": 5.980031299591064,
      "epoch": 2.0235078213755133,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004589506392079011,
      "loss": 5.2626,
      "mean_token_accuracy": 0.20704084634780884,
      "num_tokens": 46926477.0,
      "step": 23155
    },
    {
      "entropy": 5.997620010375977,
      "epoch": 2.0239447697282182,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045893269345958284,
      "loss": 5.2632,
      "mean_token_accuracy": 0.20402860790491104,
      "num_tokens": 46935807.0,
      "step": 23160
    },
    {
      "entropy": 5.894531917572022,
      "epoch": 2.0243817180809227,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004589147441833391,
      "loss": 5.2688,
      "mean_token_accuracy": 0.20136562883853912,
      "num_tokens": 46946843.0,
      "step": 23165
    },
    {
      "entropy": 5.87233452796936,
      "epoch": 2.0248186664336276,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004588967913795142,
      "loss": 5.2346,
      "mean_token_accuracy": 0.20659712702035904,
      "num_tokens": 46956289.0,
      "step": 23170
    },
    {
      "entropy": 5.9911205768585205,
      "epoch": 2.025255614786332,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045887883504845244,
      "loss": 5.2572,
      "mean_token_accuracy": 0.20373642295598984,
      "num_tokens": 46965606.0,
      "step": 23175
    },
    {
      "entropy": 5.930970239639282,
      "epoch": 2.025692563139037,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045886087519049837,
      "loss": 5.1961,
      "mean_token_accuracy": 0.20820754319429396,
      "num_tokens": 46974769.0,
      "step": 23180
    },
    {
      "entropy": 5.9625935554504395,
      "epoch": 2.0261295114917415,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045884291180599626,
      "loss": 5.2947,
      "mean_token_accuracy": 0.19292015731334686,
      "num_tokens": 46984753.0,
      "step": 23185
    },
    {
      "entropy": 5.892002201080322,
      "epoch": 2.0265664598444464,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045882494489529085,
      "loss": 5.2342,
      "mean_token_accuracy": 0.20209449380636216,
      "num_tokens": 46995403.0,
      "step": 23190
    },
    {
      "entropy": 5.981659984588623,
      "epoch": 2.0270034081971513,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004588069744587267,
      "loss": 5.2605,
      "mean_token_accuracy": 0.20321214944124222,
      "num_tokens": 47005054.0,
      "step": 23195
    },
    {
      "entropy": 5.917983150482177,
      "epoch": 2.0274403565498558,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004587890004966484,
      "loss": 5.2833,
      "mean_token_accuracy": 0.19978789687156678,
      "num_tokens": 47016315.0,
      "step": 23200
    },
    {
      "entropy": 5.949563312530517,
      "epoch": 2.0278773049025607,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045877102300940077,
      "loss": 5.2033,
      "mean_token_accuracy": 0.19868722856044768,
      "num_tokens": 47025981.0,
      "step": 23205
    },
    {
      "entropy": 5.93126425743103,
      "epoch": 2.028314253255265,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045875304199732863,
      "loss": 5.2532,
      "mean_token_accuracy": 0.2060578390955925,
      "num_tokens": 47035906.0,
      "step": 23210
    },
    {
      "entropy": 5.952390432357788,
      "epoch": 2.02875120160797,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045873505746077685,
      "loss": 5.2774,
      "mean_token_accuracy": 0.1928601309657097,
      "num_tokens": 47045884.0,
      "step": 23215
    },
    {
      "entropy": 5.966390228271484,
      "epoch": 2.0291881499606745,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004587170694000904,
      "loss": 5.2978,
      "mean_token_accuracy": 0.1994227334856987,
      "num_tokens": 47055575.0,
      "step": 23220
    },
    {
      "entropy": 5.904334545135498,
      "epoch": 2.0296250983133794,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004586990778156144,
      "loss": 5.2757,
      "mean_token_accuracy": 0.1978834867477417,
      "num_tokens": 47065484.0,
      "step": 23225
    },
    {
      "entropy": 5.9667808532714846,
      "epoch": 2.030062046666084,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045868108270769373,
      "loss": 5.228,
      "mean_token_accuracy": 0.20485471338033676,
      "num_tokens": 47076604.0,
      "step": 23230
    },
    {
      "entropy": 5.94096360206604,
      "epoch": 2.030498995018789,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004586630840766736,
      "loss": 5.2447,
      "mean_token_accuracy": 0.20742659866809846,
      "num_tokens": 47086452.0,
      "step": 23235
    },
    {
      "entropy": 5.881670522689819,
      "epoch": 2.0309359433714933,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045864508192289933,
      "loss": 5.3327,
      "mean_token_accuracy": 0.19521231055259705,
      "num_tokens": 47096425.0,
      "step": 23240
    },
    {
      "entropy": 5.924077272415161,
      "epoch": 2.031372891724198,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045862707624671616,
      "loss": 5.2778,
      "mean_token_accuracy": 0.19905679672956467,
      "num_tokens": 47107214.0,
      "step": 23245
    },
    {
      "entropy": 5.9651721000671385,
      "epoch": 2.031809840076903,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045860906704846944,
      "loss": 5.2833,
      "mean_token_accuracy": 0.1999203532934189,
      "num_tokens": 47117498.0,
      "step": 23250
    },
    {
      "entropy": 5.9095789909362795,
      "epoch": 2.0322467884296076,
      "grad_norm": 1.125,
      "learning_rate": 0.00045859105432850464,
      "loss": 5.2119,
      "mean_token_accuracy": 0.20549867749214173,
      "num_tokens": 47128031.0,
      "step": 23255
    },
    {
      "entropy": 5.973772048950195,
      "epoch": 2.0326837367823125,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004585730380871672,
      "loss": 5.2727,
      "mean_token_accuracy": 0.20344144999980926,
      "num_tokens": 47138347.0,
      "step": 23260
    },
    {
      "entropy": 5.891203355789185,
      "epoch": 2.033120685135017,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004585550183248027,
      "loss": 5.1618,
      "mean_token_accuracy": 0.2096719741821289,
      "num_tokens": 47147642.0,
      "step": 23265
    },
    {
      "entropy": 5.944439125061035,
      "epoch": 2.033557633487722,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045853699504175673,
      "loss": 5.2745,
      "mean_token_accuracy": 0.2041103407740593,
      "num_tokens": 47156855.0,
      "step": 23270
    },
    {
      "entropy": 5.966957712173462,
      "epoch": 2.0339945818404264,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045851896823837505,
      "loss": 5.2613,
      "mean_token_accuracy": 0.2051738128066063,
      "num_tokens": 47167030.0,
      "step": 23275
    },
    {
      "entropy": 5.945193958282471,
      "epoch": 2.0344315301931313,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045850093791500334,
      "loss": 5.2673,
      "mean_token_accuracy": 0.19962079972028732,
      "num_tokens": 47177471.0,
      "step": 23280
    },
    {
      "entropy": 5.964451551437378,
      "epoch": 2.0348684785458357,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045848290407198754,
      "loss": 5.3343,
      "mean_token_accuracy": 0.19921109974384307,
      "num_tokens": 47187911.0,
      "step": 23285
    },
    {
      "entropy": 5.9371977806091305,
      "epoch": 2.0353054268985407,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004584648667096736,
      "loss": 5.2252,
      "mean_token_accuracy": 0.20457464456558228,
      "num_tokens": 47198622.0,
      "step": 23290
    },
    {
      "entropy": 5.951628303527832,
      "epoch": 2.035742375251245,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004584468258284072,
      "loss": 5.3462,
      "mean_token_accuracy": 0.20051078051328658,
      "num_tokens": 47207593.0,
      "step": 23295
    },
    {
      "entropy": 5.980239486694336,
      "epoch": 2.03617932360395,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004584287814285347,
      "loss": 5.2872,
      "mean_token_accuracy": 0.1971050038933754,
      "num_tokens": 47218734.0,
      "step": 23300
    },
    {
      "entropy": 5.931703090667725,
      "epoch": 2.0366162719566545,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045841073351040196,
      "loss": 5.2903,
      "mean_token_accuracy": 0.19783456325531007,
      "num_tokens": 47228662.0,
      "step": 23305
    },
    {
      "entropy": 6.022043371200562,
      "epoch": 2.0370532203093594,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004583926820743553,
      "loss": 5.4044,
      "mean_token_accuracy": 0.19456077218055726,
      "num_tokens": 47238460.0,
      "step": 23310
    },
    {
      "entropy": 6.006273984909058,
      "epoch": 2.0374901686620643,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004583746271207409,
      "loss": 5.2945,
      "mean_token_accuracy": 0.19074565172195435,
      "num_tokens": 47247789.0,
      "step": 23315
    },
    {
      "entropy": 5.885800457000732,
      "epoch": 2.037927117014769,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045835656864990504,
      "loss": 5.3089,
      "mean_token_accuracy": 0.19545126855373382,
      "num_tokens": 47257901.0,
      "step": 23320
    },
    {
      "entropy": 5.946895790100098,
      "epoch": 2.0383640653674737,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004583385066621941,
      "loss": 5.2711,
      "mean_token_accuracy": 0.19926340728998185,
      "num_tokens": 47268362.0,
      "step": 23325
    },
    {
      "entropy": 5.94845175743103,
      "epoch": 2.038801013720178,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045832044115795464,
      "loss": 5.2704,
      "mean_token_accuracy": 0.20415778309106827,
      "num_tokens": 47278297.0,
      "step": 23330
    },
    {
      "entropy": 6.000526666641235,
      "epoch": 2.039237962072883,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045830237213753287,
      "loss": 5.2691,
      "mean_token_accuracy": 0.19455857276916505,
      "num_tokens": 47288795.0,
      "step": 23335
    },
    {
      "entropy": 5.928297472000122,
      "epoch": 2.0396749104255876,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004582842996012757,
      "loss": 5.2395,
      "mean_token_accuracy": 0.20251589864492417,
      "num_tokens": 47298373.0,
      "step": 23340
    },
    {
      "entropy": 5.912635946273804,
      "epoch": 2.0401118587782925,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045826622354952956,
      "loss": 5.3343,
      "mean_token_accuracy": 0.19835052788257598,
      "num_tokens": 47308160.0,
      "step": 23345
    },
    {
      "entropy": 5.982073068618774,
      "epoch": 2.040548807130997,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045824814398264126,
      "loss": 5.2641,
      "mean_token_accuracy": 0.2064944699406624,
      "num_tokens": 47317626.0,
      "step": 23350
    },
    {
      "entropy": 5.995806121826172,
      "epoch": 2.040985755483702,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045823006090095757,
      "loss": 5.3446,
      "mean_token_accuracy": 0.19676295667886734,
      "num_tokens": 47327221.0,
      "step": 23355
    },
    {
      "entropy": 5.979172611236573,
      "epoch": 2.0414227038364063,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004582119743048253,
      "loss": 5.3027,
      "mean_token_accuracy": 0.2039645120501518,
      "num_tokens": 47337385.0,
      "step": 23360
    },
    {
      "entropy": 5.938565015792847,
      "epoch": 2.0418596521891113,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004581938841945912,
      "loss": 5.2542,
      "mean_token_accuracy": 0.19752957820892333,
      "num_tokens": 47346894.0,
      "step": 23365
    },
    {
      "entropy": 5.943238353729248,
      "epoch": 2.042296600541816,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045817579057060254,
      "loss": 5.2911,
      "mean_token_accuracy": 0.20253728032112123,
      "num_tokens": 47356192.0,
      "step": 23370
    },
    {
      "entropy": 5.923222637176513,
      "epoch": 2.0427335488945206,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004581576934332063,
      "loss": 5.2181,
      "mean_token_accuracy": 0.20263292640447617,
      "num_tokens": 47366704.0,
      "step": 23375
    },
    {
      "entropy": 5.946924448013306,
      "epoch": 2.0431704972472255,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004581395927827494,
      "loss": 5.225,
      "mean_token_accuracy": 0.20506410002708436,
      "num_tokens": 47377068.0,
      "step": 23380
    },
    {
      "entropy": 5.89668345451355,
      "epoch": 2.04360744559993,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045812148861957923,
      "loss": 5.2765,
      "mean_token_accuracy": 0.20575304478406906,
      "num_tokens": 47387268.0,
      "step": 23385
    },
    {
      "entropy": 5.928488683700562,
      "epoch": 2.044044393952635,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004581033809440429,
      "loss": 5.2049,
      "mean_token_accuracy": 0.2044230118393898,
      "num_tokens": 47397508.0,
      "step": 23390
    },
    {
      "entropy": 5.896690368652344,
      "epoch": 2.0444813423053394,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004580852697564878,
      "loss": 5.2793,
      "mean_token_accuracy": 0.1965611308813095,
      "num_tokens": 47407871.0,
      "step": 23395
    },
    {
      "entropy": 5.9610066413879395,
      "epoch": 2.0449182906580443,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045806715505726135,
      "loss": 5.3712,
      "mean_token_accuracy": 0.1878541111946106,
      "num_tokens": 47417089.0,
      "step": 23400
    },
    {
      "entropy": 5.981172180175781,
      "epoch": 2.045355239010749,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045804903684671083,
      "loss": 5.2513,
      "mean_token_accuracy": 0.20112090557813644,
      "num_tokens": 47426746.0,
      "step": 23405
    },
    {
      "entropy": 5.99417200088501,
      "epoch": 2.0457921873634537,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045803091512518397,
      "loss": 5.3008,
      "mean_token_accuracy": 0.197466941177845,
      "num_tokens": 47437873.0,
      "step": 23410
    },
    {
      "entropy": 5.887692785263061,
      "epoch": 2.046229135716158,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045801278989302815,
      "loss": 5.2365,
      "mean_token_accuracy": 0.20404179096221925,
      "num_tokens": 47448520.0,
      "step": 23415
    },
    {
      "entropy": 5.937870168685913,
      "epoch": 2.046666084068863,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004579946611505912,
      "loss": 5.2884,
      "mean_token_accuracy": 0.19406418055295943,
      "num_tokens": 47458323.0,
      "step": 23420
    },
    {
      "entropy": 6.038191127777099,
      "epoch": 2.0471030324215675,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004579765288982207,
      "loss": 5.3723,
      "mean_token_accuracy": 0.19452205598354338,
      "num_tokens": 47468061.0,
      "step": 23425
    },
    {
      "entropy": 5.99058723449707,
      "epoch": 2.0475399807742725,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045795839313626455,
      "loss": 5.3035,
      "mean_token_accuracy": 0.20223161876201629,
      "num_tokens": 47477541.0,
      "step": 23430
    },
    {
      "entropy": 5.889900970458984,
      "epoch": 2.0479769291269774,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004579402538650705,
      "loss": 5.2303,
      "mean_token_accuracy": 0.2077564164996147,
      "num_tokens": 47488570.0,
      "step": 23435
    },
    {
      "entropy": 5.9471372127532955,
      "epoch": 2.048413877479682,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004579221110849866,
      "loss": 5.2673,
      "mean_token_accuracy": 0.19678311944007873,
      "num_tokens": 47498386.0,
      "step": 23440
    },
    {
      "entropy": 5.9431548595428465,
      "epoch": 2.0488508258323868,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004579039647963607,
      "loss": 5.1762,
      "mean_token_accuracy": 0.2038038745522499,
      "num_tokens": 47507292.0,
      "step": 23445
    },
    {
      "entropy": 5.965850591659546,
      "epoch": 2.0492877741850912,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045788581499954094,
      "loss": 5.2664,
      "mean_token_accuracy": 0.19667864441871644,
      "num_tokens": 47516971.0,
      "step": 23450
    },
    {
      "entropy": 5.861307001113891,
      "epoch": 2.049724722537796,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045786766169487536,
      "loss": 5.1623,
      "mean_token_accuracy": 0.2172908291220665,
      "num_tokens": 47527233.0,
      "step": 23455
    },
    {
      "entropy": 5.949730014801025,
      "epoch": 2.0501616708905006,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045784950488271227,
      "loss": 5.2332,
      "mean_token_accuracy": 0.20019748508930207,
      "num_tokens": 47537542.0,
      "step": 23460
    },
    {
      "entropy": 5.942405128479004,
      "epoch": 2.0505986192432055,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045783134456339985,
      "loss": 5.2303,
      "mean_token_accuracy": 0.20627484023571013,
      "num_tokens": 47546613.0,
      "step": 23465
    },
    {
      "entropy": 5.947311305999756,
      "epoch": 2.05103556759591,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045781318073728647,
      "loss": 5.2364,
      "mean_token_accuracy": 0.20394036024808884,
      "num_tokens": 47555462.0,
      "step": 23470
    },
    {
      "entropy": 5.978495073318482,
      "epoch": 2.051472515948615,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004577950134047204,
      "loss": 5.22,
      "mean_token_accuracy": 0.20241628289222718,
      "num_tokens": 47565429.0,
      "step": 23475
    },
    {
      "entropy": 5.998245477676392,
      "epoch": 2.0519094643013194,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004577768425660503,
      "loss": 5.376,
      "mean_token_accuracy": 0.19090922623872758,
      "num_tokens": 47576021.0,
      "step": 23480
    },
    {
      "entropy": 5.990873718261719,
      "epoch": 2.0523464126540243,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004577586682216245,
      "loss": 5.3031,
      "mean_token_accuracy": 0.19760984182357788,
      "num_tokens": 47586016.0,
      "step": 23485
    },
    {
      "entropy": 5.957401943206787,
      "epoch": 2.052783361006729,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004577404903717917,
      "loss": 5.272,
      "mean_token_accuracy": 0.19785157293081285,
      "num_tokens": 47595673.0,
      "step": 23490
    },
    {
      "entropy": 5.894034814834595,
      "epoch": 2.0532203093594337,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004577223090169006,
      "loss": 5.2237,
      "mean_token_accuracy": 0.20835136324167253,
      "num_tokens": 47605252.0,
      "step": 23495
    },
    {
      "entropy": 5.909328985214233,
      "epoch": 2.0536572577121386,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004577041241572998,
      "loss": 5.2516,
      "mean_token_accuracy": 0.20077119618654252,
      "num_tokens": 47614726.0,
      "step": 23500
    },
    {
      "entropy": 5.878271150588989,
      "epoch": 2.054094206064843,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045768593579333816,
      "loss": 5.2666,
      "mean_token_accuracy": 0.20857479125261308,
      "num_tokens": 47625618.0,
      "step": 23505
    },
    {
      "entropy": 5.966154098510742,
      "epoch": 2.054531154417548,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004576677439253646,
      "loss": 5.3161,
      "mean_token_accuracy": 0.19871362596750258,
      "num_tokens": 47635702.0,
      "step": 23510
    },
    {
      "entropy": 5.9190826416015625,
      "epoch": 2.0549681027702524,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045764954855372796,
      "loss": 5.2954,
      "mean_token_accuracy": 0.2019194632768631,
      "num_tokens": 47647301.0,
      "step": 23515
    },
    {
      "entropy": 5.978244161605835,
      "epoch": 2.0554050511229573,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004576313496787772,
      "loss": 5.3928,
      "mean_token_accuracy": 0.19140651375055312,
      "num_tokens": 47656844.0,
      "step": 23520
    },
    {
      "entropy": 6.02663221359253,
      "epoch": 2.055841999475662,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004576131473008616,
      "loss": 5.2399,
      "mean_token_accuracy": 0.20305257588624953,
      "num_tokens": 47666196.0,
      "step": 23525
    },
    {
      "entropy": 5.946940231323242,
      "epoch": 2.0562789478283667,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004575949414203301,
      "loss": 5.2502,
      "mean_token_accuracy": 0.200346939265728,
      "num_tokens": 47675329.0,
      "step": 23530
    },
    {
      "entropy": 5.8472271919250485,
      "epoch": 2.056715896181071,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045757673203753195,
      "loss": 5.2474,
      "mean_token_accuracy": 0.20371227711439133,
      "num_tokens": 47686806.0,
      "step": 23535
    },
    {
      "entropy": 5.927862977981567,
      "epoch": 2.057152844533776,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004575585191528164,
      "loss": 5.2321,
      "mean_token_accuracy": 0.20402391403913497,
      "num_tokens": 47696664.0,
      "step": 23540
    },
    {
      "entropy": 5.974265766143799,
      "epoch": 2.057589792886481,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045754030276653275,
      "loss": 5.3061,
      "mean_token_accuracy": 0.1981194496154785,
      "num_tokens": 47707257.0,
      "step": 23545
    },
    {
      "entropy": 5.940657138824463,
      "epoch": 2.0580267412391855,
      "grad_norm": 1.125,
      "learning_rate": 0.0004575220828790305,
      "loss": 5.2091,
      "mean_token_accuracy": 0.20988474041223526,
      "num_tokens": 47716345.0,
      "step": 23550
    },
    {
      "entropy": 5.883091449737549,
      "epoch": 2.0584636895918904,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045750385949065904,
      "loss": 5.2162,
      "mean_token_accuracy": 0.20360496938228606,
      "num_tokens": 47727779.0,
      "step": 23555
    },
    {
      "entropy": 5.948261451721192,
      "epoch": 2.058900637944595,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004574856326017679,
      "loss": 5.3144,
      "mean_token_accuracy": 0.1958341881632805,
      "num_tokens": 47737591.0,
      "step": 23560
    },
    {
      "entropy": 5.96359429359436,
      "epoch": 2.0593375862973,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045746740221270664,
      "loss": 5.2568,
      "mean_token_accuracy": 0.20040559023618698,
      "num_tokens": 47746871.0,
      "step": 23565
    },
    {
      "entropy": 5.99092960357666,
      "epoch": 2.0597745346500043,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004574491683238251,
      "loss": 5.2947,
      "mean_token_accuracy": 0.19963211864233016,
      "num_tokens": 47757089.0,
      "step": 23570
    },
    {
      "entropy": 5.987091064453125,
      "epoch": 2.060211483002709,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004574309309354729,
      "loss": 5.304,
      "mean_token_accuracy": 0.19686578214168549,
      "num_tokens": 47766853.0,
      "step": 23575
    },
    {
      "entropy": 5.943925762176514,
      "epoch": 2.0606484313554136,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045741269004799984,
      "loss": 5.2496,
      "mean_token_accuracy": 0.20334497541189195,
      "num_tokens": 47776549.0,
      "step": 23580
    },
    {
      "entropy": 5.9588018417358395,
      "epoch": 2.0610853797081186,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045739444566175584,
      "loss": 5.3265,
      "mean_token_accuracy": 0.19192838221788405,
      "num_tokens": 47786641.0,
      "step": 23585
    },
    {
      "entropy": 5.968818855285645,
      "epoch": 2.061522328060823,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045737619777709077,
      "loss": 5.3332,
      "mean_token_accuracy": 0.1874460309743881,
      "num_tokens": 47796678.0,
      "step": 23590
    },
    {
      "entropy": 5.952477836608887,
      "epoch": 2.061959276413528,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004573579463943546,
      "loss": 5.2731,
      "mean_token_accuracy": 0.20104903876781463,
      "num_tokens": 47806678.0,
      "step": 23595
    },
    {
      "entropy": 5.957982206344605,
      "epoch": 2.062396224766233,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045733969151389754,
      "loss": 5.2826,
      "mean_token_accuracy": 0.19955923408269882,
      "num_tokens": 47816619.0,
      "step": 23600
    },
    {
      "entropy": 5.888417100906372,
      "epoch": 2.0628331731189373,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004573214331360696,
      "loss": 5.2628,
      "mean_token_accuracy": 0.20416468232870102,
      "num_tokens": 47825437.0,
      "step": 23605
    },
    {
      "entropy": 5.937058067321777,
      "epoch": 2.0632701214716422,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004573031712612211,
      "loss": 5.2316,
      "mean_token_accuracy": 0.2017964869737625,
      "num_tokens": 47835604.0,
      "step": 23610
    },
    {
      "entropy": 5.918467283248901,
      "epoch": 2.0637070698243467,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004572849058897022,
      "loss": 5.2065,
      "mean_token_accuracy": 0.21175480782985687,
      "num_tokens": 47845299.0,
      "step": 23615
    },
    {
      "entropy": 5.873882913589478,
      "epoch": 2.0641440181770516,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004572666370218633,
      "loss": 5.2658,
      "mean_token_accuracy": 0.20443805903196335,
      "num_tokens": 47855293.0,
      "step": 23620
    },
    {
      "entropy": 5.9579156875610355,
      "epoch": 2.064580966529756,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045724836465805483,
      "loss": 5.2949,
      "mean_token_accuracy": 0.20496754646301268,
      "num_tokens": 47864809.0,
      "step": 23625
    },
    {
      "entropy": 5.980192804336548,
      "epoch": 2.065017914882461,
      "grad_norm": 1.125,
      "learning_rate": 0.0004572300887986272,
      "loss": 5.2939,
      "mean_token_accuracy": 0.19437238126993178,
      "num_tokens": 47874032.0,
      "step": 23630
    },
    {
      "entropy": 5.957219934463501,
      "epoch": 2.0654548632351655,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045721180944393095,
      "loss": 5.2593,
      "mean_token_accuracy": 0.20472056269645691,
      "num_tokens": 47884465.0,
      "step": 23635
    },
    {
      "entropy": 5.931257486343384,
      "epoch": 2.0658918115878704,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004571935265943168,
      "loss": 5.2334,
      "mean_token_accuracy": 0.20681965351104736,
      "num_tokens": 47894030.0,
      "step": 23640
    },
    {
      "entropy": 5.976471042633056,
      "epoch": 2.066328759940575,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045717524025013526,
      "loss": 5.2847,
      "mean_token_accuracy": 0.19934817552566528,
      "num_tokens": 47904338.0,
      "step": 23645
    },
    {
      "entropy": 5.924255323410034,
      "epoch": 2.0667657082932798,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004571569504117372,
      "loss": 5.2939,
      "mean_token_accuracy": 0.1963982105255127,
      "num_tokens": 47914384.0,
      "step": 23650
    },
    {
      "entropy": 5.899369955062866,
      "epoch": 2.0672026566459842,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004571386570794734,
      "loss": 5.2087,
      "mean_token_accuracy": 0.20342467576265336,
      "num_tokens": 47924386.0,
      "step": 23655
    },
    {
      "entropy": 5.973909997940064,
      "epoch": 2.067639604998689,
      "grad_norm": 1.0,
      "learning_rate": 0.00045712036025369465,
      "loss": 5.2978,
      "mean_token_accuracy": 0.19734818190336229,
      "num_tokens": 47935379.0,
      "step": 23660
    },
    {
      "entropy": 5.939484024047852,
      "epoch": 2.068076553351394,
      "grad_norm": 1.078125,
      "learning_rate": 0.000457102059934752,
      "loss": 5.2553,
      "mean_token_accuracy": 0.20008895099163054,
      "num_tokens": 47945657.0,
      "step": 23665
    },
    {
      "entropy": 5.934815883636475,
      "epoch": 2.0685135017040985,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045708375612299646,
      "loss": 5.2361,
      "mean_token_accuracy": 0.19928756058216096,
      "num_tokens": 47956294.0,
      "step": 23670
    },
    {
      "entropy": 5.956506299972534,
      "epoch": 2.0689504500568034,
      "grad_norm": 1.125,
      "learning_rate": 0.00045706544881877895,
      "loss": 5.3806,
      "mean_token_accuracy": 0.19210445880889893,
      "num_tokens": 47967362.0,
      "step": 23675
    },
    {
      "entropy": 5.980862236022949,
      "epoch": 2.069387398409508,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004570471380224508,
      "loss": 5.2786,
      "mean_token_accuracy": 0.20381010472774505,
      "num_tokens": 47977521.0,
      "step": 23680
    },
    {
      "entropy": 5.958224868774414,
      "epoch": 2.069824346762213,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045702882373436317,
      "loss": 5.2778,
      "mean_token_accuracy": 0.20509943664073943,
      "num_tokens": 47988485.0,
      "step": 23685
    },
    {
      "entropy": 5.908764743804932,
      "epoch": 2.0702612951149173,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004570105059548673,
      "loss": 5.3042,
      "mean_token_accuracy": 0.19646456688642502,
      "num_tokens": 47998696.0,
      "step": 23690
    },
    {
      "entropy": 5.904234170913696,
      "epoch": 2.070698243467622,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004569921846843146,
      "loss": 5.1222,
      "mean_token_accuracy": 0.21523358076810836,
      "num_tokens": 48007843.0,
      "step": 23695
    },
    {
      "entropy": 5.967300796508789,
      "epoch": 2.0711351918203267,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004569738599230563,
      "loss": 5.2389,
      "mean_token_accuracy": 0.20641682893037797,
      "num_tokens": 48017579.0,
      "step": 23700
    },
    {
      "entropy": 5.9220373153686525,
      "epoch": 2.0715721401730316,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045695553167144417,
      "loss": 5.2353,
      "mean_token_accuracy": 0.20421645045280457,
      "num_tokens": 48027267.0,
      "step": 23705
    },
    {
      "entropy": 5.93422794342041,
      "epoch": 2.072009088525736,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045693719992982956,
      "loss": 5.2714,
      "mean_token_accuracy": 0.20116518139839173,
      "num_tokens": 48038064.0,
      "step": 23710
    },
    {
      "entropy": 6.009496355056763,
      "epoch": 2.072446036878441,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045691886469856406,
      "loss": 5.3601,
      "mean_token_accuracy": 0.19242578595876694,
      "num_tokens": 48048316.0,
      "step": 23715
    },
    {
      "entropy": 5.895459365844727,
      "epoch": 2.072882985231146,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004569005259779995,
      "loss": 5.325,
      "mean_token_accuracy": 0.19640444964170456,
      "num_tokens": 48059083.0,
      "step": 23720
    },
    {
      "entropy": 5.919642114639283,
      "epoch": 2.0733199335838504,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045688218376848744,
      "loss": 5.2608,
      "mean_token_accuracy": 0.19366985708475112,
      "num_tokens": 48069184.0,
      "step": 23725
    },
    {
      "entropy": 5.979678726196289,
      "epoch": 2.0737568819365553,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004568638380703799,
      "loss": 5.3079,
      "mean_token_accuracy": 0.19822632670402526,
      "num_tokens": 48078737.0,
      "step": 23730
    },
    {
      "entropy": 5.9401264667510985,
      "epoch": 2.0741938302892597,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004568454888840285,
      "loss": 5.2911,
      "mean_token_accuracy": 0.19507412612438202,
      "num_tokens": 48088581.0,
      "step": 23735
    },
    {
      "entropy": 5.807759141921997,
      "epoch": 2.0746307786419647,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004568271362097855,
      "loss": 5.1403,
      "mean_token_accuracy": 0.22507012486457825,
      "num_tokens": 48098941.0,
      "step": 23740
    },
    {
      "entropy": 5.921372747421264,
      "epoch": 2.075067726994669,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045680878004800265,
      "loss": 5.2559,
      "mean_token_accuracy": 0.20130694806575775,
      "num_tokens": 48108773.0,
      "step": 23745
    },
    {
      "entropy": 5.954433107376099,
      "epoch": 2.075504675347374,
      "grad_norm": 1.25,
      "learning_rate": 0.0004567904203990322,
      "loss": 5.3026,
      "mean_token_accuracy": 0.20555759072303773,
      "num_tokens": 48117699.0,
      "step": 23750
    },
    {
      "entropy": 5.960046625137329,
      "epoch": 2.0759416237000785,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004567720572632261,
      "loss": 5.2122,
      "mean_token_accuracy": 0.2142866149544716,
      "num_tokens": 48127693.0,
      "step": 23755
    },
    {
      "entropy": 5.968084287643433,
      "epoch": 2.0763785720527834,
      "grad_norm": 1.125,
      "learning_rate": 0.00045675369064093686,
      "loss": 5.3252,
      "mean_token_accuracy": 0.19483716785907745,
      "num_tokens": 48138073.0,
      "step": 23760
    },
    {
      "entropy": 5.916094493865967,
      "epoch": 2.076815520405488,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004567353205325165,
      "loss": 5.1897,
      "mean_token_accuracy": 0.20931023508310317,
      "num_tokens": 48147504.0,
      "step": 23765
    },
    {
      "entropy": 5.984806776046753,
      "epoch": 2.077252468758193,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004567169469383175,
      "loss": 5.2627,
      "mean_token_accuracy": 0.19875362515449524,
      "num_tokens": 48156739.0,
      "step": 23770
    },
    {
      "entropy": 5.936259651184082,
      "epoch": 2.0776894171108973,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045669856985869225,
      "loss": 5.2706,
      "mean_token_accuracy": 0.20021311342716216,
      "num_tokens": 48166883.0,
      "step": 23775
    },
    {
      "entropy": 5.883761787414551,
      "epoch": 2.078126365463602,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004566801892939932,
      "loss": 5.1928,
      "mean_token_accuracy": 0.21361553221940993,
      "num_tokens": 48177738.0,
      "step": 23780
    },
    {
      "entropy": 5.890785503387451,
      "epoch": 2.078563313816307,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004566618052445729,
      "loss": 5.2152,
      "mean_token_accuracy": 0.20357840210199357,
      "num_tokens": 48187979.0,
      "step": 23785
    },
    {
      "entropy": 5.925275802612305,
      "epoch": 2.0790002621690116,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000456643417710784,
      "loss": 5.3054,
      "mean_token_accuracy": 0.19358107447624207,
      "num_tokens": 48198579.0,
      "step": 23790
    },
    {
      "entropy": 5.986247873306274,
      "epoch": 2.0794372105217165,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045662502669297915,
      "loss": 5.3998,
      "mean_token_accuracy": 0.1889535278081894,
      "num_tokens": 48208860.0,
      "step": 23795
    },
    {
      "entropy": 5.948847913742066,
      "epoch": 2.079874158874421,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004566066321915112,
      "loss": 5.2856,
      "mean_token_accuracy": 0.20646852850914002,
      "num_tokens": 48218842.0,
      "step": 23800
    },
    {
      "entropy": 5.962653684616089,
      "epoch": 2.080311107227126,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004565882342067328,
      "loss": 5.2353,
      "mean_token_accuracy": 0.2057206854224205,
      "num_tokens": 48229082.0,
      "step": 23805
    },
    {
      "entropy": 5.960675430297852,
      "epoch": 2.0807480555798303,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045656983273899694,
      "loss": 5.3304,
      "mean_token_accuracy": 0.19721974581480026,
      "num_tokens": 48239158.0,
      "step": 23810
    },
    {
      "entropy": 5.9309892654418945,
      "epoch": 2.0811850039325352,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004565514277886565,
      "loss": 5.2669,
      "mean_token_accuracy": 0.19722588509321212,
      "num_tokens": 48249663.0,
      "step": 23815
    },
    {
      "entropy": 5.958591604232788,
      "epoch": 2.0816219522852397,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045653301935606456,
      "loss": 5.3411,
      "mean_token_accuracy": 0.19583482444286346,
      "num_tokens": 48260319.0,
      "step": 23820
    },
    {
      "entropy": 5.974063968658447,
      "epoch": 2.0820589006379446,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045651460744157424,
      "loss": 5.3109,
      "mean_token_accuracy": 0.19356546252965928,
      "num_tokens": 48270107.0,
      "step": 23825
    },
    {
      "entropy": 5.914285135269165,
      "epoch": 2.082495848990649,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004564961920455387,
      "loss": 5.2813,
      "mean_token_accuracy": 0.2030745878815651,
      "num_tokens": 48280529.0,
      "step": 23830
    },
    {
      "entropy": 6.022377920150757,
      "epoch": 2.082932797343354,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004564777731683109,
      "loss": 5.3403,
      "mean_token_accuracy": 0.1925375536084175,
      "num_tokens": 48290672.0,
      "step": 23835
    },
    {
      "entropy": 6.063842535018921,
      "epoch": 2.083369745696059,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004564593508102445,
      "loss": 5.3423,
      "mean_token_accuracy": 0.19461796581745147,
      "num_tokens": 48301327.0,
      "step": 23840
    },
    {
      "entropy": 5.978121185302735,
      "epoch": 2.0838066940487634,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004564409249716925,
      "loss": 5.3154,
      "mean_token_accuracy": 0.19421257227659225,
      "num_tokens": 48310861.0,
      "step": 23845
    },
    {
      "entropy": 5.946794271469116,
      "epoch": 2.0842436424014683,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045642249565300866,
      "loss": 5.2632,
      "mean_token_accuracy": 0.19544033706188202,
      "num_tokens": 48320381.0,
      "step": 23850
    },
    {
      "entropy": 5.9511794090271,
      "epoch": 2.084680590754173,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045640406285454616,
      "loss": 5.2225,
      "mean_token_accuracy": 0.20642928183078765,
      "num_tokens": 48330566.0,
      "step": 23855
    },
    {
      "entropy": 5.937918186187744,
      "epoch": 2.0851175391068777,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045638562657665885,
      "loss": 5.2376,
      "mean_token_accuracy": 0.2041301116347313,
      "num_tokens": 48340773.0,
      "step": 23860
    },
    {
      "entropy": 5.927593135833741,
      "epoch": 2.085554487459582,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045636718681969996,
      "loss": 5.244,
      "mean_token_accuracy": 0.20230689346790315,
      "num_tokens": 48350972.0,
      "step": 23865
    },
    {
      "entropy": 6.000710153579712,
      "epoch": 2.085991435812287,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004563487435840235,
      "loss": 5.2613,
      "mean_token_accuracy": 0.2004084274172783,
      "num_tokens": 48360606.0,
      "step": 23870
    },
    {
      "entropy": 5.964613389968872,
      "epoch": 2.0864283841649915,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045633029686998304,
      "loss": 5.3475,
      "mean_token_accuracy": 0.2009404793381691,
      "num_tokens": 48371970.0,
      "step": 23875
    },
    {
      "entropy": 5.896227073669434,
      "epoch": 2.0868653325176965,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004563118466779326,
      "loss": 5.2303,
      "mean_token_accuracy": 0.20539893954992294,
      "num_tokens": 48382917.0,
      "step": 23880
    },
    {
      "entropy": 5.9656613826751705,
      "epoch": 2.087302280870401,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045629339300822593,
      "loss": 5.2835,
      "mean_token_accuracy": 0.2062263786792755,
      "num_tokens": 48393428.0,
      "step": 23885
    },
    {
      "entropy": 6.042111158370972,
      "epoch": 2.087739229223106,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004562749358612168,
      "loss": 5.3509,
      "mean_token_accuracy": 0.19143996983766556,
      "num_tokens": 48404616.0,
      "step": 23890
    },
    {
      "entropy": 5.910466909408569,
      "epoch": 2.0881761775758108,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004562564752372596,
      "loss": 5.2025,
      "mean_token_accuracy": 0.20915460139513015,
      "num_tokens": 48414753.0,
      "step": 23895
    },
    {
      "entropy": 5.941677379608154,
      "epoch": 2.0886131259285152,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004562380111367082,
      "loss": 5.3098,
      "mean_token_accuracy": 0.19462942630052565,
      "num_tokens": 48424759.0,
      "step": 23900
    },
    {
      "entropy": 5.95444073677063,
      "epoch": 2.08905007428122,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045621954355991675,
      "loss": 5.2642,
      "mean_token_accuracy": 0.20310911834239959,
      "num_tokens": 48435158.0,
      "step": 23905
    },
    {
      "entropy": 5.943964242935181,
      "epoch": 2.0894870226339246,
      "grad_norm": 1.125,
      "learning_rate": 0.00045620107250723946,
      "loss": 5.2434,
      "mean_token_accuracy": 0.2000228613615036,
      "num_tokens": 48445234.0,
      "step": 23910
    },
    {
      "entropy": 5.942922449111938,
      "epoch": 2.0899239709866295,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045618259797903073,
      "loss": 5.2433,
      "mean_token_accuracy": 0.20557141155004502,
      "num_tokens": 48456053.0,
      "step": 23915
    },
    {
      "entropy": 5.957898664474487,
      "epoch": 2.090360919339334,
      "grad_norm": 1.125,
      "learning_rate": 0.0004561641199756447,
      "loss": 5.2476,
      "mean_token_accuracy": 0.19949128180742265,
      "num_tokens": 48466293.0,
      "step": 23920
    },
    {
      "entropy": 5.883095932006836,
      "epoch": 2.090797867692039,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045614563849743605,
      "loss": 5.2237,
      "mean_token_accuracy": 0.20561124980449677,
      "num_tokens": 48476170.0,
      "step": 23925
    },
    {
      "entropy": 5.969257879257202,
      "epoch": 2.0912348160447434,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045612715354475907,
      "loss": 5.2528,
      "mean_token_accuracy": 0.19572177827358245,
      "num_tokens": 48485728.0,
      "step": 23930
    },
    {
      "entropy": 5.959262990951538,
      "epoch": 2.0916717643974483,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045610866511796845,
      "loss": 5.2549,
      "mean_token_accuracy": 0.20214031338691713,
      "num_tokens": 48496531.0,
      "step": 23935
    },
    {
      "entropy": 5.959681510925293,
      "epoch": 2.0921087127501528,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045609017321741863,
      "loss": 5.2233,
      "mean_token_accuracy": 0.20508576780557633,
      "num_tokens": 48506006.0,
      "step": 23940
    },
    {
      "entropy": 5.916295766830444,
      "epoch": 2.0925456611028577,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004560716778434644,
      "loss": 5.2947,
      "mean_token_accuracy": 0.20092050284147261,
      "num_tokens": 48515902.0,
      "step": 23945
    },
    {
      "entropy": 5.941633844375611,
      "epoch": 2.0929826094555626,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045605317899646066,
      "loss": 5.3282,
      "mean_token_accuracy": 0.19932926446199417,
      "num_tokens": 48526263.0,
      "step": 23950
    },
    {
      "entropy": 5.934114742279053,
      "epoch": 2.093419557808267,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045603467667676195,
      "loss": 5.2554,
      "mean_token_accuracy": 0.20804356336593627,
      "num_tokens": 48537415.0,
      "step": 23955
    },
    {
      "entropy": 5.975504541397095,
      "epoch": 2.093856506160972,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004560161708847233,
      "loss": 5.274,
      "mean_token_accuracy": 0.20147465020418168,
      "num_tokens": 48548911.0,
      "step": 23960
    },
    {
      "entropy": 5.94892725944519,
      "epoch": 2.0942934545136764,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004559976616206996,
      "loss": 5.271,
      "mean_token_accuracy": 0.20418582260608673,
      "num_tokens": 48560266.0,
      "step": 23965
    },
    {
      "entropy": 6.01814866065979,
      "epoch": 2.0947304028663813,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045597914888504597,
      "loss": 5.3499,
      "mean_token_accuracy": 0.19492668509483338,
      "num_tokens": 48570696.0,
      "step": 23970
    },
    {
      "entropy": 5.960103559494018,
      "epoch": 2.095167351219086,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045596063267811746,
      "loss": 5.2566,
      "mean_token_accuracy": 0.20353310257196428,
      "num_tokens": 48580725.0,
      "step": 23975
    },
    {
      "entropy": 5.9290989398956295,
      "epoch": 2.0956042995717907,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004559421130002691,
      "loss": 5.3197,
      "mean_token_accuracy": 0.19792591482400895,
      "num_tokens": 48591968.0,
      "step": 23980
    },
    {
      "entropy": 5.904359197616577,
      "epoch": 2.096041247924495,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004559235898518563,
      "loss": 5.2513,
      "mean_token_accuracy": 0.20692430883646012,
      "num_tokens": 48602074.0,
      "step": 23985
    },
    {
      "entropy": 5.955075979232788,
      "epoch": 2.0964781962772,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004559050632332342,
      "loss": 5.2856,
      "mean_token_accuracy": 0.20189004242420197,
      "num_tokens": 48613323.0,
      "step": 23990
    },
    {
      "entropy": 6.004605436325074,
      "epoch": 2.0969151446299046,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004558865331447581,
      "loss": 5.3238,
      "mean_token_accuracy": 0.19765252918004989,
      "num_tokens": 48623386.0,
      "step": 23995
    },
    {
      "entropy": 5.913291501998901,
      "epoch": 2.0973520929826095,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004558679995867837,
      "loss": 5.2051,
      "mean_token_accuracy": 0.2102240428328514,
      "num_tokens": 48633379.0,
      "step": 24000
    },
    {
      "epoch": 2.0973520929826095,
      "eval_entropy": 5.700214812223304,
      "eval_loss": 5.338034152984619,
      "eval_mean_token_accuracy": 0.2070745120321362,
      "eval_num_tokens": 48633379.0,
      "eval_runtime": 21.4968,
      "eval_samples_per_second": 1721.836,
      "eval_steps_per_second": 215.241,
      "step": 24000
    },
    {
      "entropy": 5.936912536621094,
      "epoch": 2.097789041335314,
      "grad_norm": 1.125,
      "learning_rate": 0.0004558494625596662,
      "loss": 5.2734,
      "mean_token_accuracy": 0.20005541294813156,
      "num_tokens": 48643918.0,
      "step": 24005
    },
    {
      "entropy": 5.972338438034058,
      "epoch": 2.098225989688019,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004558309220637612,
      "loss": 5.3604,
      "mean_token_accuracy": 0.19103100150823593,
      "num_tokens": 48654534.0,
      "step": 24010
    },
    {
      "entropy": 5.953599882125855,
      "epoch": 2.098662938040724,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004558123780994244,
      "loss": 5.2585,
      "mean_token_accuracy": 0.20396452397108078,
      "num_tokens": 48664928.0,
      "step": 24015
    },
    {
      "entropy": 5.930945301055909,
      "epoch": 2.0990998863934283,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004557938306670115,
      "loss": 5.2136,
      "mean_token_accuracy": 0.20953864753246307,
      "num_tokens": 48675441.0,
      "step": 24020
    },
    {
      "entropy": 5.899013996124268,
      "epoch": 2.099536834746133,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045577527976687813,
      "loss": 5.1939,
      "mean_token_accuracy": 0.20797068774700164,
      "num_tokens": 48685173.0,
      "step": 24025
    },
    {
      "entropy": 5.9154722690582275,
      "epoch": 2.0999737830988376,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045575672539938014,
      "loss": 5.2485,
      "mean_token_accuracy": 0.20244048088788985,
      "num_tokens": 48695539.0,
      "step": 24030
    },
    {
      "entropy": 5.9332152843475345,
      "epoch": 2.1004107314515426,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045573816756487345,
      "loss": 5.279,
      "mean_token_accuracy": 0.20369115918874742,
      "num_tokens": 48705977.0,
      "step": 24035
    },
    {
      "entropy": 5.947627115249634,
      "epoch": 2.100847679804247,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004557196062637141,
      "loss": 5.2374,
      "mean_token_accuracy": 0.19789534360170363,
      "num_tokens": 48715924.0,
      "step": 24040
    },
    {
      "entropy": 5.985813999176026,
      "epoch": 2.101284628156952,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004557010414962579,
      "loss": 5.2777,
      "mean_token_accuracy": 0.19560320675373077,
      "num_tokens": 48727071.0,
      "step": 24045
    },
    {
      "entropy": 5.944340991973877,
      "epoch": 2.1017215765096564,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045568247326286103,
      "loss": 5.2656,
      "mean_token_accuracy": 0.19789893329143524,
      "num_tokens": 48736959.0,
      "step": 24050
    },
    {
      "entropy": 5.893779659271241,
      "epoch": 2.1021585248623613,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045566390156387974,
      "loss": 5.2852,
      "mean_token_accuracy": 0.2003914624452591,
      "num_tokens": 48748419.0,
      "step": 24055
    },
    {
      "entropy": 5.949753093719482,
      "epoch": 2.102595473215066,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004556453263996701,
      "loss": 5.2455,
      "mean_token_accuracy": 0.2034192904829979,
      "num_tokens": 48758480.0,
      "step": 24060
    },
    {
      "entropy": 6.021607875823975,
      "epoch": 2.1030324215677707,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045562674777058844,
      "loss": 5.368,
      "mean_token_accuracy": 0.19729757010936738,
      "num_tokens": 48768182.0,
      "step": 24065
    },
    {
      "entropy": 5.944954109191895,
      "epoch": 2.103469369920475,
      "grad_norm": 1.125,
      "learning_rate": 0.0004556081656769911,
      "loss": 5.2352,
      "mean_token_accuracy": 0.20624670386314392,
      "num_tokens": 48777729.0,
      "step": 24070
    },
    {
      "entropy": 5.947503232955933,
      "epoch": 2.10390631827318,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045558958011923456,
      "loss": 5.2501,
      "mean_token_accuracy": 0.2045554533600807,
      "num_tokens": 48787109.0,
      "step": 24075
    },
    {
      "entropy": 5.97846941947937,
      "epoch": 2.104343266625885,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004555709910976752,
      "loss": 5.2955,
      "mean_token_accuracy": 0.194531586766243,
      "num_tokens": 48797152.0,
      "step": 24080
    },
    {
      "entropy": 5.93649697303772,
      "epoch": 2.1047802149785895,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00045555239861266964,
      "loss": 5.2315,
      "mean_token_accuracy": 0.20491039454936982,
      "num_tokens": 48806288.0,
      "step": 24085
    },
    {
      "entropy": 5.859267473220825,
      "epoch": 2.1052171633312944,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004555338026645744,
      "loss": 5.187,
      "mean_token_accuracy": 0.21962300837039947,
      "num_tokens": 48816025.0,
      "step": 24090
    },
    {
      "entropy": 5.922202253341675,
      "epoch": 2.105654111683999,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004555152032537463,
      "loss": 5.3189,
      "mean_token_accuracy": 0.19932321459054947,
      "num_tokens": 48827150.0,
      "step": 24095
    },
    {
      "entropy": 5.964849615097046,
      "epoch": 2.1060910600367038,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045549660038054205,
      "loss": 5.243,
      "mean_token_accuracy": 0.202741739153862,
      "num_tokens": 48837190.0,
      "step": 24100
    },
    {
      "entropy": 5.99328122138977,
      "epoch": 2.1065280083894082,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045547799404531826,
      "loss": 5.3056,
      "mean_token_accuracy": 0.19648630023002625,
      "num_tokens": 48847573.0,
      "step": 24105
    },
    {
      "entropy": 5.867418432235718,
      "epoch": 2.106964956742113,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045545938424843214,
      "loss": 5.2051,
      "mean_token_accuracy": 0.20027976781129836,
      "num_tokens": 48858966.0,
      "step": 24110
    },
    {
      "entropy": 5.984283018112182,
      "epoch": 2.1074019050948176,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004554407709902404,
      "loss": 5.2855,
      "mean_token_accuracy": 0.19440836757421492,
      "num_tokens": 48869027.0,
      "step": 24115
    },
    {
      "entropy": 5.941946315765381,
      "epoch": 2.1078388534475225,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004554221542711001,
      "loss": 5.2762,
      "mean_token_accuracy": 0.2087750807404518,
      "num_tokens": 48879325.0,
      "step": 24120
    },
    {
      "entropy": 5.936623191833496,
      "epoch": 2.108275801800227,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045540353409136836,
      "loss": 5.1678,
      "mean_token_accuracy": 0.20785599052906037,
      "num_tokens": 48888939.0,
      "step": 24125
    },
    {
      "entropy": 5.926378393173218,
      "epoch": 2.108712750152932,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045538491045140234,
      "loss": 5.2665,
      "mean_token_accuracy": 0.19823075979948043,
      "num_tokens": 48898997.0,
      "step": 24130
    },
    {
      "entropy": 5.957346105575562,
      "epoch": 2.109149698505637,
      "grad_norm": 1.125,
      "learning_rate": 0.00045536628335155914,
      "loss": 5.2683,
      "mean_token_accuracy": 0.20316821485757827,
      "num_tokens": 48909063.0,
      "step": 24135
    },
    {
      "entropy": 5.85595555305481,
      "epoch": 2.1095866468583413,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045534765279219626,
      "loss": 5.1474,
      "mean_token_accuracy": 0.21492865085601806,
      "num_tokens": 48918694.0,
      "step": 24140
    },
    {
      "entropy": 5.92466459274292,
      "epoch": 2.110023595211046,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004553290187736708,
      "loss": 5.3373,
      "mean_token_accuracy": 0.19514743536710738,
      "num_tokens": 48928223.0,
      "step": 24145
    },
    {
      "entropy": 5.95574722290039,
      "epoch": 2.1104605435637507,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004553103812963402,
      "loss": 5.2551,
      "mean_token_accuracy": 0.20433495193719864,
      "num_tokens": 48938193.0,
      "step": 24150
    },
    {
      "entropy": 6.018911504745484,
      "epoch": 2.1108974919164556,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045529174036056217,
      "loss": 5.3545,
      "mean_token_accuracy": 0.1899367243051529,
      "num_tokens": 48949693.0,
      "step": 24155
    },
    {
      "entropy": 5.8731447696685795,
      "epoch": 2.11133444026916,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045527309596669397,
      "loss": 5.2327,
      "mean_token_accuracy": 0.20925828516483308,
      "num_tokens": 48960199.0,
      "step": 24160
    },
    {
      "entropy": 5.942251873016358,
      "epoch": 2.111771388621865,
      "grad_norm": 1.125,
      "learning_rate": 0.00045525444811509336,
      "loss": 5.3212,
      "mean_token_accuracy": 0.20120580196380616,
      "num_tokens": 48969380.0,
      "step": 24165
    },
    {
      "entropy": 5.931191110610962,
      "epoch": 2.1122083369745694,
      "grad_norm": 1.140625,
      "learning_rate": 0.000455235796806118,
      "loss": 5.2001,
      "mean_token_accuracy": 0.20626599341630936,
      "num_tokens": 48978963.0,
      "step": 24170
    },
    {
      "entropy": 5.956633806228638,
      "epoch": 2.1126452853272744,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004552171420401256,
      "loss": 5.3463,
      "mean_token_accuracy": 0.20360541343688965,
      "num_tokens": 48989568.0,
      "step": 24175
    },
    {
      "entropy": 5.958854532241821,
      "epoch": 2.113082233679979,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045519848381747396,
      "loss": 5.3032,
      "mean_token_accuracy": 0.19827632308006288,
      "num_tokens": 49000684.0,
      "step": 24180
    },
    {
      "entropy": 6.00737247467041,
      "epoch": 2.1135191820326837,
      "grad_norm": 1.109375,
      "learning_rate": 0.000455179822138521,
      "loss": 5.353,
      "mean_token_accuracy": 0.1906353548169136,
      "num_tokens": 49010194.0,
      "step": 24185
    },
    {
      "entropy": 5.979785299301147,
      "epoch": 2.1139561303853887,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045516115700362466,
      "loss": 5.2621,
      "mean_token_accuracy": 0.20389120429754257,
      "num_tokens": 49020180.0,
      "step": 24190
    },
    {
      "entropy": 5.956661748886108,
      "epoch": 2.114393078738093,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045514248841314295,
      "loss": 5.2844,
      "mean_token_accuracy": 0.2028278201818466,
      "num_tokens": 49030665.0,
      "step": 24195
    },
    {
      "entropy": 5.9857104301452635,
      "epoch": 2.114830027090798,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004551238163674338,
      "loss": 5.2582,
      "mean_token_accuracy": 0.20374209433794022,
      "num_tokens": 49040197.0,
      "step": 24200
    },
    {
      "entropy": 5.938973474502563,
      "epoch": 2.1152669754435025,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045510514086685553,
      "loss": 5.2745,
      "mean_token_accuracy": 0.198073311150074,
      "num_tokens": 49049996.0,
      "step": 24205
    },
    {
      "entropy": 5.944878053665161,
      "epoch": 2.1157039237962074,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004550864619117664,
      "loss": 5.3551,
      "mean_token_accuracy": 0.18824660331010817,
      "num_tokens": 49061691.0,
      "step": 24210
    },
    {
      "entropy": 6.006147670745849,
      "epoch": 2.116140872148912,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004550677795025244,
      "loss": 5.384,
      "mean_token_accuracy": 0.19238684624433516,
      "num_tokens": 49071605.0,
      "step": 24215
    },
    {
      "entropy": 5.987170934677124,
      "epoch": 2.116577820501617,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004550490936394881,
      "loss": 5.2106,
      "mean_token_accuracy": 0.20762132853269577,
      "num_tokens": 49081317.0,
      "step": 24220
    },
    {
      "entropy": 5.883607673645019,
      "epoch": 2.1170147688543213,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045503040432301586,
      "loss": 5.2358,
      "mean_token_accuracy": 0.20093652456998826,
      "num_tokens": 49090662.0,
      "step": 24225
    },
    {
      "entropy": 6.019442462921143,
      "epoch": 2.117451717207026,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004550117115534661,
      "loss": 5.3825,
      "mean_token_accuracy": 0.18996863961219787,
      "num_tokens": 49102046.0,
      "step": 24230
    },
    {
      "entropy": 6.031612157821655,
      "epoch": 2.1178886655597307,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045499301533119736,
      "loss": 5.2438,
      "mean_token_accuracy": 0.20171507745981215,
      "num_tokens": 49111047.0,
      "step": 24235
    },
    {
      "entropy": 5.953808975219727,
      "epoch": 2.1183256139124356,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004549743156565684,
      "loss": 5.2321,
      "mean_token_accuracy": 0.20087315887212753,
      "num_tokens": 49120606.0,
      "step": 24240
    },
    {
      "entropy": 5.937906551361084,
      "epoch": 2.1187625622651405,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004549556125299377,
      "loss": 5.3225,
      "mean_token_accuracy": 0.20118786245584488,
      "num_tokens": 49130981.0,
      "step": 24245
    },
    {
      "entropy": 5.985271024703979,
      "epoch": 2.119199510617845,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004549369059516639,
      "loss": 5.3483,
      "mean_token_accuracy": 0.1980021193623543,
      "num_tokens": 49140843.0,
      "step": 24250
    },
    {
      "entropy": 6.029823303222656,
      "epoch": 2.11963645897055,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004549181959221061,
      "loss": 5.3312,
      "mean_token_accuracy": 0.19040898233652115,
      "num_tokens": 49150492.0,
      "step": 24255
    },
    {
      "entropy": 5.962599563598633,
      "epoch": 2.1200734073232543,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045489948244162304,
      "loss": 5.2646,
      "mean_token_accuracy": 0.20021840929985046,
      "num_tokens": 49159973.0,
      "step": 24260
    },
    {
      "entropy": 6.027772617340088,
      "epoch": 2.1205103556759592,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004548807655105736,
      "loss": 5.3409,
      "mean_token_accuracy": 0.199467633664608,
      "num_tokens": 49170711.0,
      "step": 24265
    },
    {
      "entropy": 5.995313739776611,
      "epoch": 2.1209473040286637,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045486204512931684,
      "loss": 5.338,
      "mean_token_accuracy": 0.19651162028312683,
      "num_tokens": 49181121.0,
      "step": 24270
    },
    {
      "entropy": 5.837592363357544,
      "epoch": 2.1213842523813686,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004548433212982118,
      "loss": 5.1697,
      "mean_token_accuracy": 0.21684881746768953,
      "num_tokens": 49191528.0,
      "step": 24275
    },
    {
      "entropy": 5.926962614059448,
      "epoch": 2.121821200734073,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004548245940176176,
      "loss": 5.2975,
      "mean_token_accuracy": 0.19689151793718337,
      "num_tokens": 49202244.0,
      "step": 24280
    },
    {
      "entropy": 5.960758399963379,
      "epoch": 2.122258149086778,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004548058632878935,
      "loss": 5.2743,
      "mean_token_accuracy": 0.20289418399333953,
      "num_tokens": 49212964.0,
      "step": 24285
    },
    {
      "entropy": 6.0021501064300535,
      "epoch": 2.1226950974394825,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045478712910939874,
      "loss": 5.2958,
      "mean_token_accuracy": 0.19785286188125611,
      "num_tokens": 49223478.0,
      "step": 24290
    },
    {
      "entropy": 5.952851009368897,
      "epoch": 2.1231320457921874,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045476839148249256,
      "loss": 5.3122,
      "mean_token_accuracy": 0.19686800092458726,
      "num_tokens": 49233933.0,
      "step": 24295
    },
    {
      "entropy": 5.96302638053894,
      "epoch": 2.123568994144892,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045474965040753453,
      "loss": 5.2084,
      "mean_token_accuracy": 0.21104948222637177,
      "num_tokens": 49243147.0,
      "step": 24300
    },
    {
      "entropy": 5.943764734268188,
      "epoch": 2.1240059424975968,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004547309058848839,
      "loss": 5.2159,
      "mean_token_accuracy": 0.20706461369991302,
      "num_tokens": 49252524.0,
      "step": 24305
    },
    {
      "entropy": 5.948262310028076,
      "epoch": 2.1244428908503017,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045471215791490047,
      "loss": 5.3087,
      "mean_token_accuracy": 0.1999362289905548,
      "num_tokens": 49261845.0,
      "step": 24310
    },
    {
      "entropy": 5.941661977767945,
      "epoch": 2.124879839203006,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045469340649794353,
      "loss": 5.2351,
      "mean_token_accuracy": 0.20313380807638168,
      "num_tokens": 49272437.0,
      "step": 24315
    },
    {
      "entropy": 5.937484407424927,
      "epoch": 2.125316787555711,
      "grad_norm": 1.265625,
      "learning_rate": 0.000454674651634373,
      "loss": 5.2132,
      "mean_token_accuracy": 0.21027769446372985,
      "num_tokens": 49281687.0,
      "step": 24320
    },
    {
      "entropy": 5.941183233261109,
      "epoch": 2.1257537359084155,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045465589332454844,
      "loss": 5.1846,
      "mean_token_accuracy": 0.20736036002635955,
      "num_tokens": 49291524.0,
      "step": 24325
    },
    {
      "entropy": 5.9566895961761475,
      "epoch": 2.1261906842611205,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045463713156882976,
      "loss": 5.2958,
      "mean_token_accuracy": 0.19854874163866043,
      "num_tokens": 49303124.0,
      "step": 24330
    },
    {
      "entropy": 5.982834196090698,
      "epoch": 2.126627632613825,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045461836636757674,
      "loss": 5.2301,
      "mean_token_accuracy": 0.2065385550260544,
      "num_tokens": 49313050.0,
      "step": 24335
    },
    {
      "entropy": 5.901877593994141,
      "epoch": 2.12706458096653,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045459959772114935,
      "loss": 5.2137,
      "mean_token_accuracy": 0.20753683298826217,
      "num_tokens": 49321652.0,
      "step": 24340
    },
    {
      "entropy": 5.8800609588623045,
      "epoch": 2.1275015293192343,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045458082562990754,
      "loss": 5.2203,
      "mean_token_accuracy": 0.2035089299082756,
      "num_tokens": 49331713.0,
      "step": 24345
    },
    {
      "entropy": 5.891234350204468,
      "epoch": 2.127938477671939,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004545620500942114,
      "loss": 5.2667,
      "mean_token_accuracy": 0.20020754635334015,
      "num_tokens": 49340370.0,
      "step": 24350
    },
    {
      "entropy": 5.974067449569702,
      "epoch": 2.1283754260246437,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004545432711144211,
      "loss": 5.2818,
      "mean_token_accuracy": 0.20438632220029831,
      "num_tokens": 49349894.0,
      "step": 24355
    },
    {
      "entropy": 5.9571973323822025,
      "epoch": 2.1288123743773486,
      "grad_norm": 1.125,
      "learning_rate": 0.0004545244886908967,
      "loss": 5.1801,
      "mean_token_accuracy": 0.20820298194885253,
      "num_tokens": 49359738.0,
      "step": 24360
    },
    {
      "entropy": 5.947053241729736,
      "epoch": 2.129249322730053,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045450570282399856,
      "loss": 5.2751,
      "mean_token_accuracy": 0.201566681265831,
      "num_tokens": 49370495.0,
      "step": 24365
    },
    {
      "entropy": 5.972046852111816,
      "epoch": 2.129686271082758,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000454486913514087,
      "loss": 5.2818,
      "mean_token_accuracy": 0.2052431061863899,
      "num_tokens": 49382191.0,
      "step": 24370
    },
    {
      "entropy": 5.984591436386109,
      "epoch": 2.130123219435463,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004544681207615224,
      "loss": 5.3279,
      "mean_token_accuracy": 0.19461053609848022,
      "num_tokens": 49391724.0,
      "step": 24375
    },
    {
      "entropy": 5.958333683013916,
      "epoch": 2.1305601677881674,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004544493245666652,
      "loss": 5.2337,
      "mean_token_accuracy": 0.21090283542871474,
      "num_tokens": 49401117.0,
      "step": 24380
    },
    {
      "entropy": 5.922156810760498,
      "epoch": 2.1309971161408723,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004544305249298759,
      "loss": 5.244,
      "mean_token_accuracy": 0.2000494435429573,
      "num_tokens": 49410731.0,
      "step": 24385
    },
    {
      "entropy": 5.963088846206665,
      "epoch": 2.1314340644935768,
      "grad_norm": 1.125,
      "learning_rate": 0.0004544117218515152,
      "loss": 5.3053,
      "mean_token_accuracy": 0.20320320427417754,
      "num_tokens": 49420373.0,
      "step": 24390
    },
    {
      "entropy": 5.946637153625488,
      "epoch": 2.1318710128462817,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045439291533194356,
      "loss": 5.203,
      "mean_token_accuracy": 0.20581171065568923,
      "num_tokens": 49430021.0,
      "step": 24395
    },
    {
      "entropy": 5.922928953170777,
      "epoch": 2.132307961198986,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004543741053715218,
      "loss": 5.1981,
      "mean_token_accuracy": 0.20803173184394835,
      "num_tokens": 49439687.0,
      "step": 24400
    },
    {
      "entropy": 5.952748394012451,
      "epoch": 2.132744909551691,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004543552919706107,
      "loss": 5.38,
      "mean_token_accuracy": 0.19341195672750472,
      "num_tokens": 49450320.0,
      "step": 24405
    },
    {
      "entropy": 5.952393865585327,
      "epoch": 2.1331818579043955,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004543364751295712,
      "loss": 5.2021,
      "mean_token_accuracy": 0.2034415692090988,
      "num_tokens": 49461272.0,
      "step": 24410
    },
    {
      "entropy": 6.002634429931641,
      "epoch": 2.1336188062571004,
      "grad_norm": 1.0,
      "learning_rate": 0.00045431765484876406,
      "loss": 5.3179,
      "mean_token_accuracy": 0.19207982569932938,
      "num_tokens": 49473087.0,
      "step": 24415
    },
    {
      "entropy": 5.880703020095825,
      "epoch": 2.134055754609805,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004542988311285505,
      "loss": 5.2183,
      "mean_token_accuracy": 0.2099234566092491,
      "num_tokens": 49483174.0,
      "step": 24420
    },
    {
      "entropy": 5.934034538269043,
      "epoch": 2.13449270296251,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045428000396929125,
      "loss": 5.2564,
      "mean_token_accuracy": 0.2039696216583252,
      "num_tokens": 49492267.0,
      "step": 24425
    },
    {
      "entropy": 5.934001541137695,
      "epoch": 2.1349296513152147,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045426117337134757,
      "loss": 5.2111,
      "mean_token_accuracy": 0.20237984508275986,
      "num_tokens": 49502338.0,
      "step": 24430
    },
    {
      "entropy": 5.966488838195801,
      "epoch": 2.135366599667919,
      "grad_norm": 1.390625,
      "learning_rate": 0.00045424233933508075,
      "loss": 5.2522,
      "mean_token_accuracy": 0.2087568074464798,
      "num_tokens": 49512991.0,
      "step": 24435
    },
    {
      "entropy": 5.924130296707153,
      "epoch": 2.135803548020624,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004542235018608519,
      "loss": 5.1615,
      "mean_token_accuracy": 0.2098268151283264,
      "num_tokens": 49522224.0,
      "step": 24440
    },
    {
      "entropy": 5.91452465057373,
      "epoch": 2.1362404963733286,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045420466094902234,
      "loss": 5.3064,
      "mean_token_accuracy": 0.19481339454650878,
      "num_tokens": 49531982.0,
      "step": 24445
    },
    {
      "entropy": 5.973621845245361,
      "epoch": 2.1366774447260335,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004541858165999535,
      "loss": 5.1838,
      "mean_token_accuracy": 0.2114153355360031,
      "num_tokens": 49541487.0,
      "step": 24450
    },
    {
      "entropy": 5.956585931777954,
      "epoch": 2.137114393078738,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004541669688140068,
      "loss": 5.3501,
      "mean_token_accuracy": 0.19152033627033233,
      "num_tokens": 49552052.0,
      "step": 24455
    },
    {
      "entropy": 5.915628719329834,
      "epoch": 2.137551341431443,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004541481175915438,
      "loss": 5.2948,
      "mean_token_accuracy": 0.19600663781166078,
      "num_tokens": 49561807.0,
      "step": 24460
    },
    {
      "entropy": 5.996598434448242,
      "epoch": 2.1379882897841473,
      "grad_norm": 1.328125,
      "learning_rate": 0.000454129262932926,
      "loss": 5.3075,
      "mean_token_accuracy": 0.19842616021633147,
      "num_tokens": 49570687.0,
      "step": 24465
    },
    {
      "entropy": 5.971085739135742,
      "epoch": 2.1384252381368523,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045411040483851506,
      "loss": 5.2788,
      "mean_token_accuracy": 0.19919408708810807,
      "num_tokens": 49581000.0,
      "step": 24470
    },
    {
      "entropy": 5.929480266571045,
      "epoch": 2.1388621864895567,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045409154330867274,
      "loss": 5.3248,
      "mean_token_accuracy": 0.19859050810337067,
      "num_tokens": 49590442.0,
      "step": 24475
    },
    {
      "entropy": 5.949670934677124,
      "epoch": 2.1392991348422616,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045407267834376075,
      "loss": 5.2711,
      "mean_token_accuracy": 0.20241689085960388,
      "num_tokens": 49600068.0,
      "step": 24480
    },
    {
      "entropy": 5.98914589881897,
      "epoch": 2.1397360831949666,
      "grad_norm": 1.125,
      "learning_rate": 0.000454053809944141,
      "loss": 5.3032,
      "mean_token_accuracy": 0.1951017364859581,
      "num_tokens": 49610696.0,
      "step": 24485
    },
    {
      "entropy": 5.995771646499634,
      "epoch": 2.140173031547671,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004540349381101753,
      "loss": 5.2922,
      "mean_token_accuracy": 0.20454777777194977,
      "num_tokens": 49619489.0,
      "step": 24490
    },
    {
      "entropy": 5.976050615310669,
      "epoch": 2.140609979900376,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045401606284222563,
      "loss": 5.275,
      "mean_token_accuracy": 0.20145511776208877,
      "num_tokens": 49629572.0,
      "step": 24495
    },
    {
      "entropy": 5.997118759155273,
      "epoch": 2.1410469282530804,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045399718414065415,
      "loss": 5.2326,
      "mean_token_accuracy": 0.2037989690899849,
      "num_tokens": 49640035.0,
      "step": 24500
    },
    {
      "entropy": 5.921063566207886,
      "epoch": 2.1414838766057853,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004539783020058229,
      "loss": 5.2521,
      "mean_token_accuracy": 0.20732864439487458,
      "num_tokens": 49649671.0,
      "step": 24505
    },
    {
      "entropy": 5.925938844680786,
      "epoch": 2.14192082495849,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000453959416438094,
      "loss": 5.2699,
      "mean_token_accuracy": 0.20255377292633056,
      "num_tokens": 49659122.0,
      "step": 24510
    },
    {
      "entropy": 5.9754493713378904,
      "epoch": 2.1423577733111947,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004539405274378297,
      "loss": 5.2271,
      "mean_token_accuracy": 0.2002366676926613,
      "num_tokens": 49669938.0,
      "step": 24515
    },
    {
      "entropy": 5.989089441299439,
      "epoch": 2.142794721663899,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004539216350053924,
      "loss": 5.2528,
      "mean_token_accuracy": 0.20653676241636276,
      "num_tokens": 49680783.0,
      "step": 24520
    },
    {
      "entropy": 5.960072994232178,
      "epoch": 2.143231670016604,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004539027391411443,
      "loss": 5.3194,
      "mean_token_accuracy": 0.19772512018680571,
      "num_tokens": 49691271.0,
      "step": 24525
    },
    {
      "entropy": 5.974129867553711,
      "epoch": 2.1436686183693086,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045388383984544794,
      "loss": 5.286,
      "mean_token_accuracy": 0.20037923008203506,
      "num_tokens": 49702496.0,
      "step": 24530
    },
    {
      "entropy": 5.982511711120606,
      "epoch": 2.1441055667220135,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045386493711866577,
      "loss": 5.292,
      "mean_token_accuracy": 0.20275137573480606,
      "num_tokens": 49712145.0,
      "step": 24535
    },
    {
      "entropy": 5.975623846054077,
      "epoch": 2.1445425150747184,
      "grad_norm": 1.125,
      "learning_rate": 0.00045384603096116044,
      "loss": 5.2767,
      "mean_token_accuracy": 0.2001498132944107,
      "num_tokens": 49722243.0,
      "step": 24540
    },
    {
      "entropy": 6.019931364059448,
      "epoch": 2.144979463427423,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004538271213732945,
      "loss": 5.322,
      "mean_token_accuracy": 0.19244882315397263,
      "num_tokens": 49731804.0,
      "step": 24545
    },
    {
      "entropy": 5.993660736083984,
      "epoch": 2.1454164117801278,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045380820835543066,
      "loss": 5.2234,
      "mean_token_accuracy": 0.2018257588148117,
      "num_tokens": 49741275.0,
      "step": 24550
    },
    {
      "entropy": 5.918096256256104,
      "epoch": 2.1458533601328322,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004537892919079317,
      "loss": 5.2231,
      "mean_token_accuracy": 0.20266827791929246,
      "num_tokens": 49751140.0,
      "step": 24555
    },
    {
      "entropy": 5.972182273864746,
      "epoch": 2.146290308485537,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045377037203116053,
      "loss": 5.2829,
      "mean_token_accuracy": 0.20492403209209442,
      "num_tokens": 49760747.0,
      "step": 24560
    },
    {
      "entropy": 5.994121074676514,
      "epoch": 2.1467272568382416,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045375144872547975,
      "loss": 5.3108,
      "mean_token_accuracy": 0.19376776069402696,
      "num_tokens": 49771491.0,
      "step": 24565
    },
    {
      "entropy": 5.989454698562622,
      "epoch": 2.1471642051909465,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004537325219912527,
      "loss": 5.2692,
      "mean_token_accuracy": 0.2022857651114464,
      "num_tokens": 49781523.0,
      "step": 24570
    },
    {
      "entropy": 6.002940702438354,
      "epoch": 2.147601153543651,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004537135918288421,
      "loss": 5.3541,
      "mean_token_accuracy": 0.1903823807835579,
      "num_tokens": 49791560.0,
      "step": 24575
    },
    {
      "entropy": 5.909886455535888,
      "epoch": 2.148038101896356,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004536946582386112,
      "loss": 5.2189,
      "mean_token_accuracy": 0.19900162220001222,
      "num_tokens": 49800588.0,
      "step": 24580
    },
    {
      "entropy": 5.910055685043335,
      "epoch": 2.1484750502490604,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045367572122092316,
      "loss": 5.2667,
      "mean_token_accuracy": 0.20490746051073075,
      "num_tokens": 49810381.0,
      "step": 24585
    },
    {
      "entropy": 5.998483610153198,
      "epoch": 2.1489119986017653,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004536567807761411,
      "loss": 5.2936,
      "mean_token_accuracy": 0.20589903146028518,
      "num_tokens": 49820931.0,
      "step": 24590
    },
    {
      "entropy": 5.949316501617432,
      "epoch": 2.14934894695447,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045363783690462845,
      "loss": 5.3041,
      "mean_token_accuracy": 0.19690924733877183,
      "num_tokens": 49831264.0,
      "step": 24595
    },
    {
      "entropy": 5.949942541122437,
      "epoch": 2.1497858953071747,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004536188896067483,
      "loss": 5.3112,
      "mean_token_accuracy": 0.20440936833620071,
      "num_tokens": 49841656.0,
      "step": 24600
    },
    {
      "entropy": 5.936012887954712,
      "epoch": 2.1502228436598796,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004535999388828645,
      "loss": 5.1538,
      "mean_token_accuracy": 0.21376682221889495,
      "num_tokens": 49851127.0,
      "step": 24605
    },
    {
      "entropy": 5.9712096691131595,
      "epoch": 2.150659792012584,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004535809847333402,
      "loss": 5.241,
      "mean_token_accuracy": 0.2037417247891426,
      "num_tokens": 49860563.0,
      "step": 24610
    },
    {
      "entropy": 5.911773824691773,
      "epoch": 2.151096740365289,
      "grad_norm": 1.15625,
      "learning_rate": 0.000453562027158539,
      "loss": 5.2091,
      "mean_token_accuracy": 0.20227329432964325,
      "num_tokens": 49869741.0,
      "step": 24615
    },
    {
      "entropy": 5.964188814163208,
      "epoch": 2.1515336887179934,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004535430661588245,
      "loss": 5.2979,
      "mean_token_accuracy": 0.1968265786767006,
      "num_tokens": 49879664.0,
      "step": 24620
    },
    {
      "entropy": 5.985727119445801,
      "epoch": 2.1519706370706984,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045352410173456057,
      "loss": 5.2364,
      "mean_token_accuracy": 0.2067960634827614,
      "num_tokens": 49889322.0,
      "step": 24625
    },
    {
      "entropy": 6.018596839904785,
      "epoch": 2.152407585423403,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045350513388611074,
      "loss": 5.3092,
      "mean_token_accuracy": 0.19451871365308762,
      "num_tokens": 49899172.0,
      "step": 24630
    },
    {
      "entropy": 5.972682189941406,
      "epoch": 2.1528445337761077,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045348616261383897,
      "loss": 5.2702,
      "mean_token_accuracy": 0.20226945132017135,
      "num_tokens": 49909211.0,
      "step": 24635
    },
    {
      "entropy": 5.972480344772339,
      "epoch": 2.153281482128812,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045346718791810896,
      "loss": 5.2744,
      "mean_token_accuracy": 0.21183883845806123,
      "num_tokens": 49918884.0,
      "step": 24640
    },
    {
      "entropy": 5.926026105880737,
      "epoch": 2.153718430481517,
      "grad_norm": 1.609375,
      "learning_rate": 0.00045344820979928484,
      "loss": 5.2487,
      "mean_token_accuracy": 0.2126406565308571,
      "num_tokens": 49929558.0,
      "step": 24645
    },
    {
      "entropy": 5.9438543796539305,
      "epoch": 2.1541553788342216,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045342922825773057,
      "loss": 5.2979,
      "mean_token_accuracy": 0.20477118641138076,
      "num_tokens": 49939831.0,
      "step": 24650
    },
    {
      "entropy": 5.99707293510437,
      "epoch": 2.1545923271869265,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045341024329381024,
      "loss": 5.2476,
      "mean_token_accuracy": 0.20484984517097474,
      "num_tokens": 49949690.0,
      "step": 24655
    },
    {
      "entropy": 6.016684293746948,
      "epoch": 2.1550292755396314,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004533912549078879,
      "loss": 5.2942,
      "mean_token_accuracy": 0.19786290526390077,
      "num_tokens": 49959154.0,
      "step": 24660
    },
    {
      "entropy": 5.951034307479858,
      "epoch": 2.155466223892336,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045337226310032783,
      "loss": 5.3222,
      "mean_token_accuracy": 0.1987508863210678,
      "num_tokens": 49970185.0,
      "step": 24665
    },
    {
      "entropy": 5.991044616699218,
      "epoch": 2.155903172245041,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004533532678714943,
      "loss": 5.3443,
      "mean_token_accuracy": 0.19690345078706742,
      "num_tokens": 49980489.0,
      "step": 24670
    },
    {
      "entropy": 5.965857315063476,
      "epoch": 2.1563401205977453,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045333426922175156,
      "loss": 5.2998,
      "mean_token_accuracy": 0.20120100528001786,
      "num_tokens": 49990038.0,
      "step": 24675
    },
    {
      "entropy": 5.977162218093872,
      "epoch": 2.15677706895045,
      "grad_norm": 1.25,
      "learning_rate": 0.0004533152671514641,
      "loss": 5.2811,
      "mean_token_accuracy": 0.198566997051239,
      "num_tokens": 49998888.0,
      "step": 24680
    },
    {
      "entropy": 5.89909405708313,
      "epoch": 2.1572140173031547,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045329626166099653,
      "loss": 5.2167,
      "mean_token_accuracy": 0.20657225400209428,
      "num_tokens": 50009349.0,
      "step": 24685
    },
    {
      "entropy": 6.003435373306274,
      "epoch": 2.1576509656558596,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045327725275071313,
      "loss": 5.356,
      "mean_token_accuracy": 0.19464871734380723,
      "num_tokens": 50019626.0,
      "step": 24690
    },
    {
      "entropy": 5.980347108840943,
      "epoch": 2.158087914008564,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004532582404209786,
      "loss": 5.2341,
      "mean_token_accuracy": 0.20380001217126847,
      "num_tokens": 50029281.0,
      "step": 24695
    },
    {
      "entropy": 5.946800184249878,
      "epoch": 2.158524862361269,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004532392246721576,
      "loss": 5.2503,
      "mean_token_accuracy": 0.20609516948461531,
      "num_tokens": 50038700.0,
      "step": 24700
    },
    {
      "entropy": 5.962970209121704,
      "epoch": 2.1589618107139734,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004532202055046149,
      "loss": 5.2239,
      "mean_token_accuracy": 0.20182902216911316,
      "num_tokens": 50048904.0,
      "step": 24705
    },
    {
      "entropy": 5.934461450576782,
      "epoch": 2.1593987590666783,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004532011829187153,
      "loss": 5.2708,
      "mean_token_accuracy": 0.19939713329076766,
      "num_tokens": 50059188.0,
      "step": 24710
    },
    {
      "entropy": 5.926176071166992,
      "epoch": 2.159835707419383,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004531821569148236,
      "loss": 5.2414,
      "mean_token_accuracy": 0.20804848074913024,
      "num_tokens": 50069069.0,
      "step": 24715
    },
    {
      "entropy": 5.949406480789184,
      "epoch": 2.1602726557720877,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045316312749330473,
      "loss": 5.2654,
      "mean_token_accuracy": 0.20317544043064117,
      "num_tokens": 50078904.0,
      "step": 24720
    },
    {
      "entropy": 5.886470317840576,
      "epoch": 2.1607096041247926,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045314409465452385,
      "loss": 5.2073,
      "mean_token_accuracy": 0.2064420461654663,
      "num_tokens": 50089897.0,
      "step": 24725
    },
    {
      "entropy": 5.991319179534912,
      "epoch": 2.161146552477497,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045312505839884573,
      "loss": 5.3082,
      "mean_token_accuracy": 0.20097725242376327,
      "num_tokens": 50100030.0,
      "step": 24730
    },
    {
      "entropy": 5.955930900573731,
      "epoch": 2.161583500830202,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045310601872663574,
      "loss": 5.2657,
      "mean_token_accuracy": 0.20367031544446945,
      "num_tokens": 50110043.0,
      "step": 24735
    },
    {
      "entropy": 5.922087621688843,
      "epoch": 2.1620204491829065,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045308697563825894,
      "loss": 5.2514,
      "mean_token_accuracy": 0.2054854467511177,
      "num_tokens": 50120081.0,
      "step": 24740
    },
    {
      "entropy": 5.901795864105225,
      "epoch": 2.1624573975356114,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004530679291340807,
      "loss": 5.2666,
      "mean_token_accuracy": 0.20627101957798005,
      "num_tokens": 50129834.0,
      "step": 24745
    },
    {
      "entropy": 5.93374490737915,
      "epoch": 2.162894345888316,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004530488792144662,
      "loss": 5.2919,
      "mean_token_accuracy": 0.1974508449435234,
      "num_tokens": 50140853.0,
      "step": 24750
    },
    {
      "entropy": 5.9950535774230955,
      "epoch": 2.1633312942410208,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000453029825879781,
      "loss": 5.3643,
      "mean_token_accuracy": 0.19747218787670134,
      "num_tokens": 50151341.0,
      "step": 24755
    },
    {
      "entropy": 5.953194093704224,
      "epoch": 2.1637682425937252,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045301076913039035,
      "loss": 5.2109,
      "mean_token_accuracy": 0.20803853273391723,
      "num_tokens": 50160826.0,
      "step": 24760
    },
    {
      "entropy": 5.944432544708252,
      "epoch": 2.16420519094643,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004529917089666599,
      "loss": 5.2527,
      "mean_token_accuracy": 0.1987137407064438,
      "num_tokens": 50170720.0,
      "step": 24765
    },
    {
      "entropy": 5.93590898513794,
      "epoch": 2.1646421392991346,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004529726453889552,
      "loss": 5.2746,
      "mean_token_accuracy": 0.20586954802274704,
      "num_tokens": 50180707.0,
      "step": 24770
    },
    {
      "entropy": 5.869769954681397,
      "epoch": 2.1650790876518395,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045295357839764193,
      "loss": 5.2075,
      "mean_token_accuracy": 0.2156088724732399,
      "num_tokens": 50190714.0,
      "step": 24775
    },
    {
      "entropy": 6.0113836288452145,
      "epoch": 2.1655160360045445,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004529345079930858,
      "loss": 5.3194,
      "mean_token_accuracy": 0.19854533523321152,
      "num_tokens": 50201023.0,
      "step": 24780
    },
    {
      "entropy": 5.974842643737793,
      "epoch": 2.165952984357249,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004529154341756525,
      "loss": 5.2736,
      "mean_token_accuracy": 0.19417019337415695,
      "num_tokens": 50211064.0,
      "step": 24785
    },
    {
      "entropy": 6.033513307571411,
      "epoch": 2.166389932709954,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045289635694570796,
      "loss": 5.3281,
      "mean_token_accuracy": 0.1950130581855774,
      "num_tokens": 50220696.0,
      "step": 24790
    },
    {
      "entropy": 5.883748388290405,
      "epoch": 2.1668268810626583,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045287727630361814,
      "loss": 5.2614,
      "mean_token_accuracy": 0.20407490879297258,
      "num_tokens": 50230629.0,
      "step": 24795
    },
    {
      "entropy": 5.8887269496917725,
      "epoch": 2.167263829415363,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045285819224974894,
      "loss": 5.2435,
      "mean_token_accuracy": 0.2091246634721756,
      "num_tokens": 50241042.0,
      "step": 24800
    },
    {
      "entropy": 5.943576240539551,
      "epoch": 2.1677007777680677,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004528391047844664,
      "loss": 5.2022,
      "mean_token_accuracy": 0.20197365432977676,
      "num_tokens": 50251731.0,
      "step": 24805
    },
    {
      "entropy": 5.949787950515747,
      "epoch": 2.1681377261207726,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004528200139081366,
      "loss": 5.3316,
      "mean_token_accuracy": 0.19583869874477386,
      "num_tokens": 50262893.0,
      "step": 24810
    },
    {
      "entropy": 5.976612377166748,
      "epoch": 2.168574674473477,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004528009196211258,
      "loss": 5.3785,
      "mean_token_accuracy": 0.2009825140237808,
      "num_tokens": 50273666.0,
      "step": 24815
    },
    {
      "entropy": 5.981293249130249,
      "epoch": 2.169011622826182,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004527818219238002,
      "loss": 5.2491,
      "mean_token_accuracy": 0.19874435365200044,
      "num_tokens": 50283165.0,
      "step": 24820
    },
    {
      "entropy": 5.962756681442261,
      "epoch": 2.1694485711788865,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004527627208165261,
      "loss": 5.2268,
      "mean_token_accuracy": 0.20278798639774323,
      "num_tokens": 50291958.0,
      "step": 24825
    },
    {
      "entropy": 5.944627809524536,
      "epoch": 2.1698855195315914,
      "grad_norm": 1.125,
      "learning_rate": 0.00045274361629966985,
      "loss": 5.3348,
      "mean_token_accuracy": 0.19553903341293336,
      "num_tokens": 50302944.0,
      "step": 24830
    },
    {
      "entropy": 5.954710292816162,
      "epoch": 2.1703224678842963,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045272450837359783,
      "loss": 5.267,
      "mean_token_accuracy": 0.20523503720760344,
      "num_tokens": 50313088.0,
      "step": 24835
    },
    {
      "entropy": 5.987989854812622,
      "epoch": 2.1707594162370007,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045270539703867675,
      "loss": 5.2909,
      "mean_token_accuracy": 0.19676020741462708,
      "num_tokens": 50323121.0,
      "step": 24840
    },
    {
      "entropy": 5.963884973526001,
      "epoch": 2.1711963645897057,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000452686282295273,
      "loss": 5.2821,
      "mean_token_accuracy": 0.20384474694728852,
      "num_tokens": 50333091.0,
      "step": 24845
    },
    {
      "entropy": 5.90318489074707,
      "epoch": 2.17163331294241,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045266716414375324,
      "loss": 5.204,
      "mean_token_accuracy": 0.20769517868757248,
      "num_tokens": 50342752.0,
      "step": 24850
    },
    {
      "entropy": 5.909873056411743,
      "epoch": 2.172070261295115,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045264804258448416,
      "loss": 5.2314,
      "mean_token_accuracy": 0.20466307997703553,
      "num_tokens": 50352213.0,
      "step": 24855
    },
    {
      "entropy": 5.912405776977539,
      "epoch": 2.1725072096478195,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004526289176178325,
      "loss": 5.2881,
      "mean_token_accuracy": 0.19714402705430983,
      "num_tokens": 50362586.0,
      "step": 24860
    },
    {
      "entropy": 5.906331539154053,
      "epoch": 2.1729441580005244,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004526097892441652,
      "loss": 5.1809,
      "mean_token_accuracy": 0.20440046787261962,
      "num_tokens": 50372742.0,
      "step": 24865
    },
    {
      "entropy": 5.871242904663086,
      "epoch": 2.173381106353229,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045259065746384904,
      "loss": 5.1896,
      "mean_token_accuracy": 0.2107730895280838,
      "num_tokens": 50382857.0,
      "step": 24870
    },
    {
      "entropy": 5.960904788970947,
      "epoch": 2.173818054705934,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000452571522277251,
      "loss": 5.281,
      "mean_token_accuracy": 0.2034850999712944,
      "num_tokens": 50393638.0,
      "step": 24875
    },
    {
      "entropy": 5.950558423995972,
      "epoch": 2.1742550030586383,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004525523836847382,
      "loss": 5.3007,
      "mean_token_accuracy": 0.1988300174474716,
      "num_tokens": 50404928.0,
      "step": 24880
    },
    {
      "entropy": 5.929909324645996,
      "epoch": 2.174691951411343,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004525332416866776,
      "loss": 5.2009,
      "mean_token_accuracy": 0.20341508388519286,
      "num_tokens": 50414382.0,
      "step": 24885
    },
    {
      "entropy": 5.892805337905884,
      "epoch": 2.175128899764048,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004525140962834363,
      "loss": 5.2569,
      "mean_token_accuracy": 0.2054510459303856,
      "num_tokens": 50425203.0,
      "step": 24890
    },
    {
      "entropy": 6.034569501876831,
      "epoch": 2.1755658481167526,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004524949474753817,
      "loss": 5.4184,
      "mean_token_accuracy": 0.18487895280122757,
      "num_tokens": 50436395.0,
      "step": 24895
    },
    {
      "entropy": 5.951778554916382,
      "epoch": 2.1760027964694575,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004524757952628811,
      "loss": 5.267,
      "mean_token_accuracy": 0.20682300478219987,
      "num_tokens": 50446990.0,
      "step": 24900
    },
    {
      "entropy": 5.966698360443115,
      "epoch": 2.176439744822162,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045245663964630164,
      "loss": 5.3334,
      "mean_token_accuracy": 0.19724071770906448,
      "num_tokens": 50457964.0,
      "step": 24905
    },
    {
      "entropy": 5.990730619430542,
      "epoch": 2.176876693174867,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004524374806260108,
      "loss": 5.2636,
      "mean_token_accuracy": 0.20349325686693193,
      "num_tokens": 50467329.0,
      "step": 24910
    },
    {
      "entropy": 5.984342813491821,
      "epoch": 2.1773136415275713,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004524183182023762,
      "loss": 5.2682,
      "mean_token_accuracy": 0.20908669531345367,
      "num_tokens": 50476852.0,
      "step": 24915
    },
    {
      "entropy": 5.9285187244415285,
      "epoch": 2.1777505898802763,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045239915237576525,
      "loss": 5.2692,
      "mean_token_accuracy": 0.2043532148003578,
      "num_tokens": 50486207.0,
      "step": 24920
    },
    {
      "entropy": 5.947540760040283,
      "epoch": 2.1781875382329807,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045237998314654555,
      "loss": 5.3379,
      "mean_token_accuracy": 0.20209436416625975,
      "num_tokens": 50496457.0,
      "step": 24925
    },
    {
      "entropy": 6.019744062423706,
      "epoch": 2.1786244865856856,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004523608105150849,
      "loss": 5.3089,
      "mean_token_accuracy": 0.19082404524087906,
      "num_tokens": 50506430.0,
      "step": 24930
    },
    {
      "entropy": 5.9666930675506595,
      "epoch": 2.17906143493839,
      "grad_norm": 1.078125,
      "learning_rate": 0.000452341634481751,
      "loss": 5.2203,
      "mean_token_accuracy": 0.20400639921426772,
      "num_tokens": 50516204.0,
      "step": 24935
    },
    {
      "entropy": 5.933856010437012,
      "epoch": 2.179498383291095,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004523224550469115,
      "loss": 5.2613,
      "mean_token_accuracy": 0.19939871430397033,
      "num_tokens": 50525978.0,
      "step": 24940
    },
    {
      "entropy": 5.929870653152466,
      "epoch": 2.1799353316438,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004523032722109345,
      "loss": 5.2266,
      "mean_token_accuracy": 0.2040821686387062,
      "num_tokens": 50535579.0,
      "step": 24945
    },
    {
      "entropy": 5.959495925903321,
      "epoch": 2.1803722799965044,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004522840859741878,
      "loss": 5.2292,
      "mean_token_accuracy": 0.20840755105018616,
      "num_tokens": 50545765.0,
      "step": 24950
    },
    {
      "entropy": 6.04240140914917,
      "epoch": 2.1808092283492093,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045226489633703944,
      "loss": 5.3774,
      "mean_token_accuracy": 0.19865572899580003,
      "num_tokens": 50556237.0,
      "step": 24955
    },
    {
      "entropy": 5.996370029449463,
      "epoch": 2.181246176701914,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045224570329985747,
      "loss": 5.3154,
      "mean_token_accuracy": 0.1967099592089653,
      "num_tokens": 50567720.0,
      "step": 24960
    },
    {
      "entropy": 5.976959180831909,
      "epoch": 2.1816831250546187,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045222650686301,
      "loss": 5.229,
      "mean_token_accuracy": 0.20161017626523972,
      "num_tokens": 50577234.0,
      "step": 24965
    },
    {
      "entropy": 5.915200471878052,
      "epoch": 2.182120073407323,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004522073070268654,
      "loss": 5.3024,
      "mean_token_accuracy": 0.20403348803520202,
      "num_tokens": 50587416.0,
      "step": 24970
    },
    {
      "entropy": 5.9265389919281,
      "epoch": 2.182557021760028,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004521881037917917,
      "loss": 5.1938,
      "mean_token_accuracy": 0.21717019379138947,
      "num_tokens": 50598317.0,
      "step": 24975
    },
    {
      "entropy": 6.020588064193726,
      "epoch": 2.1829939701127326,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004521688971581573,
      "loss": 5.3323,
      "mean_token_accuracy": 0.19702928215265275,
      "num_tokens": 50608507.0,
      "step": 24980
    },
    {
      "entropy": 5.9314977645874025,
      "epoch": 2.1834309184654375,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045214968712633066,
      "loss": 5.2494,
      "mean_token_accuracy": 0.20893728137016296,
      "num_tokens": 50618625.0,
      "step": 24985
    },
    {
      "entropy": 5.989011955261231,
      "epoch": 2.183867866818142,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004521304736966802,
      "loss": 5.2571,
      "mean_token_accuracy": 0.20385199338197707,
      "num_tokens": 50628696.0,
      "step": 24990
    },
    {
      "entropy": 6.00729513168335,
      "epoch": 2.184304815170847,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045211125686957443,
      "loss": 5.3306,
      "mean_token_accuracy": 0.19629994183778762,
      "num_tokens": 50639313.0,
      "step": 24995
    },
    {
      "entropy": 5.955705833435059,
      "epoch": 2.1847417635235513,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045209203664538195,
      "loss": 5.2289,
      "mean_token_accuracy": 0.20377257615327835,
      "num_tokens": 50650715.0,
      "step": 25000
    },
    {
      "entropy": 5.934905481338501,
      "epoch": 2.1851787118762562,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004520728130244714,
      "loss": 5.2288,
      "mean_token_accuracy": 0.20069743245840072,
      "num_tokens": 50660224.0,
      "step": 25005
    },
    {
      "entropy": 5.960090923309326,
      "epoch": 2.185615660228961,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004520535860072115,
      "loss": 5.2577,
      "mean_token_accuracy": 0.2013276144862175,
      "num_tokens": 50670346.0,
      "step": 25010
    },
    {
      "entropy": 5.929535579681397,
      "epoch": 2.1860526085816656,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045203435559397113,
      "loss": 5.2708,
      "mean_token_accuracy": 0.19682337939739228,
      "num_tokens": 50680274.0,
      "step": 25015
    },
    {
      "entropy": 5.854575490951538,
      "epoch": 2.1864895569343705,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004520151217851189,
      "loss": 5.2115,
      "mean_token_accuracy": 0.20722474604845048,
      "num_tokens": 50690360.0,
      "step": 25020
    },
    {
      "entropy": 5.949490880966186,
      "epoch": 2.186926505287075,
      "grad_norm": 1.140625,
      "learning_rate": 0.000451995884581024,
      "loss": 5.2912,
      "mean_token_accuracy": 0.19983508735895156,
      "num_tokens": 50700798.0,
      "step": 25025
    },
    {
      "entropy": 5.971962356567383,
      "epoch": 2.18736345363978,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004519766439820552,
      "loss": 5.2928,
      "mean_token_accuracy": 0.19717081487178803,
      "num_tokens": 50710120.0,
      "step": 25030
    },
    {
      "entropy": 6.024322414398194,
      "epoch": 2.1878004019924844,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004519573999885817,
      "loss": 5.378,
      "mean_token_accuracy": 0.19704205244779588,
      "num_tokens": 50719903.0,
      "step": 25035
    },
    {
      "entropy": 5.96846718788147,
      "epoch": 2.1882373503451893,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045193815260097253,
      "loss": 5.307,
      "mean_token_accuracy": 0.20180035829544068,
      "num_tokens": 50730632.0,
      "step": 25040
    },
    {
      "entropy": 6.0196691989898685,
      "epoch": 2.1886742986978938,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045191890181959696,
      "loss": 5.3031,
      "mean_token_accuracy": 0.19792683869600297,
      "num_tokens": 50741263.0,
      "step": 25045
    },
    {
      "entropy": 5.940884828567505,
      "epoch": 2.1891112470505987,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045189964764482407,
      "loss": 5.2567,
      "mean_token_accuracy": 0.19753177911043168,
      "num_tokens": 50751394.0,
      "step": 25050
    },
    {
      "entropy": 5.950878143310547,
      "epoch": 2.189548195403303,
      "grad_norm": 1.125,
      "learning_rate": 0.0004518803900770232,
      "loss": 5.2745,
      "mean_token_accuracy": 0.20296916663646697,
      "num_tokens": 50762038.0,
      "step": 25055
    },
    {
      "entropy": 6.026395702362061,
      "epoch": 2.189985143756008,
      "grad_norm": 1.125,
      "learning_rate": 0.00045186112911656387,
      "loss": 5.3333,
      "mean_token_accuracy": 0.1944015696644783,
      "num_tokens": 50774063.0,
      "step": 25060
    },
    {
      "entropy": 6.00710711479187,
      "epoch": 2.1904220921087125,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004518418647638153,
      "loss": 5.3108,
      "mean_token_accuracy": 0.2006281554698944,
      "num_tokens": 50784606.0,
      "step": 25065
    },
    {
      "entropy": 5.966242408752441,
      "epoch": 2.1908590404614174,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004518225970191471,
      "loss": 5.2893,
      "mean_token_accuracy": 0.19842741191387175,
      "num_tokens": 50794892.0,
      "step": 25070
    },
    {
      "entropy": 5.910374641418457,
      "epoch": 2.1912959888141224,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00045180332588292896,
      "loss": 5.3099,
      "mean_token_accuracy": 0.2027555376291275,
      "num_tokens": 50805903.0,
      "step": 25075
    },
    {
      "entropy": 5.925384569168091,
      "epoch": 2.191732937166827,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004517840513555303,
      "loss": 5.176,
      "mean_token_accuracy": 0.21003988087177278,
      "num_tokens": 50815339.0,
      "step": 25080
    },
    {
      "entropy": 5.966205406188965,
      "epoch": 2.1921698855195317,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004517647734373209,
      "loss": 5.3208,
      "mean_token_accuracy": 0.19745368212461473,
      "num_tokens": 50825575.0,
      "step": 25085
    },
    {
      "entropy": 6.025079774856567,
      "epoch": 2.192606833872236,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045174549212867053,
      "loss": 5.348,
      "mean_token_accuracy": 0.19254832565784455,
      "num_tokens": 50836106.0,
      "step": 25090
    },
    {
      "entropy": 5.980520868301392,
      "epoch": 2.193043782224941,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045172620742994904,
      "loss": 5.3367,
      "mean_token_accuracy": 0.19651081562042236,
      "num_tokens": 50847350.0,
      "step": 25095
    },
    {
      "entropy": 5.982596445083618,
      "epoch": 2.1934807305776456,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004517069193415263,
      "loss": 5.1999,
      "mean_token_accuracy": 0.20528587251901625,
      "num_tokens": 50857543.0,
      "step": 25100
    },
    {
      "entropy": 5.9817726612091064,
      "epoch": 2.1939176789303505,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045168762786377214,
      "loss": 5.3062,
      "mean_token_accuracy": 0.2001746967434883,
      "num_tokens": 50868791.0,
      "step": 25105
    },
    {
      "entropy": 5.974800157546997,
      "epoch": 2.194354627283055,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045166833299705686,
      "loss": 5.2655,
      "mean_token_accuracy": 0.20119089782238006,
      "num_tokens": 50879244.0,
      "step": 25110
    },
    {
      "entropy": 5.978437852859497,
      "epoch": 2.19479157563576,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045164903474175026,
      "loss": 5.2779,
      "mean_token_accuracy": 0.1975029617547989,
      "num_tokens": 50888935.0,
      "step": 25115
    },
    {
      "entropy": 5.9883791446685795,
      "epoch": 2.1952285239884644,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045162973309822264,
      "loss": 5.2769,
      "mean_token_accuracy": 0.20147713273763657,
      "num_tokens": 50898152.0,
      "step": 25120
    },
    {
      "entropy": 5.959103155136108,
      "epoch": 2.1956654723411693,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004516104280668442,
      "loss": 5.2723,
      "mean_token_accuracy": 0.19508961588144302,
      "num_tokens": 50908555.0,
      "step": 25125
    },
    {
      "entropy": 5.992620992660522,
      "epoch": 2.196102420693874,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045159111964798514,
      "loss": 5.4084,
      "mean_token_accuracy": 0.18842292577028275,
      "num_tokens": 50919078.0,
      "step": 25130
    },
    {
      "entropy": 5.993170499801636,
      "epoch": 2.1965393690465786,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045157180784201594,
      "loss": 5.2694,
      "mean_token_accuracy": 0.2014564260840416,
      "num_tokens": 50928879.0,
      "step": 25135
    },
    {
      "entropy": 5.984413957595825,
      "epoch": 2.1969763173992836,
      "grad_norm": 1.140625,
      "learning_rate": 0.000451552492649307,
      "loss": 5.291,
      "mean_token_accuracy": 0.20187756270170212,
      "num_tokens": 50939456.0,
      "step": 25140
    },
    {
      "entropy": 5.929342365264892,
      "epoch": 2.197413265751988,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045153317407022866,
      "loss": 5.1932,
      "mean_token_accuracy": 0.2144893392920494,
      "num_tokens": 50948789.0,
      "step": 25145
    },
    {
      "entropy": 5.941885375976563,
      "epoch": 2.197850214104693,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045151385210515153,
      "loss": 5.1986,
      "mean_token_accuracy": 0.20596335232257842,
      "num_tokens": 50958492.0,
      "step": 25150
    },
    {
      "entropy": 5.936833190917969,
      "epoch": 2.1982871624573974,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045149452675444625,
      "loss": 5.2736,
      "mean_token_accuracy": 0.20620289146900178,
      "num_tokens": 50968814.0,
      "step": 25155
    },
    {
      "entropy": 6.000827836990356,
      "epoch": 2.1987241108101023,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045147519801848346,
      "loss": 5.2726,
      "mean_token_accuracy": 0.19740806221961976,
      "num_tokens": 50979680.0,
      "step": 25160
    },
    {
      "entropy": 5.96428599357605,
      "epoch": 2.199161059162807,
      "grad_norm": 1.125,
      "learning_rate": 0.00045145586589763394,
      "loss": 5.3195,
      "mean_token_accuracy": 0.19402784407138823,
      "num_tokens": 50990827.0,
      "step": 25165
    },
    {
      "entropy": 5.932803821563721,
      "epoch": 2.1995980075155117,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004514365303922684,
      "loss": 5.2545,
      "mean_token_accuracy": 0.20123126953840256,
      "num_tokens": 51001065.0,
      "step": 25170
    },
    {
      "entropy": 5.943730354309082,
      "epoch": 2.200034955868216,
      "grad_norm": 1.125,
      "learning_rate": 0.00045141719150275773,
      "loss": 5.2592,
      "mean_token_accuracy": 0.19812546521425248,
      "num_tokens": 51011348.0,
      "step": 25175
    },
    {
      "entropy": 6.011237478256225,
      "epoch": 2.200471904220921,
      "grad_norm": 1.125,
      "learning_rate": 0.00045139784922947296,
      "loss": 5.2518,
      "mean_token_accuracy": 0.19950726330280305,
      "num_tokens": 51022033.0,
      "step": 25180
    },
    {
      "entropy": 5.942474317550659,
      "epoch": 2.200908852573626,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000451378503572785,
      "loss": 5.3049,
      "mean_token_accuracy": 0.19797243922948837,
      "num_tokens": 51032001.0,
      "step": 25185
    },
    {
      "entropy": 5.957555675506592,
      "epoch": 2.2013458009263305,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045135915453306487,
      "loss": 5.3175,
      "mean_token_accuracy": 0.20006451904773712,
      "num_tokens": 51041227.0,
      "step": 25190
    },
    {
      "entropy": 5.872471857070923,
      "epoch": 2.2017827492790354,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045133980211068374,
      "loss": 5.1256,
      "mean_token_accuracy": 0.2215285360813141,
      "num_tokens": 51050642.0,
      "step": 25195
    },
    {
      "entropy": 5.937213706970215,
      "epoch": 2.20221969763174,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004513204463060129,
      "loss": 5.2243,
      "mean_token_accuracy": 0.20632966458797455,
      "num_tokens": 51060646.0,
      "step": 25200
    },
    {
      "entropy": 5.991626787185669,
      "epoch": 2.2026566459844448,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004513010871194234,
      "loss": 5.3451,
      "mean_token_accuracy": 0.20282579958438873,
      "num_tokens": 51071762.0,
      "step": 25205
    },
    {
      "entropy": 5.962434720993042,
      "epoch": 2.2030935943371492,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045128172455128674,
      "loss": 5.2947,
      "mean_token_accuracy": 0.20092957764863967,
      "num_tokens": 51080802.0,
      "step": 25210
    },
    {
      "entropy": 5.933601808547974,
      "epoch": 2.203530542689854,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004512623586019742,
      "loss": 5.2939,
      "mean_token_accuracy": 0.19877205044031143,
      "num_tokens": 51091802.0,
      "step": 25215
    },
    {
      "entropy": 5.972301769256592,
      "epoch": 2.2039674910425586,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004512429892718573,
      "loss": 5.2586,
      "mean_token_accuracy": 0.20733703076839446,
      "num_tokens": 51100939.0,
      "step": 25220
    },
    {
      "entropy": 5.972086334228516,
      "epoch": 2.2044044393952635,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004512236165613074,
      "loss": 5.2344,
      "mean_token_accuracy": 0.20228865146636962,
      "num_tokens": 51110928.0,
      "step": 25225
    },
    {
      "entropy": 5.929602575302124,
      "epoch": 2.204841387747968,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045120424047069633,
      "loss": 5.2658,
      "mean_token_accuracy": 0.19934921115636825,
      "num_tokens": 51121126.0,
      "step": 25230
    },
    {
      "entropy": 5.988209629058838,
      "epoch": 2.205278336100673,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045118486100039553,
      "loss": 5.2099,
      "mean_token_accuracy": 0.2087031215429306,
      "num_tokens": 51130425.0,
      "step": 25235
    },
    {
      "entropy": 5.984927129745484,
      "epoch": 2.205715284453378,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004511654781507768,
      "loss": 5.2486,
      "mean_token_accuracy": 0.203528556227684,
      "num_tokens": 51139994.0,
      "step": 25240
    },
    {
      "entropy": 5.947909688949585,
      "epoch": 2.2061522328060823,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045114609192221196,
      "loss": 5.2956,
      "mean_token_accuracy": 0.197084242105484,
      "num_tokens": 51149762.0,
      "step": 25245
    },
    {
      "entropy": 5.874459457397461,
      "epoch": 2.206589181158787,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004511267023150727,
      "loss": 5.272,
      "mean_token_accuracy": 0.2122286319732666,
      "num_tokens": 51161091.0,
      "step": 25250
    },
    {
      "entropy": 6.008920001983642,
      "epoch": 2.2070261295114917,
      "grad_norm": 1.171875,
      "learning_rate": 0.000451107309329731,
      "loss": 5.2714,
      "mean_token_accuracy": 0.19802107512950898,
      "num_tokens": 51170749.0,
      "step": 25255
    },
    {
      "entropy": 6.00977897644043,
      "epoch": 2.2074630778641966,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045108791296655877,
      "loss": 5.2417,
      "mean_token_accuracy": 0.20310185104608536,
      "num_tokens": 51179515.0,
      "step": 25260
    },
    {
      "entropy": 5.964917945861816,
      "epoch": 2.207900026216901,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045106851322592817,
      "loss": 5.3092,
      "mean_token_accuracy": 0.201337568461895,
      "num_tokens": 51189036.0,
      "step": 25265
    },
    {
      "entropy": 6.003973340988159,
      "epoch": 2.208336974569606,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004510491101082112,
      "loss": 5.304,
      "mean_token_accuracy": 0.1942061096429825,
      "num_tokens": 51199967.0,
      "step": 25270
    },
    {
      "entropy": 5.935159635543823,
      "epoch": 2.2087739229223105,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004510297036137801,
      "loss": 5.2719,
      "mean_token_accuracy": 0.20440662950277327,
      "num_tokens": 51210281.0,
      "step": 25275
    },
    {
      "entropy": 5.973583793640136,
      "epoch": 2.2092108712750154,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045101029374300703,
      "loss": 5.3226,
      "mean_token_accuracy": 0.19986491203308104,
      "num_tokens": 51220209.0,
      "step": 25280
    },
    {
      "entropy": 5.989978265762329,
      "epoch": 2.20964781962772,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004509908804962643,
      "loss": 5.2861,
      "mean_token_accuracy": 0.20197409838438035,
      "num_tokens": 51231108.0,
      "step": 25285
    },
    {
      "entropy": 5.937734460830688,
      "epoch": 2.2100847679804247,
      "grad_norm": 1.125,
      "learning_rate": 0.00045097146387392423,
      "loss": 5.2412,
      "mean_token_accuracy": 0.20406871140003205,
      "num_tokens": 51241316.0,
      "step": 25290
    },
    {
      "entropy": 5.947099256515503,
      "epoch": 2.210521716333129,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004509520438763593,
      "loss": 5.2763,
      "mean_token_accuracy": 0.204260016977787,
      "num_tokens": 51250854.0,
      "step": 25295
    },
    {
      "entropy": 5.936671543121338,
      "epoch": 2.210958664685834,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045093262050394193,
      "loss": 5.2763,
      "mean_token_accuracy": 0.1994234800338745,
      "num_tokens": 51261001.0,
      "step": 25300
    },
    {
      "entropy": 5.947023630142212,
      "epoch": 2.211395613038539,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045091319375704475,
      "loss": 5.3183,
      "mean_token_accuracy": 0.20400564670562743,
      "num_tokens": 51270815.0,
      "step": 25305
    },
    {
      "entropy": 6.017109489440918,
      "epoch": 2.2118325613912435,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045089376363604037,
      "loss": 5.2661,
      "mean_token_accuracy": 0.1989959880709648,
      "num_tokens": 51281370.0,
      "step": 25310
    },
    {
      "entropy": 5.9826744556427,
      "epoch": 2.2122695097439484,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045087433014130143,
      "loss": 5.2731,
      "mean_token_accuracy": 0.2095680132508278,
      "num_tokens": 51290147.0,
      "step": 25315
    },
    {
      "entropy": 6.026210451126099,
      "epoch": 2.212706458096653,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004508548932732006,
      "loss": 5.2073,
      "mean_token_accuracy": 0.21664351224899292,
      "num_tokens": 51299487.0,
      "step": 25320
    },
    {
      "entropy": 5.981407308578492,
      "epoch": 2.213143406449358,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004508354530321108,
      "loss": 5.2636,
      "mean_token_accuracy": 0.20121206045150758,
      "num_tokens": 51309425.0,
      "step": 25325
    },
    {
      "entropy": 5.9259720802307125,
      "epoch": 2.2135803548020623,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004508160094184049,
      "loss": 5.2596,
      "mean_token_accuracy": 0.20437252521514893,
      "num_tokens": 51319962.0,
      "step": 25330
    },
    {
      "entropy": 5.936253309249878,
      "epoch": 2.214017303154767,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004507965624324558,
      "loss": 5.2604,
      "mean_token_accuracy": 0.20731020867824554,
      "num_tokens": 51330892.0,
      "step": 25335
    },
    {
      "entropy": 5.994621324539184,
      "epoch": 2.2144542515074717,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045077711207463654,
      "loss": 5.2857,
      "mean_token_accuracy": 0.20463733077049256,
      "num_tokens": 51342783.0,
      "step": 25340
    },
    {
      "entropy": 6.0160681247711185,
      "epoch": 2.2148911998601766,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045075765834532024,
      "loss": 5.3278,
      "mean_token_accuracy": 0.1967927858233452,
      "num_tokens": 51352890.0,
      "step": 25345
    },
    {
      "entropy": 6.012806367874146,
      "epoch": 2.215328148212881,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004507382012448798,
      "loss": 5.337,
      "mean_token_accuracy": 0.19714941680431367,
      "num_tokens": 51363479.0,
      "step": 25350
    },
    {
      "entropy": 5.967131662368774,
      "epoch": 2.215765096565586,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004507187407736886,
      "loss": 5.2808,
      "mean_token_accuracy": 0.1999945029616356,
      "num_tokens": 51374304.0,
      "step": 25355
    },
    {
      "entropy": 5.935908317565918,
      "epoch": 2.2162020449182904,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004506992769321199,
      "loss": 5.2916,
      "mean_token_accuracy": 0.20296127796173097,
      "num_tokens": 51384649.0,
      "step": 25360
    },
    {
      "entropy": 5.969989967346192,
      "epoch": 2.2166389932709953,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045067980972054707,
      "loss": 5.2678,
      "mean_token_accuracy": 0.20109868049621582,
      "num_tokens": 51394582.0,
      "step": 25365
    },
    {
      "entropy": 5.961095905303955,
      "epoch": 2.2170759416237003,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004506603391393434,
      "loss": 5.2609,
      "mean_token_accuracy": 0.20101435333490372,
      "num_tokens": 51404618.0,
      "step": 25370
    },
    {
      "entropy": 5.966181755065918,
      "epoch": 2.2175128899764047,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004506408651888824,
      "loss": 5.2827,
      "mean_token_accuracy": 0.20010263621807098,
      "num_tokens": 51414810.0,
      "step": 25375
    },
    {
      "entropy": 5.9522467136383055,
      "epoch": 2.2179498383291096,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004506213878695375,
      "loss": 5.3548,
      "mean_token_accuracy": 0.20055031180381774,
      "num_tokens": 51426118.0,
      "step": 25380
    },
    {
      "entropy": 5.989010524749756,
      "epoch": 2.218386786681814,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045060190718168244,
      "loss": 5.2438,
      "mean_token_accuracy": 0.20481134504079818,
      "num_tokens": 51435260.0,
      "step": 25385
    },
    {
      "entropy": 5.959049558639526,
      "epoch": 2.218823735034519,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004505824231256907,
      "loss": 5.2612,
      "mean_token_accuracy": 0.2028967782855034,
      "num_tokens": 51444778.0,
      "step": 25390
    },
    {
      "entropy": 5.940774822235108,
      "epoch": 2.2192606833872235,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004505629357019361,
      "loss": 5.2508,
      "mean_token_accuracy": 0.20684659481048584,
      "num_tokens": 51456030.0,
      "step": 25395
    },
    {
      "entropy": 6.00339527130127,
      "epoch": 2.2196976317399284,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004505434449107925,
      "loss": 5.2692,
      "mean_token_accuracy": 0.19958533346652985,
      "num_tokens": 51465836.0,
      "step": 25400
    },
    {
      "entropy": 5.912712907791137,
      "epoch": 2.220134580092633,
      "grad_norm": 1.125,
      "learning_rate": 0.0004505239507526335,
      "loss": 5.2017,
      "mean_token_accuracy": 0.20970931351184846,
      "num_tokens": 51475330.0,
      "step": 25405
    },
    {
      "entropy": 5.9231010437011715,
      "epoch": 2.220571528445338,
      "grad_norm": 1.125,
      "learning_rate": 0.00045050445322783326,
      "loss": 5.2323,
      "mean_token_accuracy": 0.20544561743736267,
      "num_tokens": 51486646.0,
      "step": 25410
    },
    {
      "entropy": 5.931829500198364,
      "epoch": 2.2210084767980423,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004504849523367656,
      "loss": 5.2531,
      "mean_token_accuracy": 0.20416360944509507,
      "num_tokens": 51496878.0,
      "step": 25415
    },
    {
      "entropy": 5.991544961929321,
      "epoch": 2.221445425150747,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004504654480798046,
      "loss": 5.3531,
      "mean_token_accuracy": 0.18977407962083817,
      "num_tokens": 51507092.0,
      "step": 25420
    },
    {
      "entropy": 5.961033916473388,
      "epoch": 2.221882373503452,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004504459404573243,
      "loss": 5.282,
      "mean_token_accuracy": 0.20402196496725084,
      "num_tokens": 51517521.0,
      "step": 25425
    },
    {
      "entropy": 5.95480432510376,
      "epoch": 2.2223193218561565,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004504264294696991,
      "loss": 5.325,
      "mean_token_accuracy": 0.19789156317710876,
      "num_tokens": 51527902.0,
      "step": 25430
    },
    {
      "entropy": 5.926574039459228,
      "epoch": 2.2227562702088615,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004504069151173029,
      "loss": 5.278,
      "mean_token_accuracy": 0.20393697321414947,
      "num_tokens": 51539062.0,
      "step": 25435
    },
    {
      "entropy": 5.943600463867187,
      "epoch": 2.223193218561566,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004503873974005102,
      "loss": 5.1853,
      "mean_token_accuracy": 0.21584951281547546,
      "num_tokens": 51548938.0,
      "step": 25440
    },
    {
      "entropy": 5.946793413162231,
      "epoch": 2.223630166914271,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004503678763196953,
      "loss": 5.2936,
      "mean_token_accuracy": 0.20007272213697433,
      "num_tokens": 51559330.0,
      "step": 25445
    },
    {
      "entropy": 5.946765470504761,
      "epoch": 2.2240671152669753,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004503483518752327,
      "loss": 5.2812,
      "mean_token_accuracy": 0.201000215113163,
      "num_tokens": 51569617.0,
      "step": 25450
    },
    {
      "entropy": 5.924801063537598,
      "epoch": 2.2245040636196802,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004503288240674968,
      "loss": 5.2507,
      "mean_token_accuracy": 0.20159258097410201,
      "num_tokens": 51579545.0,
      "step": 25455
    },
    {
      "entropy": 5.9963596820831295,
      "epoch": 2.2249410119723847,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004503092928968622,
      "loss": 5.264,
      "mean_token_accuracy": 0.1979052886366844,
      "num_tokens": 51589375.0,
      "step": 25460
    },
    {
      "entropy": 5.96301679611206,
      "epoch": 2.2253779603250896,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004502897583637035,
      "loss": 5.2569,
      "mean_token_accuracy": 0.20806915313005447,
      "num_tokens": 51600586.0,
      "step": 25465
    },
    {
      "entropy": 5.939322471618652,
      "epoch": 2.225814908677794,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045027022046839534,
      "loss": 5.2035,
      "mean_token_accuracy": 0.2023863285779953,
      "num_tokens": 51609568.0,
      "step": 25470
    },
    {
      "entropy": 5.94833345413208,
      "epoch": 2.226251857030499,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004502506792113126,
      "loss": 5.2211,
      "mean_token_accuracy": 0.21028080582618713,
      "num_tokens": 51619450.0,
      "step": 25475
    },
    {
      "entropy": 5.922131633758545,
      "epoch": 2.226688805383204,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004502311345928299,
      "loss": 5.2353,
      "mean_token_accuracy": 0.2011454612016678,
      "num_tokens": 51630528.0,
      "step": 25480
    },
    {
      "entropy": 5.926286745071411,
      "epoch": 2.2271257537359084,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045021158661332227,
      "loss": 5.2577,
      "mean_token_accuracy": 0.20690271705389024,
      "num_tokens": 51640204.0,
      "step": 25485
    },
    {
      "entropy": 6.013183450698852,
      "epoch": 2.2275627020886133,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004501920352731646,
      "loss": 5.411,
      "mean_token_accuracy": 0.1962601587176323,
      "num_tokens": 51649531.0,
      "step": 25490
    },
    {
      "entropy": 6.034527730941773,
      "epoch": 2.2279996504413178,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004501724805727319,
      "loss": 5.3452,
      "mean_token_accuracy": 0.19172675162553787,
      "num_tokens": 51660943.0,
      "step": 25495
    },
    {
      "entropy": 5.950950717926025,
      "epoch": 2.2284365987940227,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004501529225123992,
      "loss": 5.1941,
      "mean_token_accuracy": 0.20861488282680513,
      "num_tokens": 51670606.0,
      "step": 25500
    },
    {
      "entropy": 5.936872816085815,
      "epoch": 2.228873547146727,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045013336109254174,
      "loss": 5.2591,
      "mean_token_accuracy": 0.2052227646112442,
      "num_tokens": 51681634.0,
      "step": 25505
    },
    {
      "entropy": 5.972455883026123,
      "epoch": 2.229310495499432,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045011379631353457,
      "loss": 5.2886,
      "mean_token_accuracy": 0.20161950290203096,
      "num_tokens": 51691579.0,
      "step": 25510
    },
    {
      "entropy": 5.923768329620361,
      "epoch": 2.2297474438521365,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045009422817575304,
      "loss": 5.1854,
      "mean_token_accuracy": 0.21678960025310517,
      "num_tokens": 51700739.0,
      "step": 25515
    },
    {
      "entropy": 5.975208377838134,
      "epoch": 2.2301843922048414,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004500746566795725,
      "loss": 5.2894,
      "mean_token_accuracy": 0.2041110873222351,
      "num_tokens": 51711905.0,
      "step": 25520
    },
    {
      "entropy": 5.961949825286865,
      "epoch": 2.230621340557546,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045005508182536824,
      "loss": 5.2116,
      "mean_token_accuracy": 0.20604220628738404,
      "num_tokens": 51722020.0,
      "step": 25525
    },
    {
      "entropy": 5.972663116455078,
      "epoch": 2.231058288910251,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004500355036135158,
      "loss": 5.2989,
      "mean_token_accuracy": 0.20368176698684692,
      "num_tokens": 51731763.0,
      "step": 25530
    },
    {
      "entropy": 5.991360330581665,
      "epoch": 2.2314952372629557,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00045001592204439077,
      "loss": 5.3216,
      "mean_token_accuracy": 0.19664137959480285,
      "num_tokens": 51741425.0,
      "step": 25535
    },
    {
      "entropy": 5.99380898475647,
      "epoch": 2.23193218561566,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004499963371183686,
      "loss": 5.3033,
      "mean_token_accuracy": 0.1968265563249588,
      "num_tokens": 51751788.0,
      "step": 25540
    },
    {
      "entropy": 5.961472034454346,
      "epoch": 2.232369133968365,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044997674883582494,
      "loss": 5.2963,
      "mean_token_accuracy": 0.20033468008041383,
      "num_tokens": 51763205.0,
      "step": 25545
    },
    {
      "entropy": 6.079153680801392,
      "epoch": 2.2328060823210696,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004499571571971356,
      "loss": 5.2648,
      "mean_token_accuracy": 0.2024006024003029,
      "num_tokens": 51772817.0,
      "step": 25550
    },
    {
      "entropy": 6.0233011722564695,
      "epoch": 2.2332430306737745,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004499375622026763,
      "loss": 5.314,
      "mean_token_accuracy": 0.20165085196495056,
      "num_tokens": 51781596.0,
      "step": 25555
    },
    {
      "entropy": 5.894614267349243,
      "epoch": 2.233679979026479,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004499179638528229,
      "loss": 5.2218,
      "mean_token_accuracy": 0.21086497157812117,
      "num_tokens": 51792452.0,
      "step": 25560
    },
    {
      "entropy": 5.9758669376373295,
      "epoch": 2.234116927379184,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004498983621479513,
      "loss": 5.3145,
      "mean_token_accuracy": 0.19721601754426957,
      "num_tokens": 51802554.0,
      "step": 25565
    },
    {
      "entropy": 5.994680500030517,
      "epoch": 2.2345538757318884,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004498787570884374,
      "loss": 5.2849,
      "mean_token_accuracy": 0.19725495278835298,
      "num_tokens": 51812855.0,
      "step": 25570
    },
    {
      "entropy": 5.991771221160889,
      "epoch": 2.2349908240845933,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004498591486746573,
      "loss": 5.3002,
      "mean_token_accuracy": 0.20457009971141815,
      "num_tokens": 51822306.0,
      "step": 25575
    },
    {
      "entropy": 6.021661233901978,
      "epoch": 2.2354277724372977,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044983953690698724,
      "loss": 5.317,
      "mean_token_accuracy": 0.20384539365768434,
      "num_tokens": 51832209.0,
      "step": 25580
    },
    {
      "entropy": 5.9656072616577145,
      "epoch": 2.2358647207900026,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004498199217858032,
      "loss": 5.2916,
      "mean_token_accuracy": 0.20220282971858977,
      "num_tokens": 51841794.0,
      "step": 25585
    },
    {
      "entropy": 5.960112428665161,
      "epoch": 2.2363016691427076,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044980030331148144,
      "loss": 5.2995,
      "mean_token_accuracy": 0.19434671998023986,
      "num_tokens": 51851973.0,
      "step": 25590
    },
    {
      "entropy": 5.983524894714355,
      "epoch": 2.236738617495412,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044978068148439826,
      "loss": 5.3253,
      "mean_token_accuracy": 0.19407177120447158,
      "num_tokens": 51862173.0,
      "step": 25595
    },
    {
      "entropy": 5.990093231201172,
      "epoch": 2.237175565848117,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044976105630493,
      "loss": 5.2956,
      "mean_token_accuracy": 0.2044089615345001,
      "num_tokens": 51872960.0,
      "step": 25600
    },
    {
      "entropy": 5.926129150390625,
      "epoch": 2.2376125142008214,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004497414277734532,
      "loss": 5.255,
      "mean_token_accuracy": 0.19989117681980134,
      "num_tokens": 51883911.0,
      "step": 25605
    },
    {
      "entropy": 5.9796287536621096,
      "epoch": 2.2380494625535263,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044972179589034407,
      "loss": 5.2486,
      "mean_token_accuracy": 0.20566845387220384,
      "num_tokens": 51893982.0,
      "step": 25610
    },
    {
      "entropy": 5.949384260177612,
      "epoch": 2.238486410906231,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00044970216065597947,
      "loss": 5.1713,
      "mean_token_accuracy": 0.21320097744464875,
      "num_tokens": 51903808.0,
      "step": 25615
    },
    {
      "entropy": 5.8847410678863525,
      "epoch": 2.2389233592589357,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044968252207073584,
      "loss": 5.2071,
      "mean_token_accuracy": 0.2106170579791069,
      "num_tokens": 51912918.0,
      "step": 25620
    },
    {
      "entropy": 5.934665060043335,
      "epoch": 2.23936030761164,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044966288013499,
      "loss": 5.3035,
      "mean_token_accuracy": 0.20222689360380172,
      "num_tokens": 51923894.0,
      "step": 25625
    },
    {
      "entropy": 5.963602876663208,
      "epoch": 2.239797255964345,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004496432348491185,
      "loss": 5.3052,
      "mean_token_accuracy": 0.2004497990012169,
      "num_tokens": 51934569.0,
      "step": 25630
    },
    {
      "entropy": 5.978693723678589,
      "epoch": 2.2402342043170496,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004496235862134982,
      "loss": 5.2808,
      "mean_token_accuracy": 0.19691299945116042,
      "num_tokens": 51944678.0,
      "step": 25635
    },
    {
      "entropy": 6.019806241989135,
      "epoch": 2.2406711526697545,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044960393422850616,
      "loss": 5.3724,
      "mean_token_accuracy": 0.18955983519554137,
      "num_tokens": 51954785.0,
      "step": 25640
    },
    {
      "entropy": 5.939334297180176,
      "epoch": 2.241108101022459,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004495842788945191,
      "loss": 5.2131,
      "mean_token_accuracy": 0.2047238677740097,
      "num_tokens": 51964427.0,
      "step": 25645
    },
    {
      "entropy": 6.011139345169068,
      "epoch": 2.241545049375164,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044956462021191406,
      "loss": 5.432,
      "mean_token_accuracy": 0.19264955669641495,
      "num_tokens": 51974367.0,
      "step": 25650
    },
    {
      "entropy": 5.9756584644317625,
      "epoch": 2.2419819977278688,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044954495818106814,
      "loss": 5.2373,
      "mean_token_accuracy": 0.2033540651202202,
      "num_tokens": 51984935.0,
      "step": 25655
    },
    {
      "entropy": 5.970209312438965,
      "epoch": 2.2424189460805732,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004495252928023585,
      "loss": 5.2605,
      "mean_token_accuracy": 0.20725894272327422,
      "num_tokens": 51995132.0,
      "step": 25660
    },
    {
      "entropy": 5.972576093673706,
      "epoch": 2.242855894433278,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004495056240761623,
      "loss": 5.3223,
      "mean_token_accuracy": 0.19889223873615264,
      "num_tokens": 52005828.0,
      "step": 25665
    },
    {
      "entropy": 5.909350156784058,
      "epoch": 2.2432928427859826,
      "grad_norm": 1.25,
      "learning_rate": 0.00044948595200285677,
      "loss": 5.1564,
      "mean_token_accuracy": 0.2091992974281311,
      "num_tokens": 52015381.0,
      "step": 25670
    },
    {
      "entropy": 5.9646259307861325,
      "epoch": 2.2437297911386875,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044946627658281924,
      "loss": 5.3127,
      "mean_token_accuracy": 0.2008615806698799,
      "num_tokens": 52025699.0,
      "step": 25675
    },
    {
      "entropy": 5.989864301681519,
      "epoch": 2.244166739491392,
      "grad_norm": 1.125,
      "learning_rate": 0.0004494465978164271,
      "loss": 5.2573,
      "mean_token_accuracy": 0.20728893131017684,
      "num_tokens": 52035903.0,
      "step": 25680
    },
    {
      "entropy": 5.972398853302002,
      "epoch": 2.244603687844097,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004494269157040579,
      "loss": 5.2195,
      "mean_token_accuracy": 0.20863011330366135,
      "num_tokens": 52046039.0,
      "step": 25685
    },
    {
      "entropy": 6.025982332229614,
      "epoch": 2.2450406361968014,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044940723024608893,
      "loss": 5.3093,
      "mean_token_accuracy": 0.19498204290866852,
      "num_tokens": 52057146.0,
      "step": 25690
    },
    {
      "entropy": 5.936225271224975,
      "epoch": 2.2454775845495063,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044938754144289793,
      "loss": 5.2802,
      "mean_token_accuracy": 0.20351504236459733,
      "num_tokens": 52067355.0,
      "step": 25695
    },
    {
      "entropy": 5.985464334487915,
      "epoch": 2.2459145329022108,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044936784929486257,
      "loss": 5.2686,
      "mean_token_accuracy": 0.20145623981952668,
      "num_tokens": 52077062.0,
      "step": 25700
    },
    {
      "entropy": 5.943861722946167,
      "epoch": 2.2463514812549157,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004493481538023604,
      "loss": 5.2145,
      "mean_token_accuracy": 0.20596496313810347,
      "num_tokens": 52087918.0,
      "step": 25705
    },
    {
      "entropy": 5.927388143539429,
      "epoch": 2.24678842960762,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044932845496576936,
      "loss": 5.2714,
      "mean_token_accuracy": 0.19880568832159043,
      "num_tokens": 52097283.0,
      "step": 25710
    },
    {
      "entropy": 5.96006588935852,
      "epoch": 2.247225377960325,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004493087527854672,
      "loss": 5.3091,
      "mean_token_accuracy": 0.20219754725694655,
      "num_tokens": 52109037.0,
      "step": 25715
    },
    {
      "entropy": 5.988959693908692,
      "epoch": 2.24766232631303,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004492890472618318,
      "loss": 5.2703,
      "mean_token_accuracy": 0.20435430705547333,
      "num_tokens": 52119809.0,
      "step": 25720
    },
    {
      "entropy": 5.938345003128052,
      "epoch": 2.2480992746657344,
      "grad_norm": 1.125,
      "learning_rate": 0.00044926933839524116,
      "loss": 5.2149,
      "mean_token_accuracy": 0.20653061121702193,
      "num_tokens": 52129641.0,
      "step": 25725
    },
    {
      "entropy": 5.919738817214966,
      "epoch": 2.2485362230184394,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044924962618607324,
      "loss": 5.2268,
      "mean_token_accuracy": 0.20832734704017639,
      "num_tokens": 52139701.0,
      "step": 25730
    },
    {
      "entropy": 5.9564502239227295,
      "epoch": 2.248973171371144,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004492299106347062,
      "loss": 5.2115,
      "mean_token_accuracy": 0.21296195685863495,
      "num_tokens": 52148402.0,
      "step": 25735
    },
    {
      "entropy": 5.934143781661987,
      "epoch": 2.2494101197238487,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004492101917415183,
      "loss": 5.2569,
      "mean_token_accuracy": 0.19974330216646194,
      "num_tokens": 52159238.0,
      "step": 25740
    },
    {
      "entropy": 5.958623933792114,
      "epoch": 2.249847068076553,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00044919046950688744,
      "loss": 5.1974,
      "mean_token_accuracy": 0.21040650010108947,
      "num_tokens": 52169172.0,
      "step": 25745
    },
    {
      "entropy": 5.91507134437561,
      "epoch": 2.250284016429258,
      "grad_norm": 1.125,
      "learning_rate": 0.00044917074393119224,
      "loss": 5.2253,
      "mean_token_accuracy": 0.2064708426594734,
      "num_tokens": 52179432.0,
      "step": 25750
    },
    {
      "entropy": 5.93660249710083,
      "epoch": 2.2507209647819626,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044915101501481086,
      "loss": 5.2775,
      "mean_token_accuracy": 0.20357664227485656,
      "num_tokens": 52190939.0,
      "step": 25755
    },
    {
      "entropy": 5.925822687149048,
      "epoch": 2.2511579131346675,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044913128275812184,
      "loss": 5.2215,
      "mean_token_accuracy": 0.2034524440765381,
      "num_tokens": 52201644.0,
      "step": 25760
    },
    {
      "entropy": 5.98342719078064,
      "epoch": 2.251594861487372,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004491115471615035,
      "loss": 5.2289,
      "mean_token_accuracy": 0.20492924898862838,
      "num_tokens": 52211353.0,
      "step": 25765
    },
    {
      "entropy": 6.00876407623291,
      "epoch": 2.252031809840077,
      "grad_norm": 1.125,
      "learning_rate": 0.0004490918082253344,
      "loss": 5.3229,
      "mean_token_accuracy": 0.19617295563220977,
      "num_tokens": 52221717.0,
      "step": 25770
    },
    {
      "entropy": 5.9590418338775635,
      "epoch": 2.252468758192782,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044907206594999325,
      "loss": 5.3134,
      "mean_token_accuracy": 0.20222099274396896,
      "num_tokens": 52232419.0,
      "step": 25775
    },
    {
      "entropy": 5.975111961364746,
      "epoch": 2.2529057065454863,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044905232033585867,
      "loss": 5.2944,
      "mean_token_accuracy": 0.20094711184501649,
      "num_tokens": 52242386.0,
      "step": 25780
    },
    {
      "entropy": 5.986775827407837,
      "epoch": 2.253342654898191,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004490325713833094,
      "loss": 5.2838,
      "mean_token_accuracy": 0.1993423581123352,
      "num_tokens": 52252569.0,
      "step": 25785
    },
    {
      "entropy": 5.9099565029144285,
      "epoch": 2.2537796032508957,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004490128190927242,
      "loss": 5.1611,
      "mean_token_accuracy": 0.21670355051755905,
      "num_tokens": 52263101.0,
      "step": 25790
    },
    {
      "entropy": 5.956127071380616,
      "epoch": 2.2542165516036006,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044899306346448195,
      "loss": 5.2556,
      "mean_token_accuracy": 0.20383934378623964,
      "num_tokens": 52273250.0,
      "step": 25795
    },
    {
      "entropy": 5.9884134292602536,
      "epoch": 2.254653499956305,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004489733044989616,
      "loss": 5.3626,
      "mean_token_accuracy": 0.1958672970533371,
      "num_tokens": 52284351.0,
      "step": 25800
    },
    {
      "entropy": 5.925188827514648,
      "epoch": 2.25509044830901,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044895354219654204,
      "loss": 5.24,
      "mean_token_accuracy": 0.21261796057224275,
      "num_tokens": 52294642.0,
      "step": 25805
    },
    {
      "entropy": 5.9216633319854735,
      "epoch": 2.2555273966617144,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044893377655760245,
      "loss": 5.1842,
      "mean_token_accuracy": 0.20238350480794906,
      "num_tokens": 52303793.0,
      "step": 25810
    },
    {
      "entropy": 5.934630441665649,
      "epoch": 2.2559643450144193,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044891400758252187,
      "loss": 5.2876,
      "mean_token_accuracy": 0.1964072987437248,
      "num_tokens": 52313541.0,
      "step": 25815
    },
    {
      "entropy": 6.02141637802124,
      "epoch": 2.256401293367124,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004488942352716795,
      "loss": 5.281,
      "mean_token_accuracy": 0.19888078719377517,
      "num_tokens": 52325428.0,
      "step": 25820
    },
    {
      "entropy": 5.965984201431274,
      "epoch": 2.2568382417198287,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004488744596254545,
      "loss": 5.2942,
      "mean_token_accuracy": 0.20143131166696548,
      "num_tokens": 52335837.0,
      "step": 25825
    },
    {
      "entropy": 5.902680492401123,
      "epoch": 2.2572751900725336,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044885468064422636,
      "loss": 5.2503,
      "mean_token_accuracy": 0.20170735716819763,
      "num_tokens": 52346925.0,
      "step": 25830
    },
    {
      "entropy": 5.982629489898682,
      "epoch": 2.257712138425238,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004488348983283743,
      "loss": 5.3744,
      "mean_token_accuracy": 0.19373282343149184,
      "num_tokens": 52357244.0,
      "step": 25835
    },
    {
      "entropy": 5.968039417266846,
      "epoch": 2.258149086777943,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004488151126782779,
      "loss": 5.2474,
      "mean_token_accuracy": 0.20538482815027237,
      "num_tokens": 52367657.0,
      "step": 25840
    },
    {
      "entropy": 6.013935852050781,
      "epoch": 2.2585860351306475,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044879532369431635,
      "loss": 5.2924,
      "mean_token_accuracy": 0.19268314391374589,
      "num_tokens": 52377298.0,
      "step": 25845
    },
    {
      "entropy": 5.999974584579467,
      "epoch": 2.2590229834833524,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044877553137686953,
      "loss": 5.3242,
      "mean_token_accuracy": 0.2006144031882286,
      "num_tokens": 52387829.0,
      "step": 25850
    },
    {
      "entropy": 5.9491363048553465,
      "epoch": 2.259459931836057,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044875573572631706,
      "loss": 5.2713,
      "mean_token_accuracy": 0.20309037268161773,
      "num_tokens": 52398214.0,
      "step": 25855
    },
    {
      "entropy": 5.940363836288452,
      "epoch": 2.259896880188762,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004487359367430384,
      "loss": 5.2364,
      "mean_token_accuracy": 0.20308595299720764,
      "num_tokens": 52407619.0,
      "step": 25860
    },
    {
      "entropy": 5.963848114013672,
      "epoch": 2.2603338285414663,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004487161344274135,
      "loss": 5.2813,
      "mean_token_accuracy": 0.1978464901447296,
      "num_tokens": 52417465.0,
      "step": 25865
    },
    {
      "entropy": 5.9248316287994385,
      "epoch": 2.260770776894171,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004486963287798221,
      "loss": 5.2368,
      "mean_token_accuracy": 0.20848171412944794,
      "num_tokens": 52427000.0,
      "step": 25870
    },
    {
      "entropy": 5.976263093948364,
      "epoch": 2.2612077252468756,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004486765198006441,
      "loss": 5.2883,
      "mean_token_accuracy": 0.19662030190229415,
      "num_tokens": 52437449.0,
      "step": 25875
    },
    {
      "entropy": 6.052509117126465,
      "epoch": 2.2616446735995805,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044865670749025943,
      "loss": 5.3243,
      "mean_token_accuracy": 0.1989920437335968,
      "num_tokens": 52446946.0,
      "step": 25880
    },
    {
      "entropy": 5.917799615859986,
      "epoch": 2.2620816219522855,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044863689184904815,
      "loss": 5.2395,
      "mean_token_accuracy": 0.20357971489429474,
      "num_tokens": 52456993.0,
      "step": 25885
    },
    {
      "entropy": 5.992791175842285,
      "epoch": 2.26251857030499,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004486170728773903,
      "loss": 5.3054,
      "mean_token_accuracy": 0.19467682838439943,
      "num_tokens": 52467999.0,
      "step": 25890
    },
    {
      "entropy": 6.025313234329223,
      "epoch": 2.262955518657695,
      "grad_norm": 1.09375,
      "learning_rate": 0.000448597250575666,
      "loss": 5.3444,
      "mean_token_accuracy": 0.19287839531898499,
      "num_tokens": 52479556.0,
      "step": 25895
    },
    {
      "entropy": 6.042495918273926,
      "epoch": 2.2633924670103993,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004485774249442555,
      "loss": 5.2872,
      "mean_token_accuracy": 0.20390803664922713,
      "num_tokens": 52489160.0,
      "step": 25900
    },
    {
      "entropy": 5.951699542999267,
      "epoch": 2.2638294153631042,
      "grad_norm": 1.125,
      "learning_rate": 0.000448557595983539,
      "loss": 5.2417,
      "mean_token_accuracy": 0.20042531788349152,
      "num_tokens": 52500102.0,
      "step": 25905
    },
    {
      "entropy": 5.943880224227906,
      "epoch": 2.2642663637158087,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044853776369389685,
      "loss": 5.2474,
      "mean_token_accuracy": 0.2049190253019333,
      "num_tokens": 52509615.0,
      "step": 25910
    },
    {
      "entropy": 5.940184116363525,
      "epoch": 2.2647033120685136,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004485179280757095,
      "loss": 5.249,
      "mean_token_accuracy": 0.21039193421602248,
      "num_tokens": 52520796.0,
      "step": 25915
    },
    {
      "entropy": 5.9450620174407955,
      "epoch": 2.265140260421218,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044849808912935735,
      "loss": 5.2289,
      "mean_token_accuracy": 0.20482457280158997,
      "num_tokens": 52530641.0,
      "step": 25920
    },
    {
      "entropy": 5.944496965408325,
      "epoch": 2.265577208773923,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000448478246855221,
      "loss": 5.2951,
      "mean_token_accuracy": 0.20551470816135406,
      "num_tokens": 52541626.0,
      "step": 25925
    },
    {
      "entropy": 5.973793935775757,
      "epoch": 2.2660141571266275,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004484584012536809,
      "loss": 5.3222,
      "mean_token_accuracy": 0.20108033418655397,
      "num_tokens": 52552301.0,
      "step": 25930
    },
    {
      "entropy": 5.903548765182495,
      "epoch": 2.2664511054793324,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004484385523251177,
      "loss": 5.1969,
      "mean_token_accuracy": 0.21511714458465575,
      "num_tokens": 52562463.0,
      "step": 25935
    },
    {
      "entropy": 6.011064672470093,
      "epoch": 2.2668880538320373,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004484187000699123,
      "loss": 5.3105,
      "mean_token_accuracy": 0.19893949031829833,
      "num_tokens": 52572782.0,
      "step": 25940
    },
    {
      "entropy": 5.968219566345215,
      "epoch": 2.2673250021847418,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044839884448844534,
      "loss": 5.2421,
      "mean_token_accuracy": 0.20736289918422698,
      "num_tokens": 52582293.0,
      "step": 25945
    },
    {
      "entropy": 5.967358684539795,
      "epoch": 2.2677619505374462,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004483789855810977,
      "loss": 5.276,
      "mean_token_accuracy": 0.2019757568836212,
      "num_tokens": 52591950.0,
      "step": 25950
    },
    {
      "entropy": 5.929618406295776,
      "epoch": 2.268198898890151,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044835912334825024,
      "loss": 5.2823,
      "mean_token_accuracy": 0.20258513391017913,
      "num_tokens": 52603998.0,
      "step": 25955
    },
    {
      "entropy": 5.999755620956421,
      "epoch": 2.268635847242856,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044833925779028393,
      "loss": 5.2935,
      "mean_token_accuracy": 0.19620541036128997,
      "num_tokens": 52614955.0,
      "step": 25960
    },
    {
      "entropy": 6.037349653244019,
      "epoch": 2.2690727955955605,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004483193889075799,
      "loss": 5.2992,
      "mean_token_accuracy": 0.19707612097263336,
      "num_tokens": 52625305.0,
      "step": 25965
    },
    {
      "entropy": 5.954633092880249,
      "epoch": 2.2695097439482654,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004482995167005191,
      "loss": 5.2032,
      "mean_token_accuracy": 0.207632939517498,
      "num_tokens": 52635310.0,
      "step": 25970
    },
    {
      "entropy": 5.964901924133301,
      "epoch": 2.26994669230097,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044827964116948284,
      "loss": 5.2851,
      "mean_token_accuracy": 0.20242640525102615,
      "num_tokens": 52644946.0,
      "step": 25975
    },
    {
      "entropy": 5.99201226234436,
      "epoch": 2.270383640653675,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004482597623148522,
      "loss": 5.2543,
      "mean_token_accuracy": 0.1999029263854027,
      "num_tokens": 52655056.0,
      "step": 25980
    },
    {
      "entropy": 5.976499938964844,
      "epoch": 2.2708205890063793,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004482398801370085,
      "loss": 5.3032,
      "mean_token_accuracy": 0.19749280512332917,
      "num_tokens": 52665423.0,
      "step": 25985
    },
    {
      "entropy": 5.997645092010498,
      "epoch": 2.271257537359084,
      "grad_norm": 1.125,
      "learning_rate": 0.00044821999463633323,
      "loss": 5.309,
      "mean_token_accuracy": 0.1981087803840637,
      "num_tokens": 52675300.0,
      "step": 25990
    },
    {
      "entropy": 6.019380521774292,
      "epoch": 2.271694485711789,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044820010581320764,
      "loss": 5.2779,
      "mean_token_accuracy": 0.20200059562921524,
      "num_tokens": 52685431.0,
      "step": 25995
    },
    {
      "entropy": 5.976389360427857,
      "epoch": 2.2721314340644936,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044818021366801325,
      "loss": 5.3697,
      "mean_token_accuracy": 0.1955282211303711,
      "num_tokens": 52695756.0,
      "step": 26000
    },
    {
      "entropy": 5.967243814468384,
      "epoch": 2.272568382417198,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004481603182011317,
      "loss": 5.2318,
      "mean_token_accuracy": 0.20520973056554795,
      "num_tokens": 52704889.0,
      "step": 26005
    },
    {
      "entropy": 5.982651138305664,
      "epoch": 2.273005330769903,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044814041941294435,
      "loss": 5.2024,
      "mean_token_accuracy": 0.2032284051179886,
      "num_tokens": 52714227.0,
      "step": 26010
    },
    {
      "entropy": 5.882278299331665,
      "epoch": 2.273442279122608,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044812051730383316,
      "loss": 5.2027,
      "mean_token_accuracy": 0.20747416466474533,
      "num_tokens": 52725150.0,
      "step": 26015
    },
    {
      "entropy": 5.959783840179443,
      "epoch": 2.2738792274753123,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004481006118741798,
      "loss": 5.2359,
      "mean_token_accuracy": 0.20293408781290054,
      "num_tokens": 52734825.0,
      "step": 26020
    },
    {
      "entropy": 5.987322092056274,
      "epoch": 2.2743161758280173,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004480807031243658,
      "loss": 5.2357,
      "mean_token_accuracy": 0.21368785053491593,
      "num_tokens": 52745185.0,
      "step": 26025
    },
    {
      "entropy": 5.953555679321289,
      "epoch": 2.2747531241807217,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044806079105477335,
      "loss": 5.3156,
      "mean_token_accuracy": 0.2015618398785591,
      "num_tokens": 52755122.0,
      "step": 26030
    },
    {
      "entropy": 5.957521581649781,
      "epoch": 2.2751900725334266,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004480408756657842,
      "loss": 5.2836,
      "mean_token_accuracy": 0.1962753266096115,
      "num_tokens": 52765338.0,
      "step": 26035
    },
    {
      "entropy": 6.039827156066894,
      "epoch": 2.275627020886131,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044802095695778047,
      "loss": 5.3577,
      "mean_token_accuracy": 0.19219847172498702,
      "num_tokens": 52775045.0,
      "step": 26040
    },
    {
      "entropy": 5.973278379440307,
      "epoch": 2.276063969238836,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044800103493114403,
      "loss": 5.2908,
      "mean_token_accuracy": 0.19993473440408707,
      "num_tokens": 52785942.0,
      "step": 26045
    },
    {
      "entropy": 5.890877151489258,
      "epoch": 2.2765009175915405,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004479811095862571,
      "loss": 5.2259,
      "mean_token_accuracy": 0.2128657341003418,
      "num_tokens": 52794905.0,
      "step": 26050
    },
    {
      "entropy": 5.9682270050048825,
      "epoch": 2.2769378659442454,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004479611809235019,
      "loss": 5.1827,
      "mean_token_accuracy": 0.20734343975782393,
      "num_tokens": 52804091.0,
      "step": 26055
    },
    {
      "entropy": 5.998523139953614,
      "epoch": 2.27737481429695,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004479412489432605,
      "loss": 5.2968,
      "mean_token_accuracy": 0.19726764112710954,
      "num_tokens": 52814400.0,
      "step": 26060
    },
    {
      "entropy": 5.9534868717193605,
      "epoch": 2.277811762649655,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044792131364591544,
      "loss": 5.2824,
      "mean_token_accuracy": 0.20354770570993425,
      "num_tokens": 52824423.0,
      "step": 26065
    },
    {
      "entropy": 5.948024272918701,
      "epoch": 2.2782487110023597,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044790137503184895,
      "loss": 5.2811,
      "mean_token_accuracy": 0.20399956554174423,
      "num_tokens": 52833852.0,
      "step": 26070
    },
    {
      "entropy": 5.965526390075683,
      "epoch": 2.278685659355064,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004478814331014435,
      "loss": 5.2856,
      "mean_token_accuracy": 0.20133956372737885,
      "num_tokens": 52844759.0,
      "step": 26075
    },
    {
      "entropy": 5.9467058181762695,
      "epoch": 2.279122607707769,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004478614878550815,
      "loss": 5.2546,
      "mean_token_accuracy": 0.20024718791246415,
      "num_tokens": 52854302.0,
      "step": 26080
    },
    {
      "entropy": 5.975992488861084,
      "epoch": 2.2795595560604736,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004478415392931457,
      "loss": 5.2704,
      "mean_token_accuracy": 0.20082674622535707,
      "num_tokens": 52864086.0,
      "step": 26085
    },
    {
      "entropy": 5.95467438697815,
      "epoch": 2.2799965044131785,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044782158741601854,
      "loss": 5.1933,
      "mean_token_accuracy": 0.2040723368525505,
      "num_tokens": 52874894.0,
      "step": 26090
    },
    {
      "entropy": 5.874599504470825,
      "epoch": 2.280433452765883,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044780163222408277,
      "loss": 5.2517,
      "mean_token_accuracy": 0.20668382495641707,
      "num_tokens": 52885418.0,
      "step": 26095
    },
    {
      "entropy": 5.998873329162597,
      "epoch": 2.280870401118588,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044778167371772117,
      "loss": 5.3142,
      "mean_token_accuracy": 0.1982680380344391,
      "num_tokens": 52895515.0,
      "step": 26100
    },
    {
      "entropy": 5.951717233657837,
      "epoch": 2.2813073494712923,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004477617118973165,
      "loss": 5.2531,
      "mean_token_accuracy": 0.20970509499311446,
      "num_tokens": 52905599.0,
      "step": 26105
    },
    {
      "entropy": 6.015591382980347,
      "epoch": 2.2817442978239972,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004477417467632517,
      "loss": 5.3638,
      "mean_token_accuracy": 0.191654035449028,
      "num_tokens": 52916599.0,
      "step": 26110
    },
    {
      "entropy": 5.9435934066772464,
      "epoch": 2.2821812461767017,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044772177831590965,
      "loss": 5.2451,
      "mean_token_accuracy": 0.21240501403808593,
      "num_tokens": 52928131.0,
      "step": 26115
    },
    {
      "entropy": 5.96757664680481,
      "epoch": 2.2826181945294066,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044770180655567343,
      "loss": 5.3136,
      "mean_token_accuracy": 0.19528917074203492,
      "num_tokens": 52939305.0,
      "step": 26120
    },
    {
      "entropy": 5.928196001052856,
      "epoch": 2.2830551428821115,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004476818314829261,
      "loss": 5.1686,
      "mean_token_accuracy": 0.2085715964436531,
      "num_tokens": 52948814.0,
      "step": 26125
    },
    {
      "entropy": 5.952007102966308,
      "epoch": 2.283492091234816,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004476618530980506,
      "loss": 5.3218,
      "mean_token_accuracy": 0.1947695419192314,
      "num_tokens": 52958964.0,
      "step": 26130
    },
    {
      "entropy": 5.942716360092163,
      "epoch": 2.283929039587521,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004476418714014305,
      "loss": 5.3271,
      "mean_token_accuracy": 0.1946884125471115,
      "num_tokens": 52969837.0,
      "step": 26135
    },
    {
      "entropy": 5.964176797866822,
      "epoch": 2.2843659879402254,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044762188639344875,
      "loss": 5.2289,
      "mean_token_accuracy": 0.2071533501148224,
      "num_tokens": 52979184.0,
      "step": 26140
    },
    {
      "entropy": 6.028575611114502,
      "epoch": 2.2848029362929303,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044760189807448874,
      "loss": 5.389,
      "mean_token_accuracy": 0.1971719279885292,
      "num_tokens": 52989867.0,
      "step": 26145
    },
    {
      "entropy": 5.956194114685059,
      "epoch": 2.2852398846456348,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004475819064449338,
      "loss": 5.311,
      "mean_token_accuracy": 0.2002057284116745,
      "num_tokens": 53000177.0,
      "step": 26150
    },
    {
      "entropy": 5.914187955856323,
      "epoch": 2.2856768329983397,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044756191150516764,
      "loss": 5.203,
      "mean_token_accuracy": 0.19147240817546846,
      "num_tokens": 53009372.0,
      "step": 26155
    },
    {
      "entropy": 5.95326795578003,
      "epoch": 2.286113781351044,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004475419132555735,
      "loss": 5.2468,
      "mean_token_accuracy": 0.20163957476615907,
      "num_tokens": 53020405.0,
      "step": 26160
    },
    {
      "entropy": 5.965266323089599,
      "epoch": 2.286550729703749,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044752191169653504,
      "loss": 5.3147,
      "mean_token_accuracy": 0.20004501342773437,
      "num_tokens": 53030713.0,
      "step": 26165
    },
    {
      "entropy": 5.966091728210449,
      "epoch": 2.2869876780564535,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044750190682843596,
      "loss": 5.2616,
      "mean_token_accuracy": 0.20498313754796982,
      "num_tokens": 53040868.0,
      "step": 26170
    },
    {
      "entropy": 5.937581777572632,
      "epoch": 2.2874246264091584,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004474818986516599,
      "loss": 5.1633,
      "mean_token_accuracy": 0.20949238687753677,
      "num_tokens": 53050349.0,
      "step": 26175
    },
    {
      "entropy": 5.92527322769165,
      "epoch": 2.2878615747618634,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044746188716659064,
      "loss": 5.2519,
      "mean_token_accuracy": 0.20017112791538239,
      "num_tokens": 53060446.0,
      "step": 26180
    },
    {
      "entropy": 5.952855443954467,
      "epoch": 2.288298523114568,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044744187237361207,
      "loss": 5.2919,
      "mean_token_accuracy": 0.1998363420367241,
      "num_tokens": 53071099.0,
      "step": 26185
    },
    {
      "entropy": 5.9445091724395756,
      "epoch": 2.2887354714672727,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044742185427310797,
      "loss": 5.212,
      "mean_token_accuracy": 0.20551558583974838,
      "num_tokens": 53080712.0,
      "step": 26190
    },
    {
      "entropy": 5.981446218490601,
      "epoch": 2.289172419819977,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004474018328654623,
      "loss": 5.3668,
      "mean_token_accuracy": 0.19039946645498276,
      "num_tokens": 53091160.0,
      "step": 26195
    },
    {
      "entropy": 5.92049503326416,
      "epoch": 2.289609368172682,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004473818081510592,
      "loss": 5.2968,
      "mean_token_accuracy": 0.19933284372091292,
      "num_tokens": 53100767.0,
      "step": 26200
    },
    {
      "entropy": 5.9829723834991455,
      "epoch": 2.2900463165253866,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044736178013028276,
      "loss": 5.2859,
      "mean_token_accuracy": 0.2022222936153412,
      "num_tokens": 53111641.0,
      "step": 26205
    },
    {
      "entropy": 5.965549755096435,
      "epoch": 2.2904832648780915,
      "grad_norm": 1.125,
      "learning_rate": 0.000447341748803517,
      "loss": 5.261,
      "mean_token_accuracy": 0.20411663949489595,
      "num_tokens": 53121982.0,
      "step": 26210
    },
    {
      "entropy": 5.993684864044189,
      "epoch": 2.290920213230796,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044732171417114614,
      "loss": 5.3048,
      "mean_token_accuracy": 0.1976929098367691,
      "num_tokens": 53132649.0,
      "step": 26215
    },
    {
      "entropy": 5.96528115272522,
      "epoch": 2.291357161583501,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004473016762335546,
      "loss": 5.3241,
      "mean_token_accuracy": 0.20592397153377534,
      "num_tokens": 53142249.0,
      "step": 26220
    },
    {
      "entropy": 5.993523359298706,
      "epoch": 2.2917941099362054,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004472816349911266,
      "loss": 5.2484,
      "mean_token_accuracy": 0.20680857449769974,
      "num_tokens": 53152375.0,
      "step": 26225
    },
    {
      "entropy": 5.975111484527588,
      "epoch": 2.2922310582889103,
      "grad_norm": 1.125,
      "learning_rate": 0.0004472615904442465,
      "loss": 5.3315,
      "mean_token_accuracy": 0.20043864250183105,
      "num_tokens": 53162699.0,
      "step": 26230
    },
    {
      "entropy": 5.990478992462158,
      "epoch": 2.292668006641615,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044724154259329893,
      "loss": 5.3076,
      "mean_token_accuracy": 0.19685998558998108,
      "num_tokens": 53172890.0,
      "step": 26235
    },
    {
      "entropy": 5.995484447479248,
      "epoch": 2.2931049549943197,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004472214914386682,
      "loss": 5.2945,
      "mean_token_accuracy": 0.20336523056030273,
      "num_tokens": 53183735.0,
      "step": 26240
    },
    {
      "entropy": 6.018250799179077,
      "epoch": 2.2935419033470246,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004472014369807391,
      "loss": 5.2903,
      "mean_token_accuracy": 0.19737145751714708,
      "num_tokens": 53193046.0,
      "step": 26245
    },
    {
      "entropy": 5.962446641921997,
      "epoch": 2.293978851699729,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004471813792198962,
      "loss": 5.2392,
      "mean_token_accuracy": 0.2090782105922699,
      "num_tokens": 53203367.0,
      "step": 26250
    },
    {
      "entropy": 5.988530063629151,
      "epoch": 2.294415800052434,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044716131815652425,
      "loss": 5.4028,
      "mean_token_accuracy": 0.19561094641685486,
      "num_tokens": 53214207.0,
      "step": 26255
    },
    {
      "entropy": 6.035137462615967,
      "epoch": 2.2948527484051384,
      "grad_norm": 1.140625,
      "learning_rate": 0.000447141253791008,
      "loss": 5.2734,
      "mean_token_accuracy": 0.20258764326572418,
      "num_tokens": 53225061.0,
      "step": 26260
    },
    {
      "entropy": 5.995845317840576,
      "epoch": 2.2952896967578433,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044712118612373223,
      "loss": 5.3161,
      "mean_token_accuracy": 0.19787241667509078,
      "num_tokens": 53235463.0,
      "step": 26265
    },
    {
      "entropy": 6.0012853145599365,
      "epoch": 2.295726645110548,
      "grad_norm": 1.1875,
      "learning_rate": 0.000447101115155082,
      "loss": 5.3186,
      "mean_token_accuracy": 0.19806245118379592,
      "num_tokens": 53245923.0,
      "step": 26270
    },
    {
      "entropy": 5.971248960494995,
      "epoch": 2.2961635934632527,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044708104088544216,
      "loss": 5.2589,
      "mean_token_accuracy": 0.20370552986860274,
      "num_tokens": 53255756.0,
      "step": 26275
    },
    {
      "entropy": 5.950687026977539,
      "epoch": 2.296600541815957,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004470609633151978,
      "loss": 5.2652,
      "mean_token_accuracy": 0.199011167883873,
      "num_tokens": 53266176.0,
      "step": 26280
    },
    {
      "entropy": 5.962865972518921,
      "epoch": 2.297037490168662,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044704088244473404,
      "loss": 5.2799,
      "mean_token_accuracy": 0.2014361023902893,
      "num_tokens": 53276703.0,
      "step": 26285
    },
    {
      "entropy": 6.027005529403686,
      "epoch": 2.297474438521367,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000447020798274436,
      "loss": 5.3695,
      "mean_token_accuracy": 0.19316516071558,
      "num_tokens": 53286511.0,
      "step": 26290
    },
    {
      "entropy": 5.9820160388946535,
      "epoch": 2.2979113868740715,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044700071080468884,
      "loss": 5.2635,
      "mean_token_accuracy": 0.20730896294116974,
      "num_tokens": 53297261.0,
      "step": 26295
    },
    {
      "entropy": 6.001583099365234,
      "epoch": 2.298348335226776,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000446980620035878,
      "loss": 5.2979,
      "mean_token_accuracy": 0.1977737069129944,
      "num_tokens": 53307054.0,
      "step": 26300
    },
    {
      "entropy": 5.970858383178711,
      "epoch": 2.298785283579481,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044696052596838873,
      "loss": 5.272,
      "mean_token_accuracy": 0.20610567778348923,
      "num_tokens": 53318022.0,
      "step": 26305
    },
    {
      "entropy": 5.956424951553345,
      "epoch": 2.299222231932186,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004469404286026064,
      "loss": 5.2773,
      "mean_token_accuracy": 0.20874002128839492,
      "num_tokens": 53328462.0,
      "step": 26310
    },
    {
      "entropy": 5.931696605682373,
      "epoch": 2.2996591802848902,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044692032793891665,
      "loss": 5.2166,
      "mean_token_accuracy": 0.20963830649852752,
      "num_tokens": 53337837.0,
      "step": 26315
    },
    {
      "entropy": 5.918432903289795,
      "epoch": 2.300096128637595,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044690022397770484,
      "loss": 5.2313,
      "mean_token_accuracy": 0.20754361450672149,
      "num_tokens": 53348026.0,
      "step": 26320
    },
    {
      "entropy": 5.977982997894287,
      "epoch": 2.3005330769902996,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004468801167193567,
      "loss": 5.2844,
      "mean_token_accuracy": 0.20818592756986617,
      "num_tokens": 53358881.0,
      "step": 26325
    },
    {
      "entropy": 5.959980726242065,
      "epoch": 2.3009700253430045,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044686000616425795,
      "loss": 5.266,
      "mean_token_accuracy": 0.20063979029655457,
      "num_tokens": 53369732.0,
      "step": 26330
    },
    {
      "entropy": 5.946565961837768,
      "epoch": 2.301406973695709,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044683989231279406,
      "loss": 5.2846,
      "mean_token_accuracy": 0.19914212822914124,
      "num_tokens": 53379607.0,
      "step": 26335
    },
    {
      "entropy": 6.031122207641602,
      "epoch": 2.301843922048414,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044681977516535106,
      "loss": 5.3083,
      "mean_token_accuracy": 0.19219865798950195,
      "num_tokens": 53390630.0,
      "step": 26340
    },
    {
      "entropy": 5.9593569278717045,
      "epoch": 2.3022808704011184,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004467996547223148,
      "loss": 5.201,
      "mean_token_accuracy": 0.20763223618268967,
      "num_tokens": 53400830.0,
      "step": 26345
    },
    {
      "entropy": 5.95982871055603,
      "epoch": 2.3027178187538233,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044677953098407107,
      "loss": 5.2764,
      "mean_token_accuracy": 0.19661026448011398,
      "num_tokens": 53411150.0,
      "step": 26350
    },
    {
      "entropy": 5.930993461608887,
      "epoch": 2.303154767106528,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004467594039510059,
      "loss": 5.2581,
      "mean_token_accuracy": 0.20994644165039061,
      "num_tokens": 53421072.0,
      "step": 26355
    },
    {
      "entropy": 5.999410486221313,
      "epoch": 2.3035917154592327,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044673927362350545,
      "loss": 5.2224,
      "mean_token_accuracy": 0.20166186839342118,
      "num_tokens": 53430659.0,
      "step": 26360
    },
    {
      "entropy": 5.962993049621582,
      "epoch": 2.3040286638119376,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044671914000195566,
      "loss": 5.1474,
      "mean_token_accuracy": 0.20727490037679672,
      "num_tokens": 53440883.0,
      "step": 26365
    },
    {
      "entropy": 5.940031814575195,
      "epoch": 2.304465612164642,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044669900308674283,
      "loss": 5.2764,
      "mean_token_accuracy": 0.20556019097566605,
      "num_tokens": 53450491.0,
      "step": 26370
    },
    {
      "entropy": 5.914482831954956,
      "epoch": 2.304902560517347,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044667886287825316,
      "loss": 5.2743,
      "mean_token_accuracy": 0.20146131217479707,
      "num_tokens": 53460864.0,
      "step": 26375
    },
    {
      "entropy": 5.953470611572266,
      "epoch": 2.3053395088700515,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044665871937687296,
      "loss": 5.262,
      "mean_token_accuracy": 0.20339878797531127,
      "num_tokens": 53470965.0,
      "step": 26380
    },
    {
      "entropy": 5.945506191253662,
      "epoch": 2.3057764572227564,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004466385725829885,
      "loss": 5.1892,
      "mean_token_accuracy": 0.2092132732272148,
      "num_tokens": 53480272.0,
      "step": 26385
    },
    {
      "entropy": 5.930268859863281,
      "epoch": 2.306213405575461,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004466184224969863,
      "loss": 5.3017,
      "mean_token_accuracy": 0.20419214516878129,
      "num_tokens": 53490285.0,
      "step": 26390
    },
    {
      "entropy": 5.9424481868743895,
      "epoch": 2.3066503539281658,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004465982691192529,
      "loss": 5.1883,
      "mean_token_accuracy": 0.20558033883571625,
      "num_tokens": 53499712.0,
      "step": 26395
    },
    {
      "entropy": 5.936081647872925,
      "epoch": 2.3070873022808702,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004465781124501747,
      "loss": 5.2548,
      "mean_token_accuracy": 0.2037995547056198,
      "num_tokens": 53510144.0,
      "step": 26400
    },
    {
      "entropy": 5.97359447479248,
      "epoch": 2.307524250633575,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044655795249013845,
      "loss": 5.2735,
      "mean_token_accuracy": 0.2047096908092499,
      "num_tokens": 53519936.0,
      "step": 26405
    },
    {
      "entropy": 5.943789863586426,
      "epoch": 2.3079611989862796,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004465377892395308,
      "loss": 5.3052,
      "mean_token_accuracy": 0.20355853140354158,
      "num_tokens": 53530447.0,
      "step": 26410
    },
    {
      "entropy": 6.006121206283569,
      "epoch": 2.3083981473389845,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004465176226987384,
      "loss": 5.3344,
      "mean_token_accuracy": 0.20488938987255095,
      "num_tokens": 53540929.0,
      "step": 26415
    },
    {
      "entropy": 5.962171173095703,
      "epoch": 2.3088350956916894,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004464974528681481,
      "loss": 5.2108,
      "mean_token_accuracy": 0.21313857138156891,
      "num_tokens": 53551575.0,
      "step": 26420
    },
    {
      "entropy": 5.996268320083618,
      "epoch": 2.309272044044394,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004464772797481469,
      "loss": 5.3426,
      "mean_token_accuracy": 0.19480671137571334,
      "num_tokens": 53563081.0,
      "step": 26425
    },
    {
      "entropy": 5.954373550415039,
      "epoch": 2.309708992397099,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044645710333912153,
      "loss": 5.3011,
      "mean_token_accuracy": 0.20285181403160096,
      "num_tokens": 53572303.0,
      "step": 26430
    },
    {
      "entropy": 5.955782318115235,
      "epoch": 2.3101459407498033,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044643692364145914,
      "loss": 5.2334,
      "mean_token_accuracy": 0.2037407085299492,
      "num_tokens": 53582033.0,
      "step": 26435
    },
    {
      "entropy": 5.8240924835205075,
      "epoch": 2.310582889102508,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044641674065554665,
      "loss": 5.0829,
      "mean_token_accuracy": 0.23076471388339997,
      "num_tokens": 53592140.0,
      "step": 26440
    },
    {
      "entropy": 5.9721067428588865,
      "epoch": 2.3110198374552127,
      "grad_norm": 1.125,
      "learning_rate": 0.00044639655438177134,
      "loss": 5.2556,
      "mean_token_accuracy": 0.19518211036920546,
      "num_tokens": 53602601.0,
      "step": 26445
    },
    {
      "entropy": 5.9424888610839846,
      "epoch": 2.3114567858079176,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044637636482052027,
      "loss": 5.2618,
      "mean_token_accuracy": 0.20011826306581498,
      "num_tokens": 53613292.0,
      "step": 26450
    },
    {
      "entropy": 5.995673084259034,
      "epoch": 2.311893734160622,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004463561719721807,
      "loss": 5.2464,
      "mean_token_accuracy": 0.2027132675051689,
      "num_tokens": 53624493.0,
      "step": 26455
    },
    {
      "entropy": 5.9879261493682865,
      "epoch": 2.312330682513327,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044633597583713995,
      "loss": 5.2588,
      "mean_token_accuracy": 0.2047705367207527,
      "num_tokens": 53633739.0,
      "step": 26460
    },
    {
      "entropy": 5.986911964416504,
      "epoch": 2.3127676308660314,
      "grad_norm": 1.125,
      "learning_rate": 0.0004463157764157855,
      "loss": 5.2322,
      "mean_token_accuracy": 0.20419119894504548,
      "num_tokens": 53643025.0,
      "step": 26465
    },
    {
      "entropy": 5.966569662094116,
      "epoch": 2.3132045792187363,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004462955737085046,
      "loss": 5.2295,
      "mean_token_accuracy": 0.20560057312250138,
      "num_tokens": 53652905.0,
      "step": 26470
    },
    {
      "entropy": 5.890585994720459,
      "epoch": 2.3136415275714413,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004462753677156849,
      "loss": 5.2233,
      "mean_token_accuracy": 0.20625951439142226,
      "num_tokens": 53663189.0,
      "step": 26475
    },
    {
      "entropy": 5.93616509437561,
      "epoch": 2.3140784759241457,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004462551584377138,
      "loss": 5.2912,
      "mean_token_accuracy": 0.20430757850408554,
      "num_tokens": 53673068.0,
      "step": 26480
    },
    {
      "entropy": 5.975815725326538,
      "epoch": 2.3145154242768506,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004462349458749791,
      "loss": 5.3284,
      "mean_token_accuracy": 0.19725441038608552,
      "num_tokens": 53683929.0,
      "step": 26485
    },
    {
      "entropy": 5.989509916305542,
      "epoch": 2.314952372629555,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044621473002786846,
      "loss": 5.309,
      "mean_token_accuracy": 0.19584776908159257,
      "num_tokens": 53694170.0,
      "step": 26490
    },
    {
      "entropy": 5.981809520721436,
      "epoch": 2.31538932098226,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044619451089676946,
      "loss": 5.2515,
      "mean_token_accuracy": 0.20595163255929946,
      "num_tokens": 53704491.0,
      "step": 26495
    },
    {
      "entropy": 5.996604633331299,
      "epoch": 2.3158262693349645,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004461742884820701,
      "loss": 5.2909,
      "mean_token_accuracy": 0.19817458987236022,
      "num_tokens": 53714971.0,
      "step": 26500
    },
    {
      "entropy": 5.70117826461792,
      "epoch": 2.3162632176876694,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004461540627841582,
      "loss": 5.0444,
      "mean_token_accuracy": 0.2353384166955948,
      "num_tokens": 53723965.0,
      "step": 26505
    },
    {
      "entropy": 5.958590412139893,
      "epoch": 2.316700166040374,
      "grad_norm": 1.125,
      "learning_rate": 0.00044613383380342163,
      "loss": 5.2481,
      "mean_token_accuracy": 0.20474223643541337,
      "num_tokens": 53734561.0,
      "step": 26510
    },
    {
      "entropy": 6.012588977813721,
      "epoch": 2.317137114393079,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004461136015402485,
      "loss": 5.2464,
      "mean_token_accuracy": 0.20269781351089478,
      "num_tokens": 53745115.0,
      "step": 26515
    },
    {
      "entropy": 5.937128162384033,
      "epoch": 2.3175740627457833,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004460933659950269,
      "loss": 5.1689,
      "mean_token_accuracy": 0.21048847138881682,
      "num_tokens": 53754884.0,
      "step": 26520
    },
    {
      "entropy": 5.900316143035889,
      "epoch": 2.318011011098488,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004460731271681448,
      "loss": 5.2809,
      "mean_token_accuracy": 0.1981632113456726,
      "num_tokens": 53764772.0,
      "step": 26525
    },
    {
      "entropy": 6.001019906997681,
      "epoch": 2.318447959451193,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044605288505999047,
      "loss": 5.3276,
      "mean_token_accuracy": 0.1957862436771393,
      "num_tokens": 53775391.0,
      "step": 26530
    },
    {
      "entropy": 6.0049241065979,
      "epoch": 2.3188849078038976,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044603263967095224,
      "loss": 5.2805,
      "mean_token_accuracy": 0.19861340075731276,
      "num_tokens": 53785698.0,
      "step": 26535
    },
    {
      "entropy": 5.955899620056153,
      "epoch": 2.3193218561566025,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044601239100141833,
      "loss": 5.2547,
      "mean_token_accuracy": 0.20283966064453124,
      "num_tokens": 53795432.0,
      "step": 26540
    },
    {
      "entropy": 5.983546733856201,
      "epoch": 2.319758804509307,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044599213905177724,
      "loss": 5.3204,
      "mean_token_accuracy": 0.20452350229024888,
      "num_tokens": 53805295.0,
      "step": 26545
    },
    {
      "entropy": 5.970443248748779,
      "epoch": 2.320195752862012,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004459718838224171,
      "loss": 5.2923,
      "mean_token_accuracy": 0.2064556062221527,
      "num_tokens": 53814765.0,
      "step": 26550
    },
    {
      "entropy": 5.979111576080323,
      "epoch": 2.3206327012147163,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044595162531372684,
      "loss": 5.2387,
      "mean_token_accuracy": 0.20427174121141434,
      "num_tokens": 53824154.0,
      "step": 26555
    },
    {
      "entropy": 5.972795534133911,
      "epoch": 2.3210696495674212,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044593136352609475,
      "loss": 5.2801,
      "mean_token_accuracy": 0.20141192376613617,
      "num_tokens": 53834867.0,
      "step": 26560
    },
    {
      "entropy": 6.02893385887146,
      "epoch": 2.3215065979201257,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004459110984599096,
      "loss": 5.3231,
      "mean_token_accuracy": 0.19553861767053604,
      "num_tokens": 53844901.0,
      "step": 26565
    },
    {
      "entropy": 5.994007110595703,
      "epoch": 2.3219435462728306,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044589083011555995,
      "loss": 5.3573,
      "mean_token_accuracy": 0.19827143847942352,
      "num_tokens": 53854897.0,
      "step": 26570
    },
    {
      "entropy": 5.964595794677734,
      "epoch": 2.322380494625535,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004458705584934347,
      "loss": 5.3181,
      "mean_token_accuracy": 0.19411334097385408,
      "num_tokens": 53865152.0,
      "step": 26575
    },
    {
      "entropy": 6.015668296813965,
      "epoch": 2.32281744297824,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044585028359392256,
      "loss": 5.2992,
      "mean_token_accuracy": 0.1975010707974434,
      "num_tokens": 53875261.0,
      "step": 26580
    },
    {
      "entropy": 5.979590225219726,
      "epoch": 2.323254391330945,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004458300054174125,
      "loss": 5.2366,
      "mean_token_accuracy": 0.20530228912830353,
      "num_tokens": 53885281.0,
      "step": 26585
    },
    {
      "entropy": 5.9040392398834225,
      "epoch": 2.3236913396836494,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004458097239642933,
      "loss": 5.2631,
      "mean_token_accuracy": 0.20108226984739302,
      "num_tokens": 53895282.0,
      "step": 26590
    },
    {
      "entropy": 5.976267957687378,
      "epoch": 2.3241282880363543,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004457894392349542,
      "loss": 5.3048,
      "mean_token_accuracy": 0.20036439895629882,
      "num_tokens": 53905843.0,
      "step": 26595
    },
    {
      "entropy": 5.956758165359497,
      "epoch": 2.3245652363890588,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004457691512297841,
      "loss": 5.3386,
      "mean_token_accuracy": 0.19962980151176452,
      "num_tokens": 53915270.0,
      "step": 26600
    },
    {
      "entropy": 6.017842245101929,
      "epoch": 2.3250021847417637,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004457488599491723,
      "loss": 5.395,
      "mean_token_accuracy": 0.1905118480324745,
      "num_tokens": 53924548.0,
      "step": 26605
    },
    {
      "entropy": 6.020603847503662,
      "epoch": 2.325439133094468,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044572856539350784,
      "loss": 5.3213,
      "mean_token_accuracy": 0.20167311877012253,
      "num_tokens": 53935696.0,
      "step": 26610
    },
    {
      "entropy": 5.99643177986145,
      "epoch": 2.325876081447173,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044570826756318006,
      "loss": 5.2928,
      "mean_token_accuracy": 0.20029123574495317,
      "num_tokens": 53945499.0,
      "step": 26615
    },
    {
      "entropy": 5.953942346572876,
      "epoch": 2.3263130297998775,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004456879664585782,
      "loss": 5.2078,
      "mean_token_accuracy": 0.209720279276371,
      "num_tokens": 53955478.0,
      "step": 26620
    },
    {
      "entropy": 6.010342597961426,
      "epoch": 2.3267499781525824,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004456676620800917,
      "loss": 5.218,
      "mean_token_accuracy": 0.20359838157892227,
      "num_tokens": 53964908.0,
      "step": 26625
    },
    {
      "entropy": 5.963017129898072,
      "epoch": 2.327186926505287,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004456473544281101,
      "loss": 5.2242,
      "mean_token_accuracy": 0.20630244314670562,
      "num_tokens": 53974850.0,
      "step": 26630
    },
    {
      "entropy": 5.917931032180786,
      "epoch": 2.327623874857992,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004456270435030228,
      "loss": 5.201,
      "mean_token_accuracy": 0.20956290364265442,
      "num_tokens": 53984458.0,
      "step": 26635
    },
    {
      "entropy": 5.962974166870117,
      "epoch": 2.3280608232106967,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004456067293052193,
      "loss": 5.2527,
      "mean_token_accuracy": 0.1999155566096306,
      "num_tokens": 53994560.0,
      "step": 26640
    },
    {
      "entropy": 6.0151038646698,
      "epoch": 2.328497771563401,
      "grad_norm": 1.125,
      "learning_rate": 0.00044558641183508945,
      "loss": 5.2535,
      "mean_token_accuracy": 0.20367266982793808,
      "num_tokens": 54003698.0,
      "step": 26645
    },
    {
      "entropy": 5.921915292739868,
      "epoch": 2.3289347199161057,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044556609109302277,
      "loss": 5.1859,
      "mean_token_accuracy": 0.21634477525949478,
      "num_tokens": 54013405.0,
      "step": 26650
    },
    {
      "entropy": 5.990165710449219,
      "epoch": 2.3293716682688106,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004455457670794092,
      "loss": 5.2825,
      "mean_token_accuracy": 0.20301346927881242,
      "num_tokens": 54023140.0,
      "step": 26655
    },
    {
      "entropy": 5.972466230392456,
      "epoch": 2.3298086166215155,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044552543979463826,
      "loss": 5.2379,
      "mean_token_accuracy": 0.2037163943052292,
      "num_tokens": 54032506.0,
      "step": 26660
    },
    {
      "entropy": 5.910902404785157,
      "epoch": 2.33024556497422,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004455051092391002,
      "loss": 5.1579,
      "mean_token_accuracy": 0.2148728549480438,
      "num_tokens": 54042113.0,
      "step": 26665
    },
    {
      "entropy": 5.905826044082642,
      "epoch": 2.330682513326925,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044548477541318473,
      "loss": 5.2551,
      "mean_token_accuracy": 0.20448985695838928,
      "num_tokens": 54051139.0,
      "step": 26670
    },
    {
      "entropy": 5.912610244750977,
      "epoch": 2.3311194616796294,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004454644383172819,
      "loss": 5.2567,
      "mean_token_accuracy": 0.20457020103931428,
      "num_tokens": 54060929.0,
      "step": 26675
    },
    {
      "entropy": 6.011033248901367,
      "epoch": 2.3315564100323343,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004454440979517819,
      "loss": 5.2869,
      "mean_token_accuracy": 0.20203414261341096,
      "num_tokens": 54071654.0,
      "step": 26680
    },
    {
      "entropy": 5.968383026123047,
      "epoch": 2.3319933583850387,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004454237543170747,
      "loss": 5.2025,
      "mean_token_accuracy": 0.20858086496591569,
      "num_tokens": 54081497.0,
      "step": 26685
    },
    {
      "entropy": 5.978128719329834,
      "epoch": 2.3324303067377437,
      "grad_norm": 1.390625,
      "learning_rate": 0.00044540340741355063,
      "loss": 5.2566,
      "mean_token_accuracy": 0.20199504494667053,
      "num_tokens": 54091899.0,
      "step": 26690
    },
    {
      "entropy": 5.922659111022949,
      "epoch": 2.332867255090448,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004453830572416,
      "loss": 5.2315,
      "mean_token_accuracy": 0.2067328616976738,
      "num_tokens": 54101734.0,
      "step": 26695
    },
    {
      "entropy": 5.962366056442261,
      "epoch": 2.333304203443153,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000445362703801613,
      "loss": 5.2721,
      "mean_token_accuracy": 0.2009028673171997,
      "num_tokens": 54112633.0,
      "step": 26700
    },
    {
      "entropy": 5.934648275375366,
      "epoch": 2.3337411517958575,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044534234709398,
      "loss": 5.2481,
      "mean_token_accuracy": 0.2002313628792763,
      "num_tokens": 54122182.0,
      "step": 26705
    },
    {
      "entropy": 5.953657436370849,
      "epoch": 2.3341781001485624,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044532198711909153,
      "loss": 5.2944,
      "mean_token_accuracy": 0.1951600894331932,
      "num_tokens": 54131410.0,
      "step": 26710
    },
    {
      "entropy": 5.906000566482544,
      "epoch": 2.3346150485012673,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004453016238773382,
      "loss": 5.2005,
      "mean_token_accuracy": 0.20845098048448563,
      "num_tokens": 54141890.0,
      "step": 26715
    },
    {
      "entropy": 5.960325860977173,
      "epoch": 2.335051996853972,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004452812573691104,
      "loss": 5.2865,
      "mean_token_accuracy": 0.20292362123727797,
      "num_tokens": 54151579.0,
      "step": 26720
    },
    {
      "entropy": 5.97622275352478,
      "epoch": 2.3354889452066767,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044526088759479893,
      "loss": 5.2998,
      "mean_token_accuracy": 0.20106198638677597,
      "num_tokens": 54161824.0,
      "step": 26725
    },
    {
      "entropy": 5.900078439712525,
      "epoch": 2.335925893559381,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044524051455479435,
      "loss": 5.2119,
      "mean_token_accuracy": 0.20764299631118774,
      "num_tokens": 54172110.0,
      "step": 26730
    },
    {
      "entropy": 5.938591337203979,
      "epoch": 2.336362841912086,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044522013824948756,
      "loss": 5.2222,
      "mean_token_accuracy": 0.20456480234861374,
      "num_tokens": 54181852.0,
      "step": 26735
    },
    {
      "entropy": 5.9802178859710695,
      "epoch": 2.3367997902647906,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044519975867926933,
      "loss": 5.3026,
      "mean_token_accuracy": 0.20258145928382873,
      "num_tokens": 54192703.0,
      "step": 26740
    },
    {
      "entropy": 5.939910173416138,
      "epoch": 2.3372367386174955,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004451793758445305,
      "loss": 5.2083,
      "mean_token_accuracy": 0.2068295419216156,
      "num_tokens": 54201613.0,
      "step": 26745
    },
    {
      "entropy": 5.964314985275268,
      "epoch": 2.3376736869702,
      "grad_norm": 1.171875,
      "learning_rate": 0.000445158989745662,
      "loss": 5.335,
      "mean_token_accuracy": 0.19785381406545638,
      "num_tokens": 54211717.0,
      "step": 26750
    },
    {
      "entropy": 5.894619989395141,
      "epoch": 2.338110635322905,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044513860038305505,
      "loss": 5.1441,
      "mean_token_accuracy": 0.21711382269859314,
      "num_tokens": 54221481.0,
      "step": 26755
    },
    {
      "entropy": 6.004653882980347,
      "epoch": 2.3385475836756093,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004451182077571005,
      "loss": 5.2477,
      "mean_token_accuracy": 0.2077726736664772,
      "num_tokens": 54231558.0,
      "step": 26760
    },
    {
      "entropy": 5.988905143737793,
      "epoch": 2.3389845320283142,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044509781186818963,
      "loss": 5.2487,
      "mean_token_accuracy": 0.20294829159975053,
      "num_tokens": 54241150.0,
      "step": 26765
    },
    {
      "entropy": 5.9420814037323,
      "epoch": 2.339421480381019,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004450774127167136,
      "loss": 5.2611,
      "mean_token_accuracy": 0.20844826847314835,
      "num_tokens": 54251323.0,
      "step": 26770
    },
    {
      "entropy": 5.976119327545166,
      "epoch": 2.3398584287337236,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004450570103030637,
      "loss": 5.2645,
      "mean_token_accuracy": 0.19800906181335448,
      "num_tokens": 54260956.0,
      "step": 26775
    },
    {
      "entropy": 5.989648628234863,
      "epoch": 2.3402953770864285,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044503660462763116,
      "loss": 5.2616,
      "mean_token_accuracy": 0.20241255462169647,
      "num_tokens": 54270382.0,
      "step": 26780
    },
    {
      "entropy": 5.979282665252685,
      "epoch": 2.340732325439133,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004450161956908075,
      "loss": 5.2851,
      "mean_token_accuracy": 0.20182901471853257,
      "num_tokens": 54280380.0,
      "step": 26785
    },
    {
      "entropy": 5.977569818496704,
      "epoch": 2.341169273791838,
      "grad_norm": 1.125,
      "learning_rate": 0.0004449957834929841,
      "loss": 5.2685,
      "mean_token_accuracy": 0.19728201180696486,
      "num_tokens": 54290536.0,
      "step": 26790
    },
    {
      "entropy": 5.892529678344727,
      "epoch": 2.3416062221445424,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044497536803455245,
      "loss": 5.2097,
      "mean_token_accuracy": 0.20839814245700836,
      "num_tokens": 54299835.0,
      "step": 26795
    },
    {
      "entropy": 5.941550779342651,
      "epoch": 2.3420431704972473,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004449549493159043,
      "loss": 5.2477,
      "mean_token_accuracy": 0.2107363298535347,
      "num_tokens": 54309496.0,
      "step": 26800
    },
    {
      "entropy": 5.913251686096191,
      "epoch": 2.3424801188499518,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004449345273374311,
      "loss": 5.2281,
      "mean_token_accuracy": 0.20055098384618758,
      "num_tokens": 54318591.0,
      "step": 26805
    },
    {
      "entropy": 5.945567941665649,
      "epoch": 2.3429170672026567,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004449141020995246,
      "loss": 5.243,
      "mean_token_accuracy": 0.20487989336252213,
      "num_tokens": 54327995.0,
      "step": 26810
    },
    {
      "entropy": 5.994795131683349,
      "epoch": 2.343354015555361,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004448936736025765,
      "loss": 5.2999,
      "mean_token_accuracy": 0.19670633971691132,
      "num_tokens": 54339122.0,
      "step": 26815
    },
    {
      "entropy": 6.023054599761963,
      "epoch": 2.343790963908066,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004448732418469789,
      "loss": 5.316,
      "mean_token_accuracy": 0.2010682240128517,
      "num_tokens": 54349004.0,
      "step": 26820
    },
    {
      "entropy": 5.949343585968018,
      "epoch": 2.344227912260771,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044485280683312345,
      "loss": 5.2567,
      "mean_token_accuracy": 0.21017249375581742,
      "num_tokens": 54359859.0,
      "step": 26825
    },
    {
      "entropy": 5.970385265350342,
      "epoch": 2.3446648606134755,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004448323685614021,
      "loss": 5.2429,
      "mean_token_accuracy": 0.2007167875766754,
      "num_tokens": 54369386.0,
      "step": 26830
    },
    {
      "entropy": 5.915089893341064,
      "epoch": 2.3451018089661804,
      "grad_norm": 1.25,
      "learning_rate": 0.00044481192703220695,
      "loss": 5.2129,
      "mean_token_accuracy": 0.20602288991212844,
      "num_tokens": 54379250.0,
      "step": 26835
    },
    {
      "entropy": 5.955948448181152,
      "epoch": 2.345538757318885,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044479148224593014,
      "loss": 5.2768,
      "mean_token_accuracy": 0.20283494889736176,
      "num_tokens": 54388980.0,
      "step": 26840
    },
    {
      "entropy": 5.874410390853882,
      "epoch": 2.3459757056715898,
      "grad_norm": 1.4375,
      "learning_rate": 0.00044477103420296365,
      "loss": 5.1356,
      "mean_token_accuracy": 0.21990542113780975,
      "num_tokens": 54398849.0,
      "step": 26845
    },
    {
      "entropy": 6.0112803936004635,
      "epoch": 2.346412654024294,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004447505829036998,
      "loss": 5.2921,
      "mean_token_accuracy": 0.20450956076383592,
      "num_tokens": 54409426.0,
      "step": 26850
    },
    {
      "entropy": 5.99263105392456,
      "epoch": 2.346849602376999,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004447301283485308,
      "loss": 5.3131,
      "mean_token_accuracy": 0.20286180227994918,
      "num_tokens": 54420212.0,
      "step": 26855
    },
    {
      "entropy": 5.97694239616394,
      "epoch": 2.3472865507297036,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044470967053784904,
      "loss": 5.3754,
      "mean_token_accuracy": 0.19570057541131974,
      "num_tokens": 54431179.0,
      "step": 26860
    },
    {
      "entropy": 5.971631860733032,
      "epoch": 2.3477234990824085,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004446892094720468,
      "loss": 5.2688,
      "mean_token_accuracy": 0.2113860160112381,
      "num_tokens": 54442247.0,
      "step": 26865
    },
    {
      "entropy": 5.984760427474976,
      "epoch": 2.348160447435113,
      "grad_norm": 1.125,
      "learning_rate": 0.00044466874515151667,
      "loss": 5.2421,
      "mean_token_accuracy": 0.20909582376480101,
      "num_tokens": 54451475.0,
      "step": 26870
    },
    {
      "entropy": 5.953150081634521,
      "epoch": 2.348597395787818,
      "grad_norm": 1.203125,
      "learning_rate": 0.000444648277576651,
      "loss": 5.3489,
      "mean_token_accuracy": 0.19249757677316665,
      "num_tokens": 54463259.0,
      "step": 26875
    },
    {
      "entropy": 5.916848468780517,
      "epoch": 2.349034344140523,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044462780674784264,
      "loss": 5.2598,
      "mean_token_accuracy": 0.20741068869829177,
      "num_tokens": 54473573.0,
      "step": 26880
    },
    {
      "entropy": 6.081042003631592,
      "epoch": 2.3494712924932273,
      "grad_norm": 1.234375,
      "learning_rate": 0.000444607332665484,
      "loss": 5.375,
      "mean_token_accuracy": 0.18983974903821946,
      "num_tokens": 54482907.0,
      "step": 26885
    },
    {
      "entropy": 5.988726043701172,
      "epoch": 2.349908240845932,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004445868553299678,
      "loss": 5.3086,
      "mean_token_accuracy": 0.20222176015377044,
      "num_tokens": 54492993.0,
      "step": 26890
    },
    {
      "entropy": 5.933333206176758,
      "epoch": 2.3503451891986367,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004445663747416868,
      "loss": 5.3103,
      "mean_token_accuracy": 0.19639436900615692,
      "num_tokens": 54503826.0,
      "step": 26895
    },
    {
      "entropy": 5.95296196937561,
      "epoch": 2.3507821375513416,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044454589090103406,
      "loss": 5.3105,
      "mean_token_accuracy": 0.19913260191679,
      "num_tokens": 54513839.0,
      "step": 26900
    },
    {
      "entropy": 5.978499269485473,
      "epoch": 2.351219085904046,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044452540380840217,
      "loss": 5.2139,
      "mean_token_accuracy": 0.20908889621496202,
      "num_tokens": 54523742.0,
      "step": 26905
    },
    {
      "entropy": 6.016820049285888,
      "epoch": 2.351656034256751,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00044450491346418426,
      "loss": 5.3178,
      "mean_token_accuracy": 0.1964950755238533,
      "num_tokens": 54534017.0,
      "step": 26910
    },
    {
      "entropy": 5.891880798339844,
      "epoch": 2.3520929826094554,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004444844198687732,
      "loss": 5.2012,
      "mean_token_accuracy": 0.2067766234278679,
      "num_tokens": 54544445.0,
      "step": 26915
    },
    {
      "entropy": 5.9215961456298825,
      "epoch": 2.3525299309621603,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004444639230225622,
      "loss": 5.2197,
      "mean_token_accuracy": 0.20776305049657823,
      "num_tokens": 54554966.0,
      "step": 26920
    },
    {
      "entropy": 5.996667003631591,
      "epoch": 2.352966879314865,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004444434229259444,
      "loss": 5.2417,
      "mean_token_accuracy": 0.20572907477617264,
      "num_tokens": 54565005.0,
      "step": 26925
    },
    {
      "entropy": 5.979354190826416,
      "epoch": 2.3534038276675697,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004444229195793129,
      "loss": 5.2738,
      "mean_token_accuracy": 0.20435307770967484,
      "num_tokens": 54575242.0,
      "step": 26930
    },
    {
      "entropy": 5.969934606552124,
      "epoch": 2.3538407760202746,
      "grad_norm": 1.25,
      "learning_rate": 0.0004444024129830611,
      "loss": 5.2853,
      "mean_token_accuracy": 0.19297420978546143,
      "num_tokens": 54585448.0,
      "step": 26935
    },
    {
      "entropy": 5.909618330001831,
      "epoch": 2.354277724372979,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004443819031375822,
      "loss": 5.2104,
      "mean_token_accuracy": 0.2089756175875664,
      "num_tokens": 54595891.0,
      "step": 26940
    },
    {
      "entropy": 5.9971921920776365,
      "epoch": 2.3547146727256836,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004443613900432697,
      "loss": 5.3306,
      "mean_token_accuracy": 0.191923063993454,
      "num_tokens": 54607407.0,
      "step": 26945
    },
    {
      "entropy": 6.029334449768067,
      "epoch": 2.3551516210783885,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044434087370051693,
      "loss": 5.2907,
      "mean_token_accuracy": 0.20099771469831468,
      "num_tokens": 54617337.0,
      "step": 26950
    },
    {
      "entropy": 5.98782000541687,
      "epoch": 2.3555885694310934,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004443203541097175,
      "loss": 5.2675,
      "mean_token_accuracy": 0.20305315852165223,
      "num_tokens": 54627914.0,
      "step": 26955
    },
    {
      "entropy": 5.9617678165435795,
      "epoch": 2.356025517783798,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004442998312712649,
      "loss": 5.1761,
      "mean_token_accuracy": 0.20356649905443192,
      "num_tokens": 54636709.0,
      "step": 26960
    },
    {
      "entropy": 5.91498384475708,
      "epoch": 2.356462466136503,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044427930518555294,
      "loss": 5.2216,
      "mean_token_accuracy": 0.20750172734260558,
      "num_tokens": 54646167.0,
      "step": 26965
    },
    {
      "entropy": 5.978960943222046,
      "epoch": 2.3568994144892073,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004442587758529751,
      "loss": 5.2915,
      "mean_token_accuracy": 0.19665145576000215,
      "num_tokens": 54656663.0,
      "step": 26970
    },
    {
      "entropy": 5.982604455947876,
      "epoch": 2.357336362841912,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004442382432739253,
      "loss": 5.2974,
      "mean_token_accuracy": 0.19814304113388062,
      "num_tokens": 54667123.0,
      "step": 26975
    },
    {
      "entropy": 5.999262094497681,
      "epoch": 2.3577733111946166,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044421770744879725,
      "loss": 5.2744,
      "mean_token_accuracy": 0.20507195442914963,
      "num_tokens": 54677381.0,
      "step": 26980
    },
    {
      "entropy": 5.957340478897095,
      "epoch": 2.3582102595473216,
      "grad_norm": 1.125,
      "learning_rate": 0.000444197168377985,
      "loss": 5.2503,
      "mean_token_accuracy": 0.20747270286083222,
      "num_tokens": 54687190.0,
      "step": 26985
    },
    {
      "entropy": 6.042789506912231,
      "epoch": 2.3586472079000265,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044417662606188236,
      "loss": 5.3101,
      "mean_token_accuracy": 0.20170580446720124,
      "num_tokens": 54697102.0,
      "step": 26990
    },
    {
      "entropy": 5.956511640548706,
      "epoch": 2.359084156252731,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004441560805008834,
      "loss": 5.2628,
      "mean_token_accuracy": 0.2039721593260765,
      "num_tokens": 54707315.0,
      "step": 26995
    },
    {
      "entropy": 5.984962224960327,
      "epoch": 2.3595211046054354,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004441355316953822,
      "loss": 5.3321,
      "mean_token_accuracy": 0.1911090075969696,
      "num_tokens": 54717873.0,
      "step": 27000
    },
    {
      "epoch": 2.3595211046054354,
      "eval_entropy": 5.704830788009904,
      "eval_loss": 5.3230390548706055,
      "eval_mean_token_accuracy": 0.2087297729305683,
      "eval_num_tokens": 54717873.0,
      "eval_runtime": 21.5381,
      "eval_samples_per_second": 1718.539,
      "eval_steps_per_second": 214.829,
      "step": 27000
    },
    {
      "entropy": 5.982849931716919,
      "epoch": 2.3599580529581403,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004441149796457728,
      "loss": 5.3306,
      "mean_token_accuracy": 0.1975221410393715,
      "num_tokens": 54727723.0,
      "step": 27005
    },
    {
      "entropy": 5.9637621402740475,
      "epoch": 2.3603950013108452,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044409442435244947,
      "loss": 5.2406,
      "mean_token_accuracy": 0.20944924801588058,
      "num_tokens": 54738108.0,
      "step": 27010
    },
    {
      "entropy": 5.971192169189453,
      "epoch": 2.3608319496635497,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004440738658158065,
      "loss": 5.3423,
      "mean_token_accuracy": 0.20041386038064957,
      "num_tokens": 54749011.0,
      "step": 27015
    },
    {
      "entropy": 6.031109476089478,
      "epoch": 2.3612688980162546,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004440533040362382,
      "loss": 5.2936,
      "mean_token_accuracy": 0.19950413852930068,
      "num_tokens": 54759922.0,
      "step": 27020
    },
    {
      "entropy": 5.989197540283203,
      "epoch": 2.361705846368959,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044403273901413897,
      "loss": 5.2793,
      "mean_token_accuracy": 0.19582067281007767,
      "num_tokens": 54770134.0,
      "step": 27025
    },
    {
      "entropy": 5.983657932281494,
      "epoch": 2.362142794721664,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044401217074990323,
      "loss": 5.2056,
      "mean_token_accuracy": 0.20376692116260528,
      "num_tokens": 54780131.0,
      "step": 27030
    },
    {
      "entropy": 5.965873336791992,
      "epoch": 2.3625797430743685,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044399159924392545,
      "loss": 5.3194,
      "mean_token_accuracy": 0.20284462124109268,
      "num_tokens": 54790601.0,
      "step": 27035
    },
    {
      "entropy": 6.018246984481811,
      "epoch": 2.3630166914270734,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004439710244966003,
      "loss": 5.2558,
      "mean_token_accuracy": 0.20530499517917633,
      "num_tokens": 54799995.0,
      "step": 27040
    },
    {
      "entropy": 5.990733528137207,
      "epoch": 2.363453639779778,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004439504465083223,
      "loss": 5.2461,
      "mean_token_accuracy": 0.2041343629360199,
      "num_tokens": 54810299.0,
      "step": 27045
    },
    {
      "entropy": 5.972197484970093,
      "epoch": 2.3638905881324828,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004439298652794863,
      "loss": 5.2336,
      "mean_token_accuracy": 0.20192066133022307,
      "num_tokens": 54820802.0,
      "step": 27050
    },
    {
      "entropy": 5.9647118091583256,
      "epoch": 2.3643275364851872,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044390928081048693,
      "loss": 5.2461,
      "mean_token_accuracy": 0.20625963658094407,
      "num_tokens": 54831077.0,
      "step": 27055
    },
    {
      "entropy": 5.9700343132019045,
      "epoch": 2.364764484837892,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000443888693101719,
      "loss": 5.2542,
      "mean_token_accuracy": 0.20682216584682464,
      "num_tokens": 54842574.0,
      "step": 27060
    },
    {
      "entropy": 5.903518438339233,
      "epoch": 2.365201433190597,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044386810215357747,
      "loss": 5.1756,
      "mean_token_accuracy": 0.21350000351667403,
      "num_tokens": 54852365.0,
      "step": 27065
    },
    {
      "entropy": 5.95312271118164,
      "epoch": 2.3656383815433015,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044384750796645724,
      "loss": 5.2765,
      "mean_token_accuracy": 0.19836094975471497,
      "num_tokens": 54863008.0,
      "step": 27070
    },
    {
      "entropy": 6.000019216537476,
      "epoch": 2.3660753298960064,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004438269105407534,
      "loss": 5.2492,
      "mean_token_accuracy": 0.2001797989010811,
      "num_tokens": 54873134.0,
      "step": 27075
    },
    {
      "entropy": 6.03321213722229,
      "epoch": 2.366512278248711,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044380630987686086,
      "loss": 5.3115,
      "mean_token_accuracy": 0.20018837749958038,
      "num_tokens": 54883075.0,
      "step": 27080
    },
    {
      "entropy": 5.95359582901001,
      "epoch": 2.366949226601416,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044378570597517495,
      "loss": 5.2598,
      "mean_token_accuracy": 0.20041357278823851,
      "num_tokens": 54892788.0,
      "step": 27085
    },
    {
      "entropy": 5.9087730884552006,
      "epoch": 2.3673861749541203,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004437650988360906,
      "loss": 5.233,
      "mean_token_accuracy": 0.20632907301187514,
      "num_tokens": 54902708.0,
      "step": 27090
    },
    {
      "entropy": 5.9328968048095705,
      "epoch": 2.367823123306825,
      "grad_norm": 1.125,
      "learning_rate": 0.0004437444884600034,
      "loss": 5.224,
      "mean_token_accuracy": 0.20729875713586807,
      "num_tokens": 54912906.0,
      "step": 27095
    },
    {
      "entropy": 5.949405574798584,
      "epoch": 2.3682600716595297,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044372387484730835,
      "loss": 5.2856,
      "mean_token_accuracy": 0.19627260118722917,
      "num_tokens": 54922802.0,
      "step": 27100
    },
    {
      "entropy": 5.978369474411011,
      "epoch": 2.3686970200122346,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044370325799840106,
      "loss": 5.2952,
      "mean_token_accuracy": 0.2031734824180603,
      "num_tokens": 54933203.0,
      "step": 27105
    },
    {
      "entropy": 5.983736896514893,
      "epoch": 2.369133968364939,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044368263791367684,
      "loss": 5.2901,
      "mean_token_accuracy": 0.20049805343151092,
      "num_tokens": 54944081.0,
      "step": 27110
    },
    {
      "entropy": 6.01103982925415,
      "epoch": 2.369570916717644,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044366201459353124,
      "loss": 5.2692,
      "mean_token_accuracy": 0.20412447452545165,
      "num_tokens": 54953421.0,
      "step": 27115
    },
    {
      "entropy": 5.907924938201904,
      "epoch": 2.370007865070349,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044364138803835984,
      "loss": 5.2547,
      "mean_token_accuracy": 0.20586992055177689,
      "num_tokens": 54964127.0,
      "step": 27120
    },
    {
      "entropy": 5.934879159927368,
      "epoch": 2.3704448134230534,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044362075824855816,
      "loss": 5.2266,
      "mean_token_accuracy": 0.20922868251800536,
      "num_tokens": 54974636.0,
      "step": 27125
    },
    {
      "entropy": 6.032119941711426,
      "epoch": 2.3708817617757583,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044360012522452215,
      "loss": 5.3148,
      "mean_token_accuracy": 0.1971466138958931,
      "num_tokens": 54985542.0,
      "step": 27130
    },
    {
      "entropy": 6.007154750823974,
      "epoch": 2.3713187101284627,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004435794889666472,
      "loss": 5.2271,
      "mean_token_accuracy": 0.21279266178607942,
      "num_tokens": 54994973.0,
      "step": 27135
    },
    {
      "entropy": 5.949537897109986,
      "epoch": 2.3717556584811677,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044355884947532947,
      "loss": 5.3475,
      "mean_token_accuracy": 0.19813959896564484,
      "num_tokens": 55006510.0,
      "step": 27140
    },
    {
      "entropy": 5.9902795314788815,
      "epoch": 2.372192606833872,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004435382067509647,
      "loss": 5.3261,
      "mean_token_accuracy": 0.200550577044487,
      "num_tokens": 55016375.0,
      "step": 27145
    },
    {
      "entropy": 5.942539119720459,
      "epoch": 2.372629555186577,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044351756079394865,
      "loss": 5.2705,
      "mean_token_accuracy": 0.20712611079216003,
      "num_tokens": 55026213.0,
      "step": 27150
    },
    {
      "entropy": 6.031870126724243,
      "epoch": 2.3730665035392815,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004434969116046776,
      "loss": 5.3276,
      "mean_token_accuracy": 0.1961845800280571,
      "num_tokens": 55036836.0,
      "step": 27155
    },
    {
      "entropy": 6.0053221702575685,
      "epoch": 2.3735034518919864,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004434762591835475,
      "loss": 5.2084,
      "mean_token_accuracy": 0.20188466012477874,
      "num_tokens": 55046243.0,
      "step": 27160
    },
    {
      "entropy": 5.9510399341583256,
      "epoch": 2.373940400244691,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044345560353095437,
      "loss": 5.2628,
      "mean_token_accuracy": 0.1976679414510727,
      "num_tokens": 55055882.0,
      "step": 27165
    },
    {
      "entropy": 5.901238250732422,
      "epoch": 2.374377348597396,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004434349446472946,
      "loss": 5.189,
      "mean_token_accuracy": 0.20720597952604294,
      "num_tokens": 55066398.0,
      "step": 27170
    },
    {
      "entropy": 5.988975858688354,
      "epoch": 2.3748142969501007,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044341428253296436,
      "loss": 5.3403,
      "mean_token_accuracy": 0.1944714829325676,
      "num_tokens": 55077053.0,
      "step": 27175
    },
    {
      "entropy": 5.9933271408081055,
      "epoch": 2.375251245302805,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004433936171883598,
      "loss": 5.244,
      "mean_token_accuracy": 0.20705890357494355,
      "num_tokens": 55086973.0,
      "step": 27180
    },
    {
      "entropy": 6.002259254455566,
      "epoch": 2.37568819365551,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004433729486138775,
      "loss": 5.2314,
      "mean_token_accuracy": 0.2073718264698982,
      "num_tokens": 55096162.0,
      "step": 27185
    },
    {
      "entropy": 5.965226554870606,
      "epoch": 2.3761251420082146,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044335227680991385,
      "loss": 5.2268,
      "mean_token_accuracy": 0.207683527469635,
      "num_tokens": 55105852.0,
      "step": 27190
    },
    {
      "entropy": 5.914014434814453,
      "epoch": 2.3765620903609195,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004433316017768653,
      "loss": 5.182,
      "mean_token_accuracy": 0.21618347316980363,
      "num_tokens": 55116092.0,
      "step": 27195
    },
    {
      "entropy": 5.997405862808227,
      "epoch": 2.376999038713624,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004433109235151284,
      "loss": 5.2415,
      "mean_token_accuracy": 0.2110252782702446,
      "num_tokens": 55125439.0,
      "step": 27200
    },
    {
      "entropy": 6.007530975341797,
      "epoch": 2.377435987066329,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044329024202509977,
      "loss": 5.3053,
      "mean_token_accuracy": 0.1986546978354454,
      "num_tokens": 55134684.0,
      "step": 27205
    },
    {
      "entropy": 5.926865720748902,
      "epoch": 2.3778729354190333,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044326955730717614,
      "loss": 5.2085,
      "mean_token_accuracy": 0.20802289843559266,
      "num_tokens": 55143853.0,
      "step": 27210
    },
    {
      "entropy": 6.071642780303955,
      "epoch": 2.3783098837717382,
      "grad_norm": 1.25,
      "learning_rate": 0.0004432488693617542,
      "loss": 5.4024,
      "mean_token_accuracy": 0.19118579775094985,
      "num_tokens": 55154190.0,
      "step": 27215
    },
    {
      "entropy": 6.007005310058593,
      "epoch": 2.3787468321244427,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044322817818923083,
      "loss": 5.3255,
      "mean_token_accuracy": 0.20593781769275665,
      "num_tokens": 55164158.0,
      "step": 27220
    },
    {
      "entropy": 6.001749181747437,
      "epoch": 2.3791837804771476,
      "grad_norm": 1.25,
      "learning_rate": 0.00044320748379000284,
      "loss": 5.3549,
      "mean_token_accuracy": 0.1917647436261177,
      "num_tokens": 55173635.0,
      "step": 27225
    },
    {
      "entropy": 6.0066002368927,
      "epoch": 2.3796207288298525,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044318678616446717,
      "loss": 5.2534,
      "mean_token_accuracy": 0.20938255190849303,
      "num_tokens": 55182972.0,
      "step": 27230
    },
    {
      "entropy": 6.002149248123169,
      "epoch": 2.380057677182557,
      "grad_norm": 1.125,
      "learning_rate": 0.00044316608531302085,
      "loss": 5.2822,
      "mean_token_accuracy": 0.20340266972780227,
      "num_tokens": 55192769.0,
      "step": 27235
    },
    {
      "entropy": 5.995946168899536,
      "epoch": 2.380494625535262,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044314538123606083,
      "loss": 5.2971,
      "mean_token_accuracy": 0.20241550505161285,
      "num_tokens": 55203241.0,
      "step": 27240
    },
    {
      "entropy": 5.877031040191651,
      "epoch": 2.3809315738879664,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044312467393398434,
      "loss": 5.2097,
      "mean_token_accuracy": 0.21047720909118653,
      "num_tokens": 55214272.0,
      "step": 27245
    },
    {
      "entropy": 5.913450384140015,
      "epoch": 2.3813685222406713,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044310396340718853,
      "loss": 5.2051,
      "mean_token_accuracy": 0.20949835032224656,
      "num_tokens": 55223947.0,
      "step": 27250
    },
    {
      "entropy": 5.961161375045776,
      "epoch": 2.3818054705933758,
      "grad_norm": 1.125,
      "learning_rate": 0.00044308324965607054,
      "loss": 5.3259,
      "mean_token_accuracy": 0.19441009908914567,
      "num_tokens": 55234906.0,
      "step": 27255
    },
    {
      "entropy": 5.914666557312012,
      "epoch": 2.3822424189460807,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004430625326810278,
      "loss": 5.1527,
      "mean_token_accuracy": 0.21315354108810425,
      "num_tokens": 55244785.0,
      "step": 27260
    },
    {
      "entropy": 5.986311340332032,
      "epoch": 2.382679367298785,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044304181248245754,
      "loss": 5.341,
      "mean_token_accuracy": 0.1988701954483986,
      "num_tokens": 55255053.0,
      "step": 27265
    },
    {
      "entropy": 5.978615379333496,
      "epoch": 2.38311631565149,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004430210890607574,
      "loss": 5.3253,
      "mean_token_accuracy": 0.19757238179445266,
      "num_tokens": 55264525.0,
      "step": 27270
    },
    {
      "entropy": 5.99290189743042,
      "epoch": 2.3835532640041945,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044300036241632463,
      "loss": 5.2762,
      "mean_token_accuracy": 0.1983378142118454,
      "num_tokens": 55273486.0,
      "step": 27275
    },
    {
      "entropy": 5.968915033340454,
      "epoch": 2.3839902123568995,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044297963254955687,
      "loss": 5.2329,
      "mean_token_accuracy": 0.2075064331293106,
      "num_tokens": 55283164.0,
      "step": 27280
    },
    {
      "entropy": 5.987004327774048,
      "epoch": 2.3844271607096044,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044295889946085185,
      "loss": 5.2924,
      "mean_token_accuracy": 0.20133266150951384,
      "num_tokens": 55294657.0,
      "step": 27285
    },
    {
      "entropy": 5.996598720550537,
      "epoch": 2.384864109062309,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044293816315060704,
      "loss": 5.2806,
      "mean_token_accuracy": 0.19815976917743683,
      "num_tokens": 55304743.0,
      "step": 27290
    },
    {
      "entropy": 5.923869609832764,
      "epoch": 2.3853010574150133,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004429174236192203,
      "loss": 5.2435,
      "mean_token_accuracy": 0.21297921389341354,
      "num_tokens": 55314106.0,
      "step": 27295
    },
    {
      "entropy": 5.973803234100342,
      "epoch": 2.385738005767718,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044289668086708935,
      "loss": 5.2635,
      "mean_token_accuracy": 0.2007545992732048,
      "num_tokens": 55324407.0,
      "step": 27300
    },
    {
      "entropy": 5.949147081375122,
      "epoch": 2.386174954120423,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044287593489461205,
      "loss": 5.2809,
      "mean_token_accuracy": 0.20726819485425949,
      "num_tokens": 55335853.0,
      "step": 27305
    },
    {
      "entropy": 5.958419370651245,
      "epoch": 2.3866119024731276,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044285518570218635,
      "loss": 5.239,
      "mean_token_accuracy": 0.20606197118759156,
      "num_tokens": 55345947.0,
      "step": 27310
    },
    {
      "entropy": 5.982055425643921,
      "epoch": 2.3870488508258325,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004428344332902103,
      "loss": 5.2772,
      "mean_token_accuracy": 0.20437521487474442,
      "num_tokens": 55355709.0,
      "step": 27315
    },
    {
      "entropy": 5.991170930862427,
      "epoch": 2.387485799178537,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044281367765908186,
      "loss": 5.212,
      "mean_token_accuracy": 0.1990652173757553,
      "num_tokens": 55365597.0,
      "step": 27320
    },
    {
      "entropy": 5.970980024337768,
      "epoch": 2.387922747531242,
      "grad_norm": 1.25,
      "learning_rate": 0.00044279291880919915,
      "loss": 5.2866,
      "mean_token_accuracy": 0.20076058208942413,
      "num_tokens": 55375508.0,
      "step": 27325
    },
    {
      "entropy": 5.942668819427491,
      "epoch": 2.3883596958839464,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044277215674096035,
      "loss": 5.2607,
      "mean_token_accuracy": 0.20210288614034652,
      "num_tokens": 55384928.0,
      "step": 27330
    },
    {
      "entropy": 6.004765510559082,
      "epoch": 2.3887966442366513,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004427513914547637,
      "loss": 5.2543,
      "mean_token_accuracy": 0.21121509671211242,
      "num_tokens": 55394805.0,
      "step": 27335
    },
    {
      "entropy": 6.003274202346802,
      "epoch": 2.3892335925893557,
      "grad_norm": 1.125,
      "learning_rate": 0.0004427306229510074,
      "loss": 5.3104,
      "mean_token_accuracy": 0.20391065925359725,
      "num_tokens": 55405360.0,
      "step": 27340
    },
    {
      "entropy": 5.986766719818116,
      "epoch": 2.3896705409420607,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044270985123008984,
      "loss": 5.2669,
      "mean_token_accuracy": 0.19399458318948745,
      "num_tokens": 55415124.0,
      "step": 27345
    },
    {
      "entropy": 5.959262561798096,
      "epoch": 2.390107489294765,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004426890762924096,
      "loss": 5.2421,
      "mean_token_accuracy": 0.20405221432447435,
      "num_tokens": 55425621.0,
      "step": 27350
    },
    {
      "entropy": 6.0206701278686525,
      "epoch": 2.39054443764747,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004426682981383648,
      "loss": 5.3172,
      "mean_token_accuracy": 0.19685909301042556,
      "num_tokens": 55435812.0,
      "step": 27355
    },
    {
      "entropy": 5.987735652923584,
      "epoch": 2.390981386000175,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004426475167683543,
      "loss": 5.2892,
      "mean_token_accuracy": 0.19937194883823395,
      "num_tokens": 55445462.0,
      "step": 27360
    },
    {
      "entropy": 5.952284479141236,
      "epoch": 2.3914183343528794,
      "grad_norm": 1.125,
      "learning_rate": 0.00044262673218277654,
      "loss": 5.2673,
      "mean_token_accuracy": 0.19712261110544205,
      "num_tokens": 55455992.0,
      "step": 27365
    },
    {
      "entropy": 6.005670022964478,
      "epoch": 2.3918552827055843,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044260594438203025,
      "loss": 5.3362,
      "mean_token_accuracy": 0.19653766900300979,
      "num_tokens": 55466715.0,
      "step": 27370
    },
    {
      "entropy": 5.978766536712646,
      "epoch": 2.392292231058289,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044258515336651416,
      "loss": 5.2949,
      "mean_token_accuracy": 0.20065311938524247,
      "num_tokens": 55476280.0,
      "step": 27375
    },
    {
      "entropy": 5.9732804775238035,
      "epoch": 2.3927291794109937,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044256435913662696,
      "loss": 5.2494,
      "mean_token_accuracy": 0.20217181891202926,
      "num_tokens": 55486105.0,
      "step": 27380
    },
    {
      "entropy": 6.020232343673706,
      "epoch": 2.393166127763698,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004425435616927676,
      "loss": 5.2458,
      "mean_token_accuracy": 0.2030083045363426,
      "num_tokens": 55496152.0,
      "step": 27385
    },
    {
      "entropy": 5.9320024967193605,
      "epoch": 2.393603076116403,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004425227610353349,
      "loss": 5.2796,
      "mean_token_accuracy": 0.209136700630188,
      "num_tokens": 55507104.0,
      "step": 27390
    },
    {
      "entropy": 5.954079294204712,
      "epoch": 2.3940400244691076,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004425019571647279,
      "loss": 5.2718,
      "mean_token_accuracy": 0.2043379858136177,
      "num_tokens": 55517114.0,
      "step": 27395
    },
    {
      "entropy": 5.929471683502197,
      "epoch": 2.3944769728218125,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044248115008134557,
      "loss": 5.1954,
      "mean_token_accuracy": 0.2144207775592804,
      "num_tokens": 55526719.0,
      "step": 27400
    },
    {
      "entropy": 5.955900478363037,
      "epoch": 2.394913921174517,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044246033978558697,
      "loss": 5.2385,
      "mean_token_accuracy": 0.20485748797655107,
      "num_tokens": 55536760.0,
      "step": 27405
    },
    {
      "entropy": 5.964769411087036,
      "epoch": 2.395350869527222,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044243952627785136,
      "loss": 5.2714,
      "mean_token_accuracy": 0.20527591407299042,
      "num_tokens": 55547193.0,
      "step": 27410
    },
    {
      "entropy": 5.922680521011353,
      "epoch": 2.395787817879927,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044241870955853793,
      "loss": 5.2537,
      "mean_token_accuracy": 0.20538367033004762,
      "num_tokens": 55557257.0,
      "step": 27415
    },
    {
      "entropy": 5.945881700515747,
      "epoch": 2.3962247662326313,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044239788962804584,
      "loss": 5.2327,
      "mean_token_accuracy": 0.20429478585720062,
      "num_tokens": 55568191.0,
      "step": 27420
    },
    {
      "entropy": 5.980617380142212,
      "epoch": 2.396661714585336,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044237706648677456,
      "loss": 5.2463,
      "mean_token_accuracy": 0.20753553062677382,
      "num_tokens": 55578983.0,
      "step": 27425
    },
    {
      "entropy": 5.968792676925659,
      "epoch": 2.3970986629380406,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004423562401351234,
      "loss": 5.2367,
      "mean_token_accuracy": 0.20625957250595092,
      "num_tokens": 55588772.0,
      "step": 27430
    },
    {
      "entropy": 5.913537883758545,
      "epoch": 2.3975356112907455,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004423354105734919,
      "loss": 5.2512,
      "mean_token_accuracy": 0.20531272739171982,
      "num_tokens": 55599514.0,
      "step": 27435
    },
    {
      "entropy": 5.953787517547608,
      "epoch": 2.39797255964345,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004423145778022795,
      "loss": 5.2762,
      "mean_token_accuracy": 0.1969349429011345,
      "num_tokens": 55609183.0,
      "step": 27440
    },
    {
      "entropy": 6.010332345962524,
      "epoch": 2.398409507996155,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004422937418218859,
      "loss": 5.2882,
      "mean_token_accuracy": 0.20224117785692214,
      "num_tokens": 55618836.0,
      "step": 27445
    },
    {
      "entropy": 5.984930324554443,
      "epoch": 2.3988464563488594,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044227290263271066,
      "loss": 5.3077,
      "mean_token_accuracy": 0.1989605113863945,
      "num_tokens": 55629314.0,
      "step": 27450
    },
    {
      "entropy": 5.917552757263183,
      "epoch": 2.3992834047015643,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004422520602351535,
      "loss": 5.1802,
      "mean_token_accuracy": 0.22355787009000777,
      "num_tokens": 55640542.0,
      "step": 27455
    },
    {
      "entropy": 5.999539804458618,
      "epoch": 2.399720353054269,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004422312146296141,
      "loss": 5.3568,
      "mean_token_accuracy": 0.19682765752077103,
      "num_tokens": 55650165.0,
      "step": 27460
    },
    {
      "entropy": 6.028295421600342,
      "epoch": 2.4001573014069737,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044221036581649245,
      "loss": 5.2941,
      "mean_token_accuracy": 0.20216074138879775,
      "num_tokens": 55658962.0,
      "step": 27465
    },
    {
      "entropy": 5.994452810287475,
      "epoch": 2.4005942497596786,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004421895137961884,
      "loss": 5.2597,
      "mean_token_accuracy": 0.19487877190113068,
      "num_tokens": 55668357.0,
      "step": 27470
    },
    {
      "entropy": 5.982410860061646,
      "epoch": 2.401031198112383,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044216865856910176,
      "loss": 5.3447,
      "mean_token_accuracy": 0.19967369139194488,
      "num_tokens": 55678313.0,
      "step": 27475
    },
    {
      "entropy": 5.979382133483886,
      "epoch": 2.401468146465088,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044214780013563277,
      "loss": 5.287,
      "mean_token_accuracy": 0.19618748873472214,
      "num_tokens": 55689332.0,
      "step": 27480
    },
    {
      "entropy": 5.988975095748901,
      "epoch": 2.4019050948177925,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044212693849618136,
      "loss": 5.2589,
      "mean_token_accuracy": 0.20787834823131562,
      "num_tokens": 55699397.0,
      "step": 27485
    },
    {
      "entropy": 6.013682889938354,
      "epoch": 2.4023420431704974,
      "grad_norm": 1.25,
      "learning_rate": 0.0004421060736511476,
      "loss": 5.3296,
      "mean_token_accuracy": 0.19969446510076522,
      "num_tokens": 55708548.0,
      "step": 27490
    },
    {
      "entropy": 6.070343017578125,
      "epoch": 2.402778991523202,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044208520560093183,
      "loss": 5.3161,
      "mean_token_accuracy": 0.19814254641532897,
      "num_tokens": 55718043.0,
      "step": 27495
    },
    {
      "entropy": 5.986764430999756,
      "epoch": 2.4032159398759068,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004420643343459343,
      "loss": 5.2746,
      "mean_token_accuracy": 0.20370566844940186,
      "num_tokens": 55727889.0,
      "step": 27500
    },
    {
      "entropy": 5.972061586380005,
      "epoch": 2.4036528882286112,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004420434598865553,
      "loss": 5.3055,
      "mean_token_accuracy": 0.2046552926301956,
      "num_tokens": 55738993.0,
      "step": 27505
    },
    {
      "entropy": 5.976064968109131,
      "epoch": 2.404089836581316,
      "grad_norm": 1.125,
      "learning_rate": 0.0004420225822231952,
      "loss": 5.3058,
      "mean_token_accuracy": 0.20477852821350098,
      "num_tokens": 55748858.0,
      "step": 27510
    },
    {
      "entropy": 5.943922519683838,
      "epoch": 2.4045267849340206,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044200170135625446,
      "loss": 5.2361,
      "mean_token_accuracy": 0.21117992252111434,
      "num_tokens": 55758639.0,
      "step": 27515
    },
    {
      "entropy": 6.010950422286987,
      "epoch": 2.4049637332867255,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004419808172861336,
      "loss": 5.3128,
      "mean_token_accuracy": 0.20427795499563217,
      "num_tokens": 55769632.0,
      "step": 27520
    },
    {
      "entropy": 6.052605772018433,
      "epoch": 2.4054006816394304,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004419599300132331,
      "loss": 5.3334,
      "mean_token_accuracy": 0.1971154436469078,
      "num_tokens": 55780147.0,
      "step": 27525
    },
    {
      "entropy": 5.961267471313477,
      "epoch": 2.405837629992135,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044193903953795367,
      "loss": 5.2218,
      "mean_token_accuracy": 0.20476755648851394,
      "num_tokens": 55790933.0,
      "step": 27530
    },
    {
      "entropy": 5.928826236724854,
      "epoch": 2.40627457834484,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044191814586069605,
      "loss": 5.2333,
      "mean_token_accuracy": 0.2079709067940712,
      "num_tokens": 55800479.0,
      "step": 27535
    },
    {
      "entropy": 5.940217590332031,
      "epoch": 2.4067115266975443,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004418972489818609,
      "loss": 5.2465,
      "mean_token_accuracy": 0.2091786503791809,
      "num_tokens": 55810613.0,
      "step": 27540
    },
    {
      "entropy": 6.001028108596802,
      "epoch": 2.407148475050249,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044187634890184904,
      "loss": 5.3143,
      "mean_token_accuracy": 0.19674837440252305,
      "num_tokens": 55821507.0,
      "step": 27545
    },
    {
      "entropy": 5.996193075180054,
      "epoch": 2.4075854234029537,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004418554456210614,
      "loss": 5.3526,
      "mean_token_accuracy": 0.1924208715558052,
      "num_tokens": 55831878.0,
      "step": 27550
    },
    {
      "entropy": 5.968644380569458,
      "epoch": 2.4080223717556586,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044183453913989885,
      "loss": 5.2115,
      "mean_token_accuracy": 0.2054414689540863,
      "num_tokens": 55841314.0,
      "step": 27555
    },
    {
      "entropy": 6.029853105545044,
      "epoch": 2.408459320108363,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004418136294587624,
      "loss": 5.2594,
      "mean_token_accuracy": 0.20659389197826386,
      "num_tokens": 55851056.0,
      "step": 27560
    },
    {
      "entropy": 5.966867876052857,
      "epoch": 2.408896268461068,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004417927165780531,
      "loss": 5.228,
      "mean_token_accuracy": 0.20561548620462416,
      "num_tokens": 55860705.0,
      "step": 27565
    },
    {
      "entropy": 5.95285210609436,
      "epoch": 2.4093332168137724,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004417718004981723,
      "loss": 5.2979,
      "mean_token_accuracy": 0.20367242842912675,
      "num_tokens": 55870749.0,
      "step": 27570
    },
    {
      "entropy": 6.018473863601685,
      "epoch": 2.4097701651664774,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004417508812195207,
      "loss": 5.34,
      "mean_token_accuracy": 0.19647326022386552,
      "num_tokens": 55880135.0,
      "step": 27575
    },
    {
      "entropy": 6.000452661514283,
      "epoch": 2.4102071135191823,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004417299587425001,
      "loss": 5.2112,
      "mean_token_accuracy": 0.20734529942274094,
      "num_tokens": 55890357.0,
      "step": 27580
    },
    {
      "entropy": 6.009810590744019,
      "epoch": 2.4106440618718867,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004417090330675114,
      "loss": 5.2917,
      "mean_token_accuracy": 0.19983459264039993,
      "num_tokens": 55900080.0,
      "step": 27585
    },
    {
      "entropy": 5.941344451904297,
      "epoch": 2.4110810102245916,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044168810419495597,
      "loss": 5.2751,
      "mean_token_accuracy": 0.1988365352153778,
      "num_tokens": 55911326.0,
      "step": 27590
    },
    {
      "entropy": 5.978070783615112,
      "epoch": 2.411517958577296,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004416671721252355,
      "loss": 5.2103,
      "mean_token_accuracy": 0.2115773603320122,
      "num_tokens": 55921872.0,
      "step": 27595
    },
    {
      "entropy": 5.929601240158081,
      "epoch": 2.411954906930001,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044164623685875133,
      "loss": 5.237,
      "mean_token_accuracy": 0.21200128346681596,
      "num_tokens": 55932539.0,
      "step": 27600
    },
    {
      "entropy": 5.965108489990234,
      "epoch": 2.4123918552827055,
      "grad_norm": 1.328125,
      "learning_rate": 0.000441625298395905,
      "loss": 5.3117,
      "mean_token_accuracy": 0.2031942129135132,
      "num_tokens": 55942087.0,
      "step": 27605
    },
    {
      "entropy": 6.0166497230529785,
      "epoch": 2.4128288036354104,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004416043567370981,
      "loss": 5.2647,
      "mean_token_accuracy": 0.19898139387369157,
      "num_tokens": 55951604.0,
      "step": 27610
    },
    {
      "entropy": 5.991033554077148,
      "epoch": 2.413265751988115,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004415834118827323,
      "loss": 5.2588,
      "mean_token_accuracy": 0.20171024054288864,
      "num_tokens": 55961687.0,
      "step": 27615
    },
    {
      "entropy": 5.981624507904053,
      "epoch": 2.41370270034082,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004415624638332095,
      "loss": 5.2965,
      "mean_token_accuracy": 0.20424082428216933,
      "num_tokens": 55972088.0,
      "step": 27620
    },
    {
      "entropy": 5.936146926879883,
      "epoch": 2.4141396486935243,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044154151258893125,
      "loss": 5.3145,
      "mean_token_accuracy": 0.19551793187856675,
      "num_tokens": 55981857.0,
      "step": 27625
    },
    {
      "entropy": 5.939534091949463,
      "epoch": 2.414576597046229,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004415205581502996,
      "loss": 5.2038,
      "mean_token_accuracy": 0.21259273290634156,
      "num_tokens": 55992196.0,
      "step": 27630
    },
    {
      "entropy": 5.9879899501800535,
      "epoch": 2.415013545398934,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044149960051771643,
      "loss": 5.2374,
      "mean_token_accuracy": 0.2057455599308014,
      "num_tokens": 56002295.0,
      "step": 27635
    },
    {
      "entropy": 6.00427279472351,
      "epoch": 2.4154504937516386,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044147863969158353,
      "loss": 5.2404,
      "mean_token_accuracy": 0.19927303344011307,
      "num_tokens": 56011847.0,
      "step": 27640
    },
    {
      "entropy": 5.945450305938721,
      "epoch": 2.415887442104343,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004414576756723032,
      "loss": 5.2067,
      "mean_token_accuracy": 0.20953072905540465,
      "num_tokens": 56021717.0,
      "step": 27645
    },
    {
      "entropy": 5.987042379379273,
      "epoch": 2.416324390457048,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004414367084602774,
      "loss": 5.2777,
      "mean_token_accuracy": 0.20314341634511948,
      "num_tokens": 56031845.0,
      "step": 27650
    },
    {
      "entropy": 5.973042297363281,
      "epoch": 2.416761338809753,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004414157380559083,
      "loss": 5.2742,
      "mean_token_accuracy": 0.20535965859889985,
      "num_tokens": 56042082.0,
      "step": 27655
    },
    {
      "entropy": 5.950490999221802,
      "epoch": 2.4171982871624573,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044139476445959817,
      "loss": 5.271,
      "mean_token_accuracy": 0.1980453684926033,
      "num_tokens": 56052592.0,
      "step": 27660
    },
    {
      "entropy": 5.9896735668182375,
      "epoch": 2.4176352355151622,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004413737876717492,
      "loss": 5.3492,
      "mean_token_accuracy": 0.19451983273029327,
      "num_tokens": 56062500.0,
      "step": 27665
    },
    {
      "entropy": 5.9700404644012455,
      "epoch": 2.4180721838678667,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004413528076927639,
      "loss": 5.1866,
      "mean_token_accuracy": 0.20411760807037355,
      "num_tokens": 56072912.0,
      "step": 27670
    },
    {
      "entropy": 6.000873851776123,
      "epoch": 2.4185091322205716,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004413318245230446,
      "loss": 5.3257,
      "mean_token_accuracy": 0.191184900701046,
      "num_tokens": 56082591.0,
      "step": 27675
    },
    {
      "entropy": 5.897507905960083,
      "epoch": 2.418946080573276,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004413108381629936,
      "loss": 5.1679,
      "mean_token_accuracy": 0.21185963302850724,
      "num_tokens": 56091738.0,
      "step": 27680
    },
    {
      "entropy": 5.94038667678833,
      "epoch": 2.419383028925981,
      "grad_norm": 1.375,
      "learning_rate": 0.0004412898486130137,
      "loss": 5.2929,
      "mean_token_accuracy": 0.20092814266681672,
      "num_tokens": 56102011.0,
      "step": 27685
    },
    {
      "entropy": 5.9888725757598875,
      "epoch": 2.4198199772786855,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044126885587350735,
      "loss": 5.2737,
      "mean_token_accuracy": 0.20897216349840164,
      "num_tokens": 56112901.0,
      "step": 27690
    },
    {
      "entropy": 6.003911542892456,
      "epoch": 2.4202569256313904,
      "grad_norm": 1.125,
      "learning_rate": 0.0004412478599448772,
      "loss": 5.2606,
      "mean_token_accuracy": 0.2044934332370758,
      "num_tokens": 56123598.0,
      "step": 27695
    },
    {
      "entropy": 5.943812561035156,
      "epoch": 2.420693873984095,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000441226860827526,
      "loss": 5.2774,
      "mean_token_accuracy": 0.1983986333012581,
      "num_tokens": 56133686.0,
      "step": 27700
    },
    {
      "entropy": 5.995155906677246,
      "epoch": 2.4211308223367998,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044120585852185654,
      "loss": 5.2856,
      "mean_token_accuracy": 0.19937517046928405,
      "num_tokens": 56143722.0,
      "step": 27705
    },
    {
      "entropy": 5.98631443977356,
      "epoch": 2.4215677706895047,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044118485302827156,
      "loss": 5.2328,
      "mean_token_accuracy": 0.20925222635269164,
      "num_tokens": 56153781.0,
      "step": 27710
    },
    {
      "entropy": 5.88372573852539,
      "epoch": 2.422004719042209,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044116384434717405,
      "loss": 5.2435,
      "mean_token_accuracy": 0.2069736272096634,
      "num_tokens": 56163398.0,
      "step": 27715
    },
    {
      "entropy": 5.994554805755615,
      "epoch": 2.422441667394914,
      "grad_norm": 1.25,
      "learning_rate": 0.0004411428324789669,
      "loss": 5.3096,
      "mean_token_accuracy": 0.20300259739160537,
      "num_tokens": 56173418.0,
      "step": 27720
    },
    {
      "entropy": 6.034494924545288,
      "epoch": 2.4228786157476185,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044112181742405324,
      "loss": 5.3549,
      "mean_token_accuracy": 0.1930885449051857,
      "num_tokens": 56184547.0,
      "step": 27725
    },
    {
      "entropy": 6.017813062667846,
      "epoch": 2.4233155641003234,
      "grad_norm": 1.203125,
      "learning_rate": 0.000441100799182836,
      "loss": 5.2569,
      "mean_token_accuracy": 0.2021833524107933,
      "num_tokens": 56195404.0,
      "step": 27730
    },
    {
      "entropy": 5.891440391540527,
      "epoch": 2.423752512453028,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004410797777557184,
      "loss": 5.2167,
      "mean_token_accuracy": 0.21284371614456177,
      "num_tokens": 56205552.0,
      "step": 27735
    },
    {
      "entropy": 5.976535177230835,
      "epoch": 2.424189460805733,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044105875314310374,
      "loss": 5.2252,
      "mean_token_accuracy": 0.20540145486593248,
      "num_tokens": 56215581.0,
      "step": 27740
    },
    {
      "entropy": 5.9424175262451175,
      "epoch": 2.4246264091584373,
      "grad_norm": 1.375,
      "learning_rate": 0.00044103772534539505,
      "loss": 5.1862,
      "mean_token_accuracy": 0.2100246876478195,
      "num_tokens": 56225428.0,
      "step": 27745
    },
    {
      "entropy": 5.968746376037598,
      "epoch": 2.425063357511142,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004410166943629959,
      "loss": 5.2835,
      "mean_token_accuracy": 0.20053628981113433,
      "num_tokens": 56235284.0,
      "step": 27750
    },
    {
      "entropy": 5.982584619522095,
      "epoch": 2.4255003058638467,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004409956601963095,
      "loss": 5.2352,
      "mean_token_accuracy": 0.20928251445293428,
      "num_tokens": 56244444.0,
      "step": 27755
    },
    {
      "entropy": 5.940780735015869,
      "epoch": 2.4259372542165516,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004409746228457393,
      "loss": 5.2692,
      "mean_token_accuracy": 0.20864402502775192,
      "num_tokens": 56254983.0,
      "step": 27760
    },
    {
      "entropy": 5.979168939590454,
      "epoch": 2.4263742025692565,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004409535823116889,
      "loss": 5.2691,
      "mean_token_accuracy": 0.20385584831237794,
      "num_tokens": 56264307.0,
      "step": 27765
    },
    {
      "entropy": 5.983766746520996,
      "epoch": 2.426811150921961,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004409325385945619,
      "loss": 5.2781,
      "mean_token_accuracy": 0.19677289873361586,
      "num_tokens": 56274418.0,
      "step": 27770
    },
    {
      "entropy": 6.065314531326294,
      "epoch": 2.427248099274666,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004409114916947618,
      "loss": 5.3743,
      "mean_token_accuracy": 0.1919599324464798,
      "num_tokens": 56285665.0,
      "step": 27775
    },
    {
      "entropy": 5.961397171020508,
      "epoch": 2.4276850476273704,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044089044161269237,
      "loss": 5.2574,
      "mean_token_accuracy": 0.20390213429927825,
      "num_tokens": 56296320.0,
      "step": 27780
    },
    {
      "entropy": 5.980436134338379,
      "epoch": 2.4281219959800753,
      "grad_norm": 1.0,
      "learning_rate": 0.0004408693883487574,
      "loss": 5.364,
      "mean_token_accuracy": 0.1961873322725296,
      "num_tokens": 56308009.0,
      "step": 27785
    },
    {
      "entropy": 5.921389818191528,
      "epoch": 2.4285589443327797,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004408483319033606,
      "loss": 5.2705,
      "mean_token_accuracy": 0.2055211618542671,
      "num_tokens": 56318341.0,
      "step": 27790
    },
    {
      "entropy": 5.89641056060791,
      "epoch": 2.4289958926854847,
      "grad_norm": 1.125,
      "learning_rate": 0.0004408272722769059,
      "loss": 5.1951,
      "mean_token_accuracy": 0.2184273511171341,
      "num_tokens": 56328808.0,
      "step": 27795
    },
    {
      "entropy": 6.038774156570435,
      "epoch": 2.429432841038189,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004408062094697972,
      "loss": 5.2741,
      "mean_token_accuracy": 0.20639655888080596,
      "num_tokens": 56339043.0,
      "step": 27800
    },
    {
      "entropy": 5.998224496841431,
      "epoch": 2.429869789390894,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044078514348243846,
      "loss": 5.2879,
      "mean_token_accuracy": 0.2008794963359833,
      "num_tokens": 56349518.0,
      "step": 27805
    },
    {
      "entropy": 5.920193481445312,
      "epoch": 2.4303067377435985,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044076407431523394,
      "loss": 5.2337,
      "mean_token_accuracy": 0.20135708004236222,
      "num_tokens": 56358739.0,
      "step": 27810
    },
    {
      "entropy": 5.9422468662261965,
      "epoch": 2.4307436860963034,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004407430019685875,
      "loss": 5.2829,
      "mean_token_accuracy": 0.2028033822774887,
      "num_tokens": 56368049.0,
      "step": 27815
    },
    {
      "entropy": 6.020836448669433,
      "epoch": 2.4311806344490083,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044072192644290355,
      "loss": 5.2476,
      "mean_token_accuracy": 0.20227947682142258,
      "num_tokens": 56377607.0,
      "step": 27820
    },
    {
      "entropy": 5.959865522384644,
      "epoch": 2.431617582801713,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004407008477385862,
      "loss": 5.24,
      "mean_token_accuracy": 0.20960080921649932,
      "num_tokens": 56387188.0,
      "step": 27825
    },
    {
      "entropy": 5.991115283966065,
      "epoch": 2.4320545311544177,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044067976585603964,
      "loss": 5.358,
      "mean_token_accuracy": 0.1927933394908905,
      "num_tokens": 56396389.0,
      "step": 27830
    },
    {
      "entropy": 5.985091257095337,
      "epoch": 2.432491479507122,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004406586807956684,
      "loss": 5.3018,
      "mean_token_accuracy": 0.20266139954328538,
      "num_tokens": 56406909.0,
      "step": 27835
    },
    {
      "entropy": 5.986688756942749,
      "epoch": 2.432928427859827,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044063759255787686,
      "loss": 5.296,
      "mean_token_accuracy": 0.1975620687007904,
      "num_tokens": 56416373.0,
      "step": 27840
    },
    {
      "entropy": 6.016581296920776,
      "epoch": 2.4333653762125316,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044061650114306954,
      "loss": 5.3286,
      "mean_token_accuracy": 0.19839755594730377,
      "num_tokens": 56426759.0,
      "step": 27845
    },
    {
      "entropy": 6.013709163665771,
      "epoch": 2.4338023245652365,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044059540655165097,
      "loss": 5.2987,
      "mean_token_accuracy": 0.20241316705942153,
      "num_tokens": 56436783.0,
      "step": 27850
    },
    {
      "entropy": 6.016373777389527,
      "epoch": 2.434239272917941,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044057430878402564,
      "loss": 5.2825,
      "mean_token_accuracy": 0.19659881442785263,
      "num_tokens": 56447191.0,
      "step": 27855
    },
    {
      "entropy": 5.916089296340942,
      "epoch": 2.434676221270646,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004405532078405984,
      "loss": 5.3258,
      "mean_token_accuracy": 0.19390515834093094,
      "num_tokens": 56457119.0,
      "step": 27860
    },
    {
      "entropy": 5.991366338729859,
      "epoch": 2.4351131696233503,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004405321037217739,
      "loss": 5.2225,
      "mean_token_accuracy": 0.20819111764431,
      "num_tokens": 56467139.0,
      "step": 27865
    },
    {
      "entropy": 6.039913177490234,
      "epoch": 2.4355501179760553,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004405109964279568,
      "loss": 5.2578,
      "mean_token_accuracy": 0.20478453040122985,
      "num_tokens": 56477188.0,
      "step": 27870
    },
    {
      "entropy": 6.016380167007446,
      "epoch": 2.43598706632876,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044048988595955216,
      "loss": 5.4088,
      "mean_token_accuracy": 0.1892642393708229,
      "num_tokens": 56487594.0,
      "step": 27875
    },
    {
      "entropy": 5.96146788597107,
      "epoch": 2.4364240146814646,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004404687723169647,
      "loss": 5.2851,
      "mean_token_accuracy": 0.20335555523633958,
      "num_tokens": 56498008.0,
      "step": 27880
    },
    {
      "entropy": 5.9409363746643065,
      "epoch": 2.4368609630341695,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044044765550059955,
      "loss": 5.1828,
      "mean_token_accuracy": 0.20481159090995787,
      "num_tokens": 56508409.0,
      "step": 27885
    },
    {
      "entropy": 5.987059545516968,
      "epoch": 2.437297911386874,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004404265355108616,
      "loss": 5.2714,
      "mean_token_accuracy": 0.1995624214410782,
      "num_tokens": 56519513.0,
      "step": 27890
    },
    {
      "entropy": 5.987036180496216,
      "epoch": 2.437734859739579,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004404054123481561,
      "loss": 5.2397,
      "mean_token_accuracy": 0.20134842544794082,
      "num_tokens": 56528549.0,
      "step": 27895
    },
    {
      "entropy": 5.948874759674072,
      "epoch": 2.4381718080922834,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044038428601288817,
      "loss": 5.2512,
      "mean_token_accuracy": 0.20307360887527465,
      "num_tokens": 56537859.0,
      "step": 27900
    },
    {
      "entropy": 5.957953691482544,
      "epoch": 2.4386087564449883,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044036315650546287,
      "loss": 5.3209,
      "mean_token_accuracy": 0.19439662843942643,
      "num_tokens": 56547326.0,
      "step": 27905
    },
    {
      "entropy": 6.055893325805664,
      "epoch": 2.439045704797693,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044034202382628565,
      "loss": 5.3429,
      "mean_token_accuracy": 0.19971491992473603,
      "num_tokens": 56557460.0,
      "step": 27910
    },
    {
      "entropy": 5.934233331680298,
      "epoch": 2.4394826531503977,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004403208879757617,
      "loss": 5.2497,
      "mean_token_accuracy": 0.20713906586170197,
      "num_tokens": 56566928.0,
      "step": 27915
    },
    {
      "entropy": 5.863408851623535,
      "epoch": 2.439919601503102,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004402997489542966,
      "loss": 5.2266,
      "mean_token_accuracy": 0.2136493593454361,
      "num_tokens": 56577576.0,
      "step": 27920
    },
    {
      "entropy": 6.051094675064087,
      "epoch": 2.440356549855807,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004402786067622955,
      "loss": 5.367,
      "mean_token_accuracy": 0.19621023535728455,
      "num_tokens": 56587744.0,
      "step": 27925
    },
    {
      "entropy": 6.025964307785034,
      "epoch": 2.440793498208512,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004402574614001643,
      "loss": 5.2717,
      "mean_token_accuracy": 0.20673415213823318,
      "num_tokens": 56598272.0,
      "step": 27930
    },
    {
      "entropy": 5.8980628490448,
      "epoch": 2.4412304465612165,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044023631286830835,
      "loss": 5.2297,
      "mean_token_accuracy": 0.20686972588300706,
      "num_tokens": 56609346.0,
      "step": 27935
    },
    {
      "entropy": 6.007078647613525,
      "epoch": 2.441667394913921,
      "grad_norm": 1.25,
      "learning_rate": 0.00044021516116713334,
      "loss": 5.2392,
      "mean_token_accuracy": 0.2038705825805664,
      "num_tokens": 56619210.0,
      "step": 27940
    },
    {
      "entropy": 6.032521772384643,
      "epoch": 2.442104343266626,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000440194006297045,
      "loss": 5.2774,
      "mean_token_accuracy": 0.20414779484272003,
      "num_tokens": 56629135.0,
      "step": 27945
    },
    {
      "entropy": 6.00271635055542,
      "epoch": 2.4425412916193308,
      "grad_norm": 1.1875,
      "learning_rate": 0.000440172848258449,
      "loss": 5.2334,
      "mean_token_accuracy": 0.20344859808683396,
      "num_tokens": 56638900.0,
      "step": 27950
    },
    {
      "entropy": 6.008350133895874,
      "epoch": 2.4429782399720352,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044015168705175117,
      "loss": 5.3044,
      "mean_token_accuracy": 0.20024186074733735,
      "num_tokens": 56648605.0,
      "step": 27955
    },
    {
      "entropy": 5.983518981933594,
      "epoch": 2.44341518832474,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004401305226773576,
      "loss": 5.3375,
      "mean_token_accuracy": 0.20226116627454757,
      "num_tokens": 56659556.0,
      "step": 27960
    },
    {
      "entropy": 6.065122413635254,
      "epoch": 2.4438521366774446,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004401093551356739,
      "loss": 5.3903,
      "mean_token_accuracy": 0.1931208312511444,
      "num_tokens": 56669727.0,
      "step": 27965
    },
    {
      "entropy": 6.011923456192017,
      "epoch": 2.4442890850301495,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004400881844271064,
      "loss": 5.3509,
      "mean_token_accuracy": 0.19224983155727388,
      "num_tokens": 56681308.0,
      "step": 27970
    },
    {
      "entropy": 6.021063804626465,
      "epoch": 2.444726033382854,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044006701055206095,
      "loss": 5.3036,
      "mean_token_accuracy": 0.20315070897340776,
      "num_tokens": 56691089.0,
      "step": 27975
    },
    {
      "entropy": 5.970992612838745,
      "epoch": 2.445162981735559,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004400458335109437,
      "loss": 5.2502,
      "mean_token_accuracy": 0.2092625916004181,
      "num_tokens": 56701599.0,
      "step": 27980
    },
    {
      "entropy": 5.9319713592529295,
      "epoch": 2.4455999300882634,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044002465330416095,
      "loss": 5.1766,
      "mean_token_accuracy": 0.21666321456432341,
      "num_tokens": 56711890.0,
      "step": 27985
    },
    {
      "entropy": 5.966093873977661,
      "epoch": 2.4460368784409683,
      "grad_norm": 1.125,
      "learning_rate": 0.0004400034699321189,
      "loss": 5.2322,
      "mean_token_accuracy": 0.20880955308675767,
      "num_tokens": 56722003.0,
      "step": 27990
    },
    {
      "entropy": 5.986715936660767,
      "epoch": 2.4464738267936728,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004399822833952238,
      "loss": 5.2056,
      "mean_token_accuracy": 0.20819285660982131,
      "num_tokens": 56731636.0,
      "step": 27995
    },
    {
      "entropy": 5.917466402053833,
      "epoch": 2.4469107751463777,
      "grad_norm": 1.140625,
      "learning_rate": 0.000439961093693882,
      "loss": 5.1927,
      "mean_token_accuracy": 0.20923107266426086,
      "num_tokens": 56741973.0,
      "step": 28000
    },
    {
      "entropy": 6.000338125228882,
      "epoch": 2.4473477234990826,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004399399008285001,
      "loss": 5.3544,
      "mean_token_accuracy": 0.19914155453443527,
      "num_tokens": 56751441.0,
      "step": 28005
    },
    {
      "entropy": 5.987693166732788,
      "epoch": 2.447784671851787,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004399187047994844,
      "loss": 5.283,
      "mean_token_accuracy": 0.20798686295747756,
      "num_tokens": 56761156.0,
      "step": 28010
    },
    {
      "entropy": 5.994379711151123,
      "epoch": 2.448221620204492,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043989750560724153,
      "loss": 5.2745,
      "mean_token_accuracy": 0.20323581993579865,
      "num_tokens": 56771414.0,
      "step": 28015
    },
    {
      "entropy": 5.87394528388977,
      "epoch": 2.4486585685571964,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004398763032521781,
      "loss": 5.1818,
      "mean_token_accuracy": 0.2118681937456131,
      "num_tokens": 56781572.0,
      "step": 28020
    },
    {
      "entropy": 6.017564487457276,
      "epoch": 2.4490955169099013,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043985509773470077,
      "loss": 5.3263,
      "mean_token_accuracy": 0.1956464648246765,
      "num_tokens": 56791299.0,
      "step": 28025
    },
    {
      "entropy": 5.980851125717163,
      "epoch": 2.449532465262606,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004398338890552163,
      "loss": 5.2262,
      "mean_token_accuracy": 0.2012437880039215,
      "num_tokens": 56802485.0,
      "step": 28030
    },
    {
      "entropy": 5.976202392578125,
      "epoch": 2.4499694136153107,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004398126772141315,
      "loss": 5.3005,
      "mean_token_accuracy": 0.193277907371521,
      "num_tokens": 56811694.0,
      "step": 28035
    },
    {
      "entropy": 5.938247394561768,
      "epoch": 2.450406361968015,
      "grad_norm": 1.25,
      "learning_rate": 0.00043979146221185313,
      "loss": 5.2053,
      "mean_token_accuracy": 0.20650685727596282,
      "num_tokens": 56821814.0,
      "step": 28040
    },
    {
      "entropy": 5.984222364425659,
      "epoch": 2.45084331032072,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004397702440487882,
      "loss": 5.271,
      "mean_token_accuracy": 0.20605750679969786,
      "num_tokens": 56832257.0,
      "step": 28045
    },
    {
      "entropy": 6.003148221969605,
      "epoch": 2.4512802586734246,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004397490227253436,
      "loss": 5.2929,
      "mean_token_accuracy": 0.2001851886510849,
      "num_tokens": 56841789.0,
      "step": 28050
    },
    {
      "entropy": 5.984490299224854,
      "epoch": 2.4517172070261295,
      "grad_norm": 1.125,
      "learning_rate": 0.0004397277982419265,
      "loss": 5.2533,
      "mean_token_accuracy": 0.20134703814983368,
      "num_tokens": 56851612.0,
      "step": 28055
    },
    {
      "entropy": 5.964906692504883,
      "epoch": 2.4521541553788344,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043970657059894385,
      "loss": 5.2124,
      "mean_token_accuracy": 0.20949293226003646,
      "num_tokens": 56861812.0,
      "step": 28060
    },
    {
      "entropy": 5.961755466461182,
      "epoch": 2.452591103731539,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000439685339796803,
      "loss": 5.2042,
      "mean_token_accuracy": 0.21225310117006302,
      "num_tokens": 56871873.0,
      "step": 28065
    },
    {
      "entropy": 6.038943767547607,
      "epoch": 2.453028052084244,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004396641058359109,
      "loss": 5.3135,
      "mean_token_accuracy": 0.19489022642374038,
      "num_tokens": 56883579.0,
      "step": 28070
    },
    {
      "entropy": 5.940277194976806,
      "epoch": 2.4534650004369483,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043964286871667495,
      "loss": 5.2349,
      "mean_token_accuracy": 0.2060253754258156,
      "num_tokens": 56894305.0,
      "step": 28075
    },
    {
      "entropy": 5.981283473968506,
      "epoch": 2.453901948789653,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004396216284395026,
      "loss": 5.2335,
      "mean_token_accuracy": 0.20085926204919816,
      "num_tokens": 56904393.0,
      "step": 28080
    },
    {
      "entropy": 5.988172197341919,
      "epoch": 2.4543388971423576,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004396003850048011,
      "loss": 5.2646,
      "mean_token_accuracy": 0.20238866060972213,
      "num_tokens": 56914600.0,
      "step": 28085
    },
    {
      "entropy": 6.010832357406616,
      "epoch": 2.4547758454950626,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004395791384129779,
      "loss": 5.3429,
      "mean_token_accuracy": 0.20506860464811325,
      "num_tokens": 56925948.0,
      "step": 28090
    },
    {
      "entropy": 5.961394929885865,
      "epoch": 2.455212793847767,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043955788866444073,
      "loss": 5.1958,
      "mean_token_accuracy": 0.20157287269830704,
      "num_tokens": 56935395.0,
      "step": 28095
    },
    {
      "entropy": 5.931991815567017,
      "epoch": 2.455649742200472,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043953663575959696,
      "loss": 5.2011,
      "mean_token_accuracy": 0.21531972289085388,
      "num_tokens": 56944769.0,
      "step": 28100
    },
    {
      "entropy": 5.918155765533447,
      "epoch": 2.4560866905531764,
      "grad_norm": 1.125,
      "learning_rate": 0.0004395153796988543,
      "loss": 5.2181,
      "mean_token_accuracy": 0.20924417823553085,
      "num_tokens": 56954352.0,
      "step": 28105
    },
    {
      "entropy": 5.950512886047363,
      "epoch": 2.4565236389058813,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004394941204826205,
      "loss": 5.219,
      "mean_token_accuracy": 0.20201366990804673,
      "num_tokens": 56963759.0,
      "step": 28110
    },
    {
      "entropy": 5.966155004501343,
      "epoch": 2.4569605872585862,
      "grad_norm": 1.125,
      "learning_rate": 0.00043947285811130315,
      "loss": 5.2133,
      "mean_token_accuracy": 0.20377452671527863,
      "num_tokens": 56974962.0,
      "step": 28115
    },
    {
      "entropy": 5.969319295883179,
      "epoch": 2.4573975356112907,
      "grad_norm": 1.125,
      "learning_rate": 0.00043945159258531035,
      "loss": 5.2889,
      "mean_token_accuracy": 0.19842729419469834,
      "num_tokens": 56985637.0,
      "step": 28120
    },
    {
      "entropy": 5.983480310440063,
      "epoch": 2.4578344839639956,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004394303239050498,
      "loss": 5.2682,
      "mean_token_accuracy": 0.2097742110490799,
      "num_tokens": 56995779.0,
      "step": 28125
    },
    {
      "entropy": 5.972747373580932,
      "epoch": 2.4582714323167,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004394090520709294,
      "loss": 5.2338,
      "mean_token_accuracy": 0.20566672682762147,
      "num_tokens": 57005506.0,
      "step": 28130
    },
    {
      "entropy": 5.995728540420532,
      "epoch": 2.458708380669405,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043938777708335726,
      "loss": 5.3373,
      "mean_token_accuracy": 0.19448368698358537,
      "num_tokens": 57016001.0,
      "step": 28135
    },
    {
      "entropy": 5.985873031616211,
      "epoch": 2.4591453290221095,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043936649894274146,
      "loss": 5.284,
      "mean_token_accuracy": 0.20568411499261857,
      "num_tokens": 57026418.0,
      "step": 28140
    },
    {
      "entropy": 6.022682857513428,
      "epoch": 2.4595822773748144,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043934521764949,
      "loss": 5.297,
      "mean_token_accuracy": 0.20218710452318192,
      "num_tokens": 57037170.0,
      "step": 28145
    },
    {
      "entropy": 5.97733588218689,
      "epoch": 2.460019225727519,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004393239332040113,
      "loss": 5.2026,
      "mean_token_accuracy": 0.20673907995224,
      "num_tokens": 57046775.0,
      "step": 28150
    },
    {
      "entropy": 5.926603078842163,
      "epoch": 2.4604561740802238,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043930264560671336,
      "loss": 5.2111,
      "mean_token_accuracy": 0.2074948102235794,
      "num_tokens": 57056360.0,
      "step": 28155
    },
    {
      "entropy": 5.958480453491211,
      "epoch": 2.4608931224329282,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004392813548580045,
      "loss": 5.225,
      "mean_token_accuracy": 0.2050098404288292,
      "num_tokens": 57066355.0,
      "step": 28160
    },
    {
      "entropy": 5.945051765441894,
      "epoch": 2.461330070785633,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004392600609582933,
      "loss": 5.2698,
      "mean_token_accuracy": 0.20578551590442656,
      "num_tokens": 57077158.0,
      "step": 28165
    },
    {
      "entropy": 5.987356185913086,
      "epoch": 2.461767019138338,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000439238763907988,
      "loss": 5.3318,
      "mean_token_accuracy": 0.204252253472805,
      "num_tokens": 57088005.0,
      "step": 28170
    },
    {
      "entropy": 5.946875715255738,
      "epoch": 2.4622039674910425,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004392174637074972,
      "loss": 5.2435,
      "mean_token_accuracy": 0.20690279453992844,
      "num_tokens": 57098399.0,
      "step": 28175
    },
    {
      "entropy": 5.976794242858887,
      "epoch": 2.4626409158437474,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004391961603572294,
      "loss": 5.2233,
      "mean_token_accuracy": 0.20992447584867477,
      "num_tokens": 57108949.0,
      "step": 28180
    },
    {
      "entropy": 5.959930562973023,
      "epoch": 2.463077864196452,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043917485385759316,
      "loss": 5.2914,
      "mean_token_accuracy": 0.20248962193727493,
      "num_tokens": 57119441.0,
      "step": 28185
    },
    {
      "entropy": 5.978952264785766,
      "epoch": 2.463514812549157,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004391535442089971,
      "loss": 5.2792,
      "mean_token_accuracy": 0.21067195534706115,
      "num_tokens": 57129507.0,
      "step": 28190
    },
    {
      "entropy": 5.973205947875977,
      "epoch": 2.4639517609018613,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004391322314118502,
      "loss": 5.25,
      "mean_token_accuracy": 0.20356573015451432,
      "num_tokens": 57140490.0,
      "step": 28195
    },
    {
      "entropy": 5.939560604095459,
      "epoch": 2.464388709254566,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043911091546656105,
      "loss": 5.2066,
      "mean_token_accuracy": 0.20802740901708602,
      "num_tokens": 57150706.0,
      "step": 28200
    },
    {
      "entropy": 5.946043872833252,
      "epoch": 2.4648256576072707,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043908959637353864,
      "loss": 5.2606,
      "mean_token_accuracy": 0.20026109665632247,
      "num_tokens": 57160259.0,
      "step": 28205
    },
    {
      "entropy": 5.955775880813599,
      "epoch": 2.4652626059599756,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043906827413319166,
      "loss": 5.2491,
      "mean_token_accuracy": 0.20829489529132844,
      "num_tokens": 57171831.0,
      "step": 28210
    },
    {
      "entropy": 6.019066286087036,
      "epoch": 2.46569955431268,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004390469487459292,
      "loss": 5.3535,
      "mean_token_accuracy": 0.1980710431933403,
      "num_tokens": 57182172.0,
      "step": 28215
    },
    {
      "entropy": 5.939724969863891,
      "epoch": 2.466136502665385,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043902562021216033,
      "loss": 5.2338,
      "mean_token_accuracy": 0.2107057422399521,
      "num_tokens": 57191633.0,
      "step": 28220
    },
    {
      "entropy": 5.998600482940674,
      "epoch": 2.46657345101809,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004390042885322942,
      "loss": 5.2648,
      "mean_token_accuracy": 0.20537398755550385,
      "num_tokens": 57200953.0,
      "step": 28225
    },
    {
      "entropy": 6.028683185577393,
      "epoch": 2.4670103993707944,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043898295370673986,
      "loss": 5.3347,
      "mean_token_accuracy": 0.19980351775884628,
      "num_tokens": 57210893.0,
      "step": 28230
    },
    {
      "entropy": 5.934898948669433,
      "epoch": 2.4674473477234993,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043896161573590646,
      "loss": 5.2554,
      "mean_token_accuracy": 0.21118145287036896,
      "num_tokens": 57221458.0,
      "step": 28235
    },
    {
      "entropy": 5.984973526000976,
      "epoch": 2.4678842960762037,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004389402746202034,
      "loss": 5.3053,
      "mean_token_accuracy": 0.19925461560487748,
      "num_tokens": 57231450.0,
      "step": 28240
    },
    {
      "entropy": 6.012554407119751,
      "epoch": 2.4683212444289087,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043891893036004,
      "loss": 5.2821,
      "mean_token_accuracy": 0.20782667249441147,
      "num_tokens": 57242820.0,
      "step": 28245
    },
    {
      "entropy": 6.0824981212615965,
      "epoch": 2.468758192781613,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004388975829558256,
      "loss": 5.2919,
      "mean_token_accuracy": 0.20268598943948746,
      "num_tokens": 57253465.0,
      "step": 28250
    },
    {
      "entropy": 5.942614936828614,
      "epoch": 2.469195141134318,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043887623240796965,
      "loss": 5.2579,
      "mean_token_accuracy": 0.20501359701156616,
      "num_tokens": 57263521.0,
      "step": 28255
    },
    {
      "entropy": 5.9546120166778564,
      "epoch": 2.4696320894870225,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004388548787168817,
      "loss": 5.2632,
      "mean_token_accuracy": 0.19704711586236953,
      "num_tokens": 57273568.0,
      "step": 28260
    },
    {
      "entropy": 6.066166830062866,
      "epoch": 2.4700690378397274,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004388335218829713,
      "loss": 5.3221,
      "mean_token_accuracy": 0.20259978473186493,
      "num_tokens": 57283960.0,
      "step": 28265
    },
    {
      "entropy": 5.974500370025635,
      "epoch": 2.470505986192432,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004388121619066482,
      "loss": 5.2592,
      "mean_token_accuracy": 0.20231667757034302,
      "num_tokens": 57294672.0,
      "step": 28270
    },
    {
      "entropy": 5.964369630813598,
      "epoch": 2.470942934545137,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004387907987883219,
      "loss": 5.264,
      "mean_token_accuracy": 0.20730752795934676,
      "num_tokens": 57305594.0,
      "step": 28275
    },
    {
      "entropy": 5.951103925704956,
      "epoch": 2.4713798828978417,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004387694325284023,
      "loss": 5.2525,
      "mean_token_accuracy": 0.21027169227600098,
      "num_tokens": 57315250.0,
      "step": 28280
    },
    {
      "entropy": 5.99584264755249,
      "epoch": 2.471816831250546,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004387480631272991,
      "loss": 5.2965,
      "mean_token_accuracy": 0.1988244742155075,
      "num_tokens": 57326750.0,
      "step": 28285
    },
    {
      "entropy": 6.0372536182403564,
      "epoch": 2.4722537796032507,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004387266905854223,
      "loss": 5.2552,
      "mean_token_accuracy": 0.20225838273763658,
      "num_tokens": 57336151.0,
      "step": 28290
    },
    {
      "entropy": 5.979277753829956,
      "epoch": 2.4726907279559556,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004387053149031819,
      "loss": 5.2205,
      "mean_token_accuracy": 0.20809998363256454,
      "num_tokens": 57346761.0,
      "step": 28295
    },
    {
      "entropy": 5.977178001403809,
      "epoch": 2.4731276763086605,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004386839360809876,
      "loss": 5.3153,
      "mean_token_accuracy": 0.2022350996732712,
      "num_tokens": 57356654.0,
      "step": 28300
    },
    {
      "entropy": 5.9331374168396,
      "epoch": 2.473564624661365,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004386625541192497,
      "loss": 5.2841,
      "mean_token_accuracy": 0.199643774330616,
      "num_tokens": 57367888.0,
      "step": 28305
    },
    {
      "entropy": 5.974944591522217,
      "epoch": 2.47400157301407,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043864116901837826,
      "loss": 5.2305,
      "mean_token_accuracy": 0.21333955824375153,
      "num_tokens": 57377167.0,
      "step": 28310
    },
    {
      "entropy": 5.931625556945801,
      "epoch": 2.4744385213667743,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043861978077878343,
      "loss": 5.2355,
      "mean_token_accuracy": 0.20630238801240922,
      "num_tokens": 57389076.0,
      "step": 28315
    },
    {
      "entropy": 5.970355463027954,
      "epoch": 2.4748754697194792,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004385983894008755,
      "loss": 5.2723,
      "mean_token_accuracy": 0.20190111100673674,
      "num_tokens": 57401136.0,
      "step": 28320
    },
    {
      "entropy": 5.96897439956665,
      "epoch": 2.4753124180721837,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043857699488506467,
      "loss": 5.2569,
      "mean_token_accuracy": 0.20297173857688905,
      "num_tokens": 57410548.0,
      "step": 28325
    },
    {
      "entropy": 5.9833556652069095,
      "epoch": 2.4757493664248886,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004385555972317615,
      "loss": 5.3153,
      "mean_token_accuracy": 0.1985708326101303,
      "num_tokens": 57420371.0,
      "step": 28330
    },
    {
      "entropy": 6.013678550720215,
      "epoch": 2.476186314777593,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043853419644137616,
      "loss": 5.3219,
      "mean_token_accuracy": 0.19805074781179427,
      "num_tokens": 57430079.0,
      "step": 28335
    },
    {
      "entropy": 5.989920139312744,
      "epoch": 2.476623263130298,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043851279251431937,
      "loss": 5.2623,
      "mean_token_accuracy": 0.20243007838726043,
      "num_tokens": 57440763.0,
      "step": 28340
    },
    {
      "entropy": 6.0696498394012455,
      "epoch": 2.4770602114830025,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043849138545100136,
      "loss": 5.3446,
      "mean_token_accuracy": 0.19775958806276323,
      "num_tokens": 57451008.0,
      "step": 28345
    },
    {
      "entropy": 6.004605293273926,
      "epoch": 2.4774971598357074,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043846997525183305,
      "loss": 5.3401,
      "mean_token_accuracy": 0.19361809343099595,
      "num_tokens": 57460852.0,
      "step": 28350
    },
    {
      "entropy": 6.032968664169312,
      "epoch": 2.4779341081884123,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043844856191722486,
      "loss": 5.2929,
      "mean_token_accuracy": 0.19733521938323975,
      "num_tokens": 57470027.0,
      "step": 28355
    },
    {
      "entropy": 6.011757516860962,
      "epoch": 2.478371056541117,
      "grad_norm": 1.125,
      "learning_rate": 0.00043842714544758764,
      "loss": 5.2012,
      "mean_token_accuracy": 0.20309304893016816,
      "num_tokens": 57479846.0,
      "step": 28360
    },
    {
      "entropy": 6.023183536529541,
      "epoch": 2.4788080048938217,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043840572584333216,
      "loss": 5.3262,
      "mean_token_accuracy": 0.1980222374200821,
      "num_tokens": 57490569.0,
      "step": 28365
    },
    {
      "entropy": 5.958676099777222,
      "epoch": 2.479244953246526,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004383843031048692,
      "loss": 5.3097,
      "mean_token_accuracy": 0.19732238054275514,
      "num_tokens": 57501048.0,
      "step": 28370
    },
    {
      "entropy": 6.032219171524048,
      "epoch": 2.479681901599231,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043836287723260977,
      "loss": 5.3183,
      "mean_token_accuracy": 0.19193998575210572,
      "num_tokens": 57511510.0,
      "step": 28375
    },
    {
      "entropy": 5.970440673828125,
      "epoch": 2.4801188499519355,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043834144822696463,
      "loss": 5.2806,
      "mean_token_accuracy": 0.20047815442085265,
      "num_tokens": 57522229.0,
      "step": 28380
    },
    {
      "entropy": 5.983261632919311,
      "epoch": 2.4805557983046405,
      "grad_norm": 1.109375,
      "learning_rate": 0.000438320016088345,
      "loss": 5.2222,
      "mean_token_accuracy": 0.2052251547574997,
      "num_tokens": 57532618.0,
      "step": 28385
    },
    {
      "entropy": 5.9771812915802,
      "epoch": 2.480992746657345,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004382985808171619,
      "loss": 5.279,
      "mean_token_accuracy": 0.20493919551372528,
      "num_tokens": 57542698.0,
      "step": 28390
    },
    {
      "entropy": 5.922773838043213,
      "epoch": 2.48142969501005,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043827714241382644,
      "loss": 5.1436,
      "mean_token_accuracy": 0.21135137230157852,
      "num_tokens": 57552556.0,
      "step": 28395
    },
    {
      "entropy": 6.075367832183838,
      "epoch": 2.4818666433627543,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004382557008787498,
      "loss": 5.3016,
      "mean_token_accuracy": 0.20356569439172745,
      "num_tokens": 57562830.0,
      "step": 28400
    },
    {
      "entropy": 6.002661609649659,
      "epoch": 2.4823035917154592,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004382342562123433,
      "loss": 5.2552,
      "mean_token_accuracy": 0.20457461774349212,
      "num_tokens": 57572584.0,
      "step": 28405
    },
    {
      "entropy": 5.908700656890869,
      "epoch": 2.482740540068164,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004382128084150182,
      "loss": 5.247,
      "mean_token_accuracy": 0.19934695810079575,
      "num_tokens": 57582571.0,
      "step": 28410
    },
    {
      "entropy": 5.967271423339843,
      "epoch": 2.4831774884208686,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043819135748718595,
      "loss": 5.3095,
      "mean_token_accuracy": 0.2024248108267784,
      "num_tokens": 57593267.0,
      "step": 28415
    },
    {
      "entropy": 5.98675684928894,
      "epoch": 2.4836144367735735,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043816990342925797,
      "loss": 5.3277,
      "mean_token_accuracy": 0.1903754726052284,
      "num_tokens": 57603708.0,
      "step": 28420
    },
    {
      "entropy": 6.028821563720703,
      "epoch": 2.484051385126278,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043814844624164576,
      "loss": 5.3514,
      "mean_token_accuracy": 0.197670379281044,
      "num_tokens": 57613523.0,
      "step": 28425
    },
    {
      "entropy": 5.954929637908935,
      "epoch": 2.484488333478983,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004381269859247607,
      "loss": 5.244,
      "mean_token_accuracy": 0.21379541456699372,
      "num_tokens": 57624181.0,
      "step": 28430
    },
    {
      "entropy": 5.9918437004089355,
      "epoch": 2.4849252818316874,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043810552247901475,
      "loss": 5.2325,
      "mean_token_accuracy": 0.20053109377622605,
      "num_tokens": 57633674.0,
      "step": 28435
    },
    {
      "entropy": 5.986803102493286,
      "epoch": 2.4853622301843923,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043808405590481934,
      "loss": 5.2763,
      "mean_token_accuracy": 0.2057914361357689,
      "num_tokens": 57644723.0,
      "step": 28440
    },
    {
      "entropy": 5.963625764846801,
      "epoch": 2.4857991785370968,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004380625862025863,
      "loss": 5.2419,
      "mean_token_accuracy": 0.20289125442504882,
      "num_tokens": 57654274.0,
      "step": 28445
    },
    {
      "entropy": 5.946555423736572,
      "epoch": 2.4862361268898017,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004380411133727274,
      "loss": 5.2055,
      "mean_token_accuracy": 0.20631152242422104,
      "num_tokens": 57664572.0,
      "step": 28450
    },
    {
      "entropy": 6.005164670944214,
      "epoch": 2.486673075242506,
      "grad_norm": 1.25,
      "learning_rate": 0.00043801963741565454,
      "loss": 5.33,
      "mean_token_accuracy": 0.19168623834848403,
      "num_tokens": 57674048.0,
      "step": 28455
    },
    {
      "entropy": 6.0123162269592285,
      "epoch": 2.487110023595211,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004379981583317796,
      "loss": 5.2885,
      "mean_token_accuracy": 0.20053085833787918,
      "num_tokens": 57685488.0,
      "step": 28460
    },
    {
      "entropy": 6.021380758285522,
      "epoch": 2.487546971947916,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043797667612151454,
      "loss": 5.2896,
      "mean_token_accuracy": 0.2044860988855362,
      "num_tokens": 57696044.0,
      "step": 28465
    },
    {
      "entropy": 5.968987560272216,
      "epoch": 2.4879839203006204,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043795519078527144,
      "loss": 5.2573,
      "mean_token_accuracy": 0.20973459482192994,
      "num_tokens": 57705702.0,
      "step": 28470
    },
    {
      "entropy": 5.9587338924407955,
      "epoch": 2.4884208686533253,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043793370232346246,
      "loss": 5.318,
      "mean_token_accuracy": 0.19669504314661027,
      "num_tokens": 57715628.0,
      "step": 28475
    },
    {
      "entropy": 6.022745132446289,
      "epoch": 2.48885781700603,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043791221073649964,
      "loss": 5.2912,
      "mean_token_accuracy": 0.20191202014684678,
      "num_tokens": 57725319.0,
      "step": 28480
    },
    {
      "entropy": 5.9843017101287845,
      "epoch": 2.4892947653587347,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043789071602479525,
      "loss": 5.3123,
      "mean_token_accuracy": 0.19947897493839264,
      "num_tokens": 57736483.0,
      "step": 28485
    },
    {
      "entropy": 6.007945394515991,
      "epoch": 2.489731713711439,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004378692181887616,
      "loss": 5.297,
      "mean_token_accuracy": 0.19833699464797974,
      "num_tokens": 57746667.0,
      "step": 28490
    },
    {
      "entropy": 5.941132354736328,
      "epoch": 2.490168662064144,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043784771722881095,
      "loss": 5.1732,
      "mean_token_accuracy": 0.21723474115133284,
      "num_tokens": 57755971.0,
      "step": 28495
    },
    {
      "entropy": 5.941589736938477,
      "epoch": 2.4906056104168486,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004378262131453558,
      "loss": 5.2265,
      "mean_token_accuracy": 0.21016687899827957,
      "num_tokens": 57765858.0,
      "step": 28500
    },
    {
      "entropy": 5.952618885040283,
      "epoch": 2.4910425587695535,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004378047059388086,
      "loss": 5.2774,
      "mean_token_accuracy": 0.20242150276899337,
      "num_tokens": 57775154.0,
      "step": 28505
    },
    {
      "entropy": 5.960286331176758,
      "epoch": 2.491479507122258,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004377831956095817,
      "loss": 5.2744,
      "mean_token_accuracy": 0.20340730249881744,
      "num_tokens": 57785416.0,
      "step": 28510
    },
    {
      "entropy": 5.9659607887268065,
      "epoch": 2.491916455474963,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000437761682158088,
      "loss": 5.2752,
      "mean_token_accuracy": 0.20187558233737946,
      "num_tokens": 57795483.0,
      "step": 28515
    },
    {
      "entropy": 6.009956359863281,
      "epoch": 2.492353403827668,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043774016558473976,
      "loss": 5.2666,
      "mean_token_accuracy": 0.20390488356351852,
      "num_tokens": 57805772.0,
      "step": 28520
    },
    {
      "entropy": 5.9951029300689695,
      "epoch": 2.4927903521803723,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043771864588994995,
      "loss": 5.2949,
      "mean_token_accuracy": 0.20097257941961288,
      "num_tokens": 57815514.0,
      "step": 28525
    },
    {
      "entropy": 5.904044342041016,
      "epoch": 2.493227300533077,
      "grad_norm": 1.125,
      "learning_rate": 0.0004376971230741313,
      "loss": 5.224,
      "mean_token_accuracy": 0.21041263937950133,
      "num_tokens": 57826312.0,
      "step": 28530
    },
    {
      "entropy": 6.004065275192261,
      "epoch": 2.4936642488857816,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004376755971376965,
      "loss": 5.3328,
      "mean_token_accuracy": 0.19513718783855438,
      "num_tokens": 57836718.0,
      "step": 28535
    },
    {
      "entropy": 6.037451267242432,
      "epoch": 2.4941011972384866,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004376540680810585,
      "loss": 5.2437,
      "mean_token_accuracy": 0.2065552979707718,
      "num_tokens": 57846833.0,
      "step": 28540
    },
    {
      "entropy": 6.038604831695556,
      "epoch": 2.494538145591191,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004376325359046304,
      "loss": 5.3301,
      "mean_token_accuracy": 0.19880367666482926,
      "num_tokens": 57856716.0,
      "step": 28545
    },
    {
      "entropy": 6.0363829135894775,
      "epoch": 2.494975093943896,
      "grad_norm": 1.375,
      "learning_rate": 0.00043761100060882486,
      "loss": 5.301,
      "mean_token_accuracy": 0.19664810746908187,
      "num_tokens": 57867079.0,
      "step": 28550
    },
    {
      "entropy": 5.978106164932251,
      "epoch": 2.4954120422966004,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004375894621940552,
      "loss": 5.2493,
      "mean_token_accuracy": 0.20764082819223403,
      "num_tokens": 57876593.0,
      "step": 28555
    },
    {
      "entropy": 5.9746791362762455,
      "epoch": 2.4958489906493053,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004375679206607344,
      "loss": 5.2717,
      "mean_token_accuracy": 0.20531344562768936,
      "num_tokens": 57886834.0,
      "step": 28560
    },
    {
      "entropy": 5.973783731460571,
      "epoch": 2.49628593900201,
      "grad_norm": 1.125,
      "learning_rate": 0.0004375463760092758,
      "loss": 5.2319,
      "mean_token_accuracy": 0.2087125912308693,
      "num_tokens": 57896766.0,
      "step": 28565
    },
    {
      "entropy": 5.981206941604614,
      "epoch": 2.4967228873547147,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004375248282400925,
      "loss": 5.2744,
      "mean_token_accuracy": 0.2034983217716217,
      "num_tokens": 57907259.0,
      "step": 28570
    },
    {
      "entropy": 6.0184996128082275,
      "epoch": 2.4971598357074196,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004375032773535977,
      "loss": 5.3938,
      "mean_token_accuracy": 0.191384856402874,
      "num_tokens": 57917940.0,
      "step": 28575
    },
    {
      "entropy": 6.037058734893799,
      "epoch": 2.497596784060124,
      "grad_norm": 1.109375,
      "learning_rate": 0.000437481723350205,
      "loss": 5.2838,
      "mean_token_accuracy": 0.20193682610988617,
      "num_tokens": 57927256.0,
      "step": 28580
    },
    {
      "entropy": 5.973752975463867,
      "epoch": 2.4980337324128286,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043746016623032776,
      "loss": 5.2514,
      "mean_token_accuracy": 0.20195016711950303,
      "num_tokens": 57936718.0,
      "step": 28585
    },
    {
      "entropy": 6.02273588180542,
      "epoch": 2.4984706807655335,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004374386059943794,
      "loss": 5.272,
      "mean_token_accuracy": 0.20313145369291305,
      "num_tokens": 57947548.0,
      "step": 28590
    },
    {
      "entropy": 5.984759712219239,
      "epoch": 2.4989076291182384,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004374170426427734,
      "loss": 5.2114,
      "mean_token_accuracy": 0.21048180460929872,
      "num_tokens": 57957598.0,
      "step": 28595
    },
    {
      "entropy": 6.023036623001099,
      "epoch": 2.499344577470943,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043739547617592344,
      "loss": 5.3409,
      "mean_token_accuracy": 0.1969071537256241,
      "num_tokens": 57967790.0,
      "step": 28600
    },
    {
      "entropy": 5.969751787185669,
      "epoch": 2.4997815258236478,
      "grad_norm": 1.125,
      "learning_rate": 0.00043737390659424316,
      "loss": 5.2638,
      "mean_token_accuracy": 0.20804286152124404,
      "num_tokens": 57977904.0,
      "step": 28605
    },
    {
      "entropy": 5.922207546234131,
      "epoch": 2.5002184741763522,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043735233389814623,
      "loss": 5.1512,
      "mean_token_accuracy": 0.21463654637336732,
      "num_tokens": 57987882.0,
      "step": 28610
    },
    {
      "entropy": 5.931307697296143,
      "epoch": 2.500655422529057,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043733075808804643,
      "loss": 5.2741,
      "mean_token_accuracy": 0.21171950548887253,
      "num_tokens": 57997376.0,
      "step": 28615
    },
    {
      "entropy": 5.924736976623535,
      "epoch": 2.5010923708817616,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043730917916435776,
      "loss": 5.1992,
      "mean_token_accuracy": 0.21713083684444429,
      "num_tokens": 58008115.0,
      "step": 28620
    },
    {
      "entropy": 6.0283119678497314,
      "epoch": 2.5015293192344665,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004372875971274939,
      "loss": 5.3514,
      "mean_token_accuracy": 0.1961859717965126,
      "num_tokens": 58017922.0,
      "step": 28625
    },
    {
      "entropy": 5.983604431152344,
      "epoch": 2.5019662675871714,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004372660119778689,
      "loss": 5.3045,
      "mean_token_accuracy": 0.1994615077972412,
      "num_tokens": 58028208.0,
      "step": 28630
    },
    {
      "entropy": 6.01278657913208,
      "epoch": 2.502403215939876,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004372444237158968,
      "loss": 5.3176,
      "mean_token_accuracy": 0.19670614898204802,
      "num_tokens": 58037237.0,
      "step": 28635
    },
    {
      "entropy": 5.959597492218018,
      "epoch": 2.5028401642925804,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004372228323419916,
      "loss": 5.2807,
      "mean_token_accuracy": 0.20451522022485732,
      "num_tokens": 58048083.0,
      "step": 28640
    },
    {
      "entropy": 6.017225456237793,
      "epoch": 2.5032771126452853,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004372012378565674,
      "loss": 5.3219,
      "mean_token_accuracy": 0.1985000878572464,
      "num_tokens": 58058593.0,
      "step": 28645
    },
    {
      "entropy": 6.01740665435791,
      "epoch": 2.50371406099799,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043717964026003856,
      "loss": 5.2577,
      "mean_token_accuracy": 0.21215178966522216,
      "num_tokens": 58068996.0,
      "step": 28650
    },
    {
      "entropy": 6.001143932342529,
      "epoch": 2.5041510093506947,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004371580395528192,
      "loss": 5.2645,
      "mean_token_accuracy": 0.20043449252843856,
      "num_tokens": 58078546.0,
      "step": 28655
    },
    {
      "entropy": 6.034889507293701,
      "epoch": 2.5045879577033996,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043713643573532377,
      "loss": 5.3595,
      "mean_token_accuracy": 0.18924068063497543,
      "num_tokens": 58087824.0,
      "step": 28660
    },
    {
      "entropy": 5.941547155380249,
      "epoch": 2.505024906056104,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043711482880796643,
      "loss": 5.2988,
      "mean_token_accuracy": 0.2063297986984253,
      "num_tokens": 58097940.0,
      "step": 28665
    },
    {
      "entropy": 6.0182465553283695,
      "epoch": 2.505461854408809,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004370932187711618,
      "loss": 5.2983,
      "mean_token_accuracy": 0.19423691630363465,
      "num_tokens": 58109274.0,
      "step": 28670
    },
    {
      "entropy": 6.017317390441894,
      "epoch": 2.5058988027615134,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004370716056253242,
      "loss": 5.2776,
      "mean_token_accuracy": 0.19998184889554976,
      "num_tokens": 58119061.0,
      "step": 28675
    },
    {
      "entropy": 5.992775583267212,
      "epoch": 2.5063357511142184,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043704998937086844,
      "loss": 5.2691,
      "mean_token_accuracy": 0.20180677026510238,
      "num_tokens": 58128949.0,
      "step": 28680
    },
    {
      "entropy": 5.925647163391114,
      "epoch": 2.5067726994669233,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004370283700082088,
      "loss": 5.1725,
      "mean_token_accuracy": 0.21054780036211013,
      "num_tokens": 58139807.0,
      "step": 28685
    },
    {
      "entropy": 6.01763391494751,
      "epoch": 2.5072096478196277,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004370067475377602,
      "loss": 5.3032,
      "mean_token_accuracy": 0.20091084986925126,
      "num_tokens": 58150878.0,
      "step": 28690
    },
    {
      "entropy": 5.944634532928466,
      "epoch": 2.507646596172332,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004369851219599373,
      "loss": 5.2804,
      "mean_token_accuracy": 0.20275641977787018,
      "num_tokens": 58160805.0,
      "step": 28695
    },
    {
      "entropy": 6.021055126190186,
      "epoch": 2.508083544525037,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043696349327515484,
      "loss": 5.229,
      "mean_token_accuracy": 0.20099234879016875,
      "num_tokens": 58170298.0,
      "step": 28700
    },
    {
      "entropy": 5.9543030738830565,
      "epoch": 2.508520492877742,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043694186148382773,
      "loss": 5.2423,
      "mean_token_accuracy": 0.21296296417713165,
      "num_tokens": 58180497.0,
      "step": 28705
    },
    {
      "entropy": 6.0059648036956785,
      "epoch": 2.5089574412304465,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043692022658637085,
      "loss": 5.3115,
      "mean_token_accuracy": 0.19620503932237626,
      "num_tokens": 58190871.0,
      "step": 28710
    },
    {
      "entropy": 5.965127944946289,
      "epoch": 2.5093943895831514,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043689858858319923,
      "loss": 5.2674,
      "mean_token_accuracy": 0.20455760210752488,
      "num_tokens": 58201654.0,
      "step": 28715
    },
    {
      "entropy": 5.975850296020508,
      "epoch": 2.509831337935856,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043687694747472785,
      "loss": 5.3262,
      "mean_token_accuracy": 0.20539195835590363,
      "num_tokens": 58211647.0,
      "step": 28720
    },
    {
      "entropy": 5.946395826339722,
      "epoch": 2.510268286288561,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043685530326137173,
      "loss": 5.2472,
      "mean_token_accuracy": 0.2070670947432518,
      "num_tokens": 58221945.0,
      "step": 28725
    },
    {
      "entropy": 5.986625337600708,
      "epoch": 2.5107052346412653,
      "grad_norm": 1.25,
      "learning_rate": 0.00043683365594354614,
      "loss": 5.2972,
      "mean_token_accuracy": 0.20650529116392136,
      "num_tokens": 58231702.0,
      "step": 28730
    },
    {
      "entropy": 6.024110651016235,
      "epoch": 2.51114218299397,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004368120055216662,
      "loss": 5.3433,
      "mean_token_accuracy": 0.19526453018188478,
      "num_tokens": 58242822.0,
      "step": 28735
    },
    {
      "entropy": 5.9817485332489015,
      "epoch": 2.511579131346675,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004367903519961472,
      "loss": 5.2165,
      "mean_token_accuracy": 0.21209161430597306,
      "num_tokens": 58253221.0,
      "step": 28740
    },
    {
      "entropy": 5.964937973022461,
      "epoch": 2.5120160796993796,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004367686953674045,
      "loss": 5.2225,
      "mean_token_accuracy": 0.21015130281448363,
      "num_tokens": 58263739.0,
      "step": 28745
    },
    {
      "entropy": 5.940247821807861,
      "epoch": 2.512453028052084,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043674703563585345,
      "loss": 5.2446,
      "mean_token_accuracy": 0.20864369124174117,
      "num_tokens": 58273526.0,
      "step": 28750
    },
    {
      "entropy": 5.910269069671631,
      "epoch": 2.512889976404789,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004367253728019095,
      "loss": 5.2565,
      "mean_token_accuracy": 0.2102193608880043,
      "num_tokens": 58283373.0,
      "step": 28755
    },
    {
      "entropy": 6.005427074432373,
      "epoch": 2.513326924757494,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043670370686598814,
      "loss": 5.3191,
      "mean_token_accuracy": 0.20326435714960098,
      "num_tokens": 58293365.0,
      "step": 28760
    },
    {
      "entropy": 5.958084678649902,
      "epoch": 2.5137638731101983,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043668203782850495,
      "loss": 5.2515,
      "mean_token_accuracy": 0.20628005415201187,
      "num_tokens": 58303750.0,
      "step": 28765
    },
    {
      "entropy": 5.985069465637207,
      "epoch": 2.5142008214629032,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004366603656898755,
      "loss": 5.2915,
      "mean_token_accuracy": 0.2036924257874489,
      "num_tokens": 58313209.0,
      "step": 28770
    },
    {
      "entropy": 5.994743919372558,
      "epoch": 2.5146377698156077,
      "grad_norm": 1.25,
      "learning_rate": 0.00043663869045051565,
      "loss": 5.3033,
      "mean_token_accuracy": 0.20486003458499907,
      "num_tokens": 58322641.0,
      "step": 28775
    },
    {
      "entropy": 6.019469308853149,
      "epoch": 2.5150747181683126,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043661701211084096,
      "loss": 5.1806,
      "mean_token_accuracy": 0.21578887701034546,
      "num_tokens": 58331865.0,
      "step": 28780
    },
    {
      "entropy": 5.996821498870849,
      "epoch": 2.515511666521017,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043659533067126735,
      "loss": 5.3379,
      "mean_token_accuracy": 0.20205118358135224,
      "num_tokens": 58341188.0,
      "step": 28785
    },
    {
      "entropy": 5.936635589599609,
      "epoch": 2.515948614873722,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004365736461322105,
      "loss": 5.3175,
      "mean_token_accuracy": 0.19872253835201265,
      "num_tokens": 58351856.0,
      "step": 28790
    },
    {
      "entropy": 5.977356576919556,
      "epoch": 2.5163855632264265,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004365519584940866,
      "loss": 5.3087,
      "mean_token_accuracy": 0.20018353164196015,
      "num_tokens": 58363066.0,
      "step": 28795
    },
    {
      "entropy": 6.0489410877227785,
      "epoch": 2.5168225115791314,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004365302677573114,
      "loss": 5.2736,
      "mean_token_accuracy": 0.1990484818816185,
      "num_tokens": 58374646.0,
      "step": 28800
    },
    {
      "entropy": 5.952631044387817,
      "epoch": 2.517259459931836,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000436508573922301,
      "loss": 5.258,
      "mean_token_accuracy": 0.20308532267808915,
      "num_tokens": 58385192.0,
      "step": 28805
    },
    {
      "entropy": 5.9975409507751465,
      "epoch": 2.5176964082845408,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043648687698947153,
      "loss": 5.2665,
      "mean_token_accuracy": 0.21112236976623536,
      "num_tokens": 58394237.0,
      "step": 28810
    },
    {
      "entropy": 5.914026737213135,
      "epoch": 2.5181333566372457,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043646517695923923,
      "loss": 5.2418,
      "mean_token_accuracy": 0.2123783439397812,
      "num_tokens": 58404229.0,
      "step": 28815
    },
    {
      "entropy": 5.96253924369812,
      "epoch": 2.51857030498995,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004364434738320201,
      "loss": 5.2881,
      "mean_token_accuracy": 0.20284906774759293,
      "num_tokens": 58414706.0,
      "step": 28820
    },
    {
      "entropy": 6.025110673904419,
      "epoch": 2.5190072533426546,
      "grad_norm": 1.125,
      "learning_rate": 0.00043642176760823066,
      "loss": 5.3076,
      "mean_token_accuracy": 0.19742999821901322,
      "num_tokens": 58425699.0,
      "step": 28825
    },
    {
      "entropy": 6.024270820617676,
      "epoch": 2.5194442016953595,
      "grad_norm": 1.203125,
      "learning_rate": 0.000436400058288287,
      "loss": 5.3062,
      "mean_token_accuracy": 0.19424470514059067,
      "num_tokens": 58435923.0,
      "step": 28830
    },
    {
      "entropy": 5.967609643936157,
      "epoch": 2.5198811500480645,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004363783458726058,
      "loss": 5.285,
      "mean_token_accuracy": 0.19926431477069856,
      "num_tokens": 58445569.0,
      "step": 28835
    },
    {
      "entropy": 5.996568059921264,
      "epoch": 2.520318098400769,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043635663036160324,
      "loss": 5.2517,
      "mean_token_accuracy": 0.21130514442920684,
      "num_tokens": 58456686.0,
      "step": 28840
    },
    {
      "entropy": 5.930530595779419,
      "epoch": 2.520755046753474,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000436334911755696,
      "loss": 5.2159,
      "mean_token_accuracy": 0.20813927948474883,
      "num_tokens": 58466148.0,
      "step": 28845
    },
    {
      "entropy": 6.000898313522339,
      "epoch": 2.5211919951061783,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004363131900553006,
      "loss": 5.2686,
      "mean_token_accuracy": 0.19988810569047927,
      "num_tokens": 58475161.0,
      "step": 28850
    },
    {
      "entropy": 6.009205055236817,
      "epoch": 2.521628943458883,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004362914652608338,
      "loss": 5.2457,
      "mean_token_accuracy": 0.20390670001506805,
      "num_tokens": 58483631.0,
      "step": 28855
    },
    {
      "entropy": 5.939361143112182,
      "epoch": 2.5220658918115877,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004362697373727121,
      "loss": 5.2445,
      "mean_token_accuracy": 0.20476087778806687,
      "num_tokens": 58493063.0,
      "step": 28860
    },
    {
      "entropy": 5.984193992614746,
      "epoch": 2.5225028401642926,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004362480063913524,
      "loss": 5.347,
      "mean_token_accuracy": 0.196442611515522,
      "num_tokens": 58503658.0,
      "step": 28865
    },
    {
      "entropy": 5.970314264297485,
      "epoch": 2.5229397885169975,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043622627231717126,
      "loss": 5.2381,
      "mean_token_accuracy": 0.20619446486234666,
      "num_tokens": 58515083.0,
      "step": 28870
    },
    {
      "entropy": 6.066597747802734,
      "epoch": 2.523376736869702,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004362045351505859,
      "loss": 5.3016,
      "mean_token_accuracy": 0.20571998357772828,
      "num_tokens": 58524984.0,
      "step": 28875
    },
    {
      "entropy": 6.1106654644012455,
      "epoch": 2.5238136852224065,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004361827948920131,
      "loss": 5.3692,
      "mean_token_accuracy": 0.19427465051412582,
      "num_tokens": 58535651.0,
      "step": 28880
    },
    {
      "entropy": 6.00118317604065,
      "epoch": 2.5242506335751114,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043616105154186974,
      "loss": 5.3526,
      "mean_token_accuracy": 0.20063704550266265,
      "num_tokens": 58546380.0,
      "step": 28885
    },
    {
      "entropy": 5.970354270935059,
      "epoch": 2.5246875819278163,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043613930510057305,
      "loss": 5.2366,
      "mean_token_accuracy": 0.20892452001571654,
      "num_tokens": 58555639.0,
      "step": 28890
    },
    {
      "entropy": 6.005766773223877,
      "epoch": 2.5251245302805208,
      "grad_norm": 1.453125,
      "learning_rate": 0.00043611755556853993,
      "loss": 5.2816,
      "mean_token_accuracy": 0.20010433495044708,
      "num_tokens": 58564948.0,
      "step": 28895
    },
    {
      "entropy": 6.033193206787109,
      "epoch": 2.5255614786332257,
      "grad_norm": 1.125,
      "learning_rate": 0.0004360958029461878,
      "loss": 5.3535,
      "mean_token_accuracy": 0.1958629682660103,
      "num_tokens": 58574146.0,
      "step": 28900
    },
    {
      "entropy": 6.039344072341919,
      "epoch": 2.52599842698593,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004360740472339336,
      "loss": 5.2495,
      "mean_token_accuracy": 0.21151082068681717,
      "num_tokens": 58583438.0,
      "step": 28905
    },
    {
      "entropy": 5.950890016555786,
      "epoch": 2.526435375338635,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043605228843219494,
      "loss": 5.2342,
      "mean_token_accuracy": 0.2068863809108734,
      "num_tokens": 58594552.0,
      "step": 28910
    },
    {
      "entropy": 5.999192333221435,
      "epoch": 2.5268723236913395,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004360305265413889,
      "loss": 5.246,
      "mean_token_accuracy": 0.19886558800935744,
      "num_tokens": 58604162.0,
      "step": 28915
    },
    {
      "entropy": 5.983382987976074,
      "epoch": 2.5273092720440444,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004360087615619331,
      "loss": 5.203,
      "mean_token_accuracy": 0.20505042672157286,
      "num_tokens": 58613487.0,
      "step": 28920
    },
    {
      "entropy": 5.8968418598175045,
      "epoch": 2.5277462203967493,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043598699349424474,
      "loss": 5.1501,
      "mean_token_accuracy": 0.2217172607779503,
      "num_tokens": 58623252.0,
      "step": 28925
    },
    {
      "entropy": 6.024511098861694,
      "epoch": 2.528183168749454,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004359652223387415,
      "loss": 5.3306,
      "mean_token_accuracy": 0.19383296519517898,
      "num_tokens": 58633585.0,
      "step": 28930
    },
    {
      "entropy": 5.982134532928467,
      "epoch": 2.5286201171021583,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004359434480958411,
      "loss": 5.2969,
      "mean_token_accuracy": 0.19676992446184158,
      "num_tokens": 58643202.0,
      "step": 28935
    },
    {
      "entropy": 5.89625186920166,
      "epoch": 2.529057065454863,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004359216707659608,
      "loss": 5.2113,
      "mean_token_accuracy": 0.21171680092811584,
      "num_tokens": 58653153.0,
      "step": 28940
    },
    {
      "entropy": 5.973930215835571,
      "epoch": 2.529494013807568,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004358998903495188,
      "loss": 5.1411,
      "mean_token_accuracy": 0.21809392273426056,
      "num_tokens": 58663835.0,
      "step": 28945
    },
    {
      "entropy": 6.0264771461486815,
      "epoch": 2.5299309621602726,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004358781068469324,
      "loss": 5.2485,
      "mean_token_accuracy": 0.20422210395336152,
      "num_tokens": 58673114.0,
      "step": 28950
    },
    {
      "entropy": 5.994345760345459,
      "epoch": 2.5303679105129775,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043585632025861963,
      "loss": 5.2529,
      "mean_token_accuracy": 0.2051296576857567,
      "num_tokens": 58683121.0,
      "step": 28955
    },
    {
      "entropy": 6.0246742248535154,
      "epoch": 2.530804858865682,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004358345305849984,
      "loss": 5.2958,
      "mean_token_accuracy": 0.19811339378356935,
      "num_tokens": 58693656.0,
      "step": 28960
    },
    {
      "entropy": 5.9394073486328125,
      "epoch": 2.531241807218387,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043581273782648664,
      "loss": 5.2579,
      "mean_token_accuracy": 0.2096821293234825,
      "num_tokens": 58703061.0,
      "step": 28965
    },
    {
      "entropy": 5.988442087173462,
      "epoch": 2.5316787555710913,
      "grad_norm": 1.25,
      "learning_rate": 0.00043579094198350224,
      "loss": 5.2948,
      "mean_token_accuracy": 0.20351819396018983,
      "num_tokens": 58712463.0,
      "step": 28970
    },
    {
      "entropy": 5.99011754989624,
      "epoch": 2.5321157039237963,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00043576914305646334,
      "loss": 5.2314,
      "mean_token_accuracy": 0.20366568863391876,
      "num_tokens": 58722214.0,
      "step": 28975
    },
    {
      "entropy": 6.005238056182861,
      "epoch": 2.532552652276501,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000435747341045788,
      "loss": 5.3145,
      "mean_token_accuracy": 0.1972338542342186,
      "num_tokens": 58732042.0,
      "step": 28980
    },
    {
      "entropy": 6.001090860366821,
      "epoch": 2.5329896006292056,
      "grad_norm": 1.125,
      "learning_rate": 0.00043572553595189444,
      "loss": 5.3772,
      "mean_token_accuracy": 0.19225098043680192,
      "num_tokens": 58743139.0,
      "step": 28985
    },
    {
      "entropy": 6.053582286834716,
      "epoch": 2.53342654898191,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043570372777520096,
      "loss": 5.3249,
      "mean_token_accuracy": 0.19485745280981065,
      "num_tokens": 58752994.0,
      "step": 28990
    },
    {
      "entropy": 5.968592309951783,
      "epoch": 2.533863497334615,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043568191651612573,
      "loss": 5.3068,
      "mean_token_accuracy": 0.19468299746513368,
      "num_tokens": 58762664.0,
      "step": 28995
    },
    {
      "entropy": 5.983875179290772,
      "epoch": 2.53430044568732,
      "grad_norm": 1.125,
      "learning_rate": 0.00043566010217508706,
      "loss": 5.3027,
      "mean_token_accuracy": 0.2028774991631508,
      "num_tokens": 58772992.0,
      "step": 29000
    },
    {
      "entropy": 6.004227256774902,
      "epoch": 2.5347373940400244,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004356382847525035,
      "loss": 5.2884,
      "mean_token_accuracy": 0.20322806984186173,
      "num_tokens": 58783157.0,
      "step": 29005
    },
    {
      "entropy": 6.040749645233154,
      "epoch": 2.5351743423927293,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004356164642487935,
      "loss": 5.3426,
      "mean_token_accuracy": 0.20086099803447724,
      "num_tokens": 58794020.0,
      "step": 29010
    },
    {
      "entropy": 5.9366849899292,
      "epoch": 2.535611290745434,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004355946406643755,
      "loss": 5.2313,
      "mean_token_accuracy": 0.208686164021492,
      "num_tokens": 58804524.0,
      "step": 29015
    },
    {
      "entropy": 5.942982530593872,
      "epoch": 2.5360482390981387,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043557281399966814,
      "loss": 5.2874,
      "mean_token_accuracy": 0.2005661889910698,
      "num_tokens": 58814278.0,
      "step": 29020
    },
    {
      "entropy": 5.977720403671265,
      "epoch": 2.536485187450843,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004355509842550901,
      "loss": 5.2073,
      "mean_token_accuracy": 0.20942481756210327,
      "num_tokens": 58824258.0,
      "step": 29025
    },
    {
      "entropy": 5.978753471374512,
      "epoch": 2.536922135803548,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004355291514310601,
      "loss": 5.3046,
      "mean_token_accuracy": 0.20384543985128403,
      "num_tokens": 58834316.0,
      "step": 29030
    },
    {
      "entropy": 6.0105420589447025,
      "epoch": 2.537359084156253,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043550731552799676,
      "loss": 5.2995,
      "mean_token_accuracy": 0.20375483930110933,
      "num_tokens": 58844201.0,
      "step": 29035
    },
    {
      "entropy": 5.986966514587403,
      "epoch": 2.5377960325089575,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043548547654631917,
      "loss": 5.2293,
      "mean_token_accuracy": 0.2089112877845764,
      "num_tokens": 58853675.0,
      "step": 29040
    },
    {
      "entropy": 5.973989105224609,
      "epoch": 2.538232980861662,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004354636344864459,
      "loss": 5.2718,
      "mean_token_accuracy": 0.20168411433696748,
      "num_tokens": 58863316.0,
      "step": 29045
    },
    {
      "entropy": 5.999681520462036,
      "epoch": 2.538669929214367,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004354417893487961,
      "loss": 5.2282,
      "mean_token_accuracy": 0.20124090015888213,
      "num_tokens": 58873479.0,
      "step": 29050
    },
    {
      "entropy": 5.969886207580567,
      "epoch": 2.5391068775670718,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004354199411337886,
      "loss": 5.2644,
      "mean_token_accuracy": 0.20273020416498183,
      "num_tokens": 58882540.0,
      "step": 29055
    },
    {
      "entropy": 5.978857135772705,
      "epoch": 2.5395438259197762,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004353980898418427,
      "loss": 5.2625,
      "mean_token_accuracy": 0.19896504133939744,
      "num_tokens": 58891565.0,
      "step": 29060
    },
    {
      "entropy": 5.986374616622925,
      "epoch": 2.539980774272481,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004353762354733773,
      "loss": 5.2512,
      "mean_token_accuracy": 0.20501862168312074,
      "num_tokens": 58900689.0,
      "step": 29065
    },
    {
      "entropy": 5.959392118453979,
      "epoch": 2.5404177226251856,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043535437802881175,
      "loss": 5.2447,
      "mean_token_accuracy": 0.20915958881378174,
      "num_tokens": 58910588.0,
      "step": 29070
    },
    {
      "entropy": 5.987724685668946,
      "epoch": 2.5408546709778905,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004353325175085651,
      "loss": 5.3472,
      "mean_token_accuracy": 0.19736105501651763,
      "num_tokens": 58921189.0,
      "step": 29075
    },
    {
      "entropy": 5.996506977081299,
      "epoch": 2.541291619330595,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043531065391305685,
      "loss": 5.2621,
      "mean_token_accuracy": 0.19815252274274825,
      "num_tokens": 58930941.0,
      "step": 29080
    },
    {
      "entropy": 6.011729145050049,
      "epoch": 2.5417285676833,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004352887872427063,
      "loss": 5.3755,
      "mean_token_accuracy": 0.20364423245191574,
      "num_tokens": 58940371.0,
      "step": 29085
    },
    {
      "entropy": 5.966240549087525,
      "epoch": 2.5421655160360044,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043526691749793273,
      "loss": 5.274,
      "mean_token_accuracy": 0.19934715926647187,
      "num_tokens": 58950446.0,
      "step": 29090
    },
    {
      "entropy": 5.987056541442871,
      "epoch": 2.5426024643887093,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004352450446791557,
      "loss": 5.2513,
      "mean_token_accuracy": 0.20416356325149537,
      "num_tokens": 58960757.0,
      "step": 29095
    },
    {
      "entropy": 6.030523347854614,
      "epoch": 2.5430394127414138,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043522316878679475,
      "loss": 5.3128,
      "mean_token_accuracy": 0.20383866131305695,
      "num_tokens": 58971621.0,
      "step": 29100
    },
    {
      "entropy": 6.033510780334472,
      "epoch": 2.5434763610941187,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043520128982126944,
      "loss": 5.2871,
      "mean_token_accuracy": 0.20518697500228883,
      "num_tokens": 58982505.0,
      "step": 29105
    },
    {
      "entropy": 5.956657457351684,
      "epoch": 2.5439133094468236,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004351794077829995,
      "loss": 5.2024,
      "mean_token_accuracy": 0.2135899543762207,
      "num_tokens": 58992270.0,
      "step": 29110
    },
    {
      "entropy": 5.89670991897583,
      "epoch": 2.544350257799528,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004351575226724046,
      "loss": 5.1596,
      "mean_token_accuracy": 0.20991332828998566,
      "num_tokens": 59002510.0,
      "step": 29115
    },
    {
      "entropy": 5.999687004089355,
      "epoch": 2.544787206152233,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004351356344899045,
      "loss": 5.2666,
      "mean_token_accuracy": 0.20020217895507814,
      "num_tokens": 59012942.0,
      "step": 29120
    },
    {
      "entropy": 6.030916786193847,
      "epoch": 2.5452241545049374,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043511374323591904,
      "loss": 5.3744,
      "mean_token_accuracy": 0.19929156750440596,
      "num_tokens": 59023509.0,
      "step": 29125
    },
    {
      "entropy": 6.053931665420532,
      "epoch": 2.5456611028576424,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043509184891086795,
      "loss": 5.2489,
      "mean_token_accuracy": 0.2050583317875862,
      "num_tokens": 59033782.0,
      "step": 29130
    },
    {
      "entropy": 5.975520181655884,
      "epoch": 2.546098051210347,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043506995151517135,
      "loss": 5.3229,
      "mean_token_accuracy": 0.19427622705698014,
      "num_tokens": 59045791.0,
      "step": 29135
    },
    {
      "entropy": 5.981488370895386,
      "epoch": 2.5465349995630517,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043504805104924925,
      "loss": 5.234,
      "mean_token_accuracy": 0.20786426067352295,
      "num_tokens": 59056578.0,
      "step": 29140
    },
    {
      "entropy": 6.023945283889771,
      "epoch": 2.546971947915756,
      "grad_norm": 1.25,
      "learning_rate": 0.0004350261475135217,
      "loss": 5.3064,
      "mean_token_accuracy": 0.1964884355664253,
      "num_tokens": 59066712.0,
      "step": 29145
    },
    {
      "entropy": 5.991926574707032,
      "epoch": 2.547408896268461,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004350042409084088,
      "loss": 5.224,
      "mean_token_accuracy": 0.21289344429969786,
      "num_tokens": 59077557.0,
      "step": 29150
    },
    {
      "entropy": 5.971440744400025,
      "epoch": 2.5478458446211656,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043498233123433064,
      "loss": 5.2567,
      "mean_token_accuracy": 0.20512089282274246,
      "num_tokens": 59086649.0,
      "step": 29155
    },
    {
      "entropy": 5.9947654724121096,
      "epoch": 2.5482827929738705,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043496041849170757,
      "loss": 5.2457,
      "mean_token_accuracy": 0.2076978340744972,
      "num_tokens": 59096838.0,
      "step": 29160
    },
    {
      "entropy": 5.994300127029419,
      "epoch": 2.5487197413265754,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004349385026809599,
      "loss": 5.2702,
      "mean_token_accuracy": 0.20787379145622253,
      "num_tokens": 59107884.0,
      "step": 29165
    },
    {
      "entropy": 5.992727899551392,
      "epoch": 2.54915668967928,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043491658380250784,
      "loss": 5.3037,
      "mean_token_accuracy": 0.19744227677583695,
      "num_tokens": 59118315.0,
      "step": 29170
    },
    {
      "entropy": 5.933763742446899,
      "epoch": 2.5495936380319844,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043489466185677193,
      "loss": 5.2613,
      "mean_token_accuracy": 0.2047365665435791,
      "num_tokens": 59128331.0,
      "step": 29175
    },
    {
      "entropy": 5.991111135482788,
      "epoch": 2.5500305863846893,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004348727368441727,
      "loss": 5.2397,
      "mean_token_accuracy": 0.2108575239777565,
      "num_tokens": 59138643.0,
      "step": 29180
    },
    {
      "entropy": 6.009386253356934,
      "epoch": 2.550467534737394,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004348508087651306,
      "loss": 5.2274,
      "mean_token_accuracy": 0.2048812434077263,
      "num_tokens": 59148364.0,
      "step": 29185
    },
    {
      "entropy": 5.95018630027771,
      "epoch": 2.5509044830900987,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043482887762006614,
      "loss": 5.2776,
      "mean_token_accuracy": 0.20437239557504655,
      "num_tokens": 59158317.0,
      "step": 29190
    },
    {
      "entropy": 6.0410229682922365,
      "epoch": 2.5513414314428036,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043480694340940017,
      "loss": 5.3438,
      "mean_token_accuracy": 0.19779048562049867,
      "num_tokens": 59168879.0,
      "step": 29195
    },
    {
      "entropy": 6.053788375854492,
      "epoch": 2.551778379795508,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004347850061335532,
      "loss": 5.2362,
      "mean_token_accuracy": 0.2120053380727768,
      "num_tokens": 59179248.0,
      "step": 29200
    },
    {
      "entropy": 5.908783674240112,
      "epoch": 2.552215328148213,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004347630657929461,
      "loss": 5.1748,
      "mean_token_accuracy": 0.21820689141750335,
      "num_tokens": 59189217.0,
      "step": 29205
    },
    {
      "entropy": 6.007916736602783,
      "epoch": 2.5526522765009174,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043474112238799974,
      "loss": 5.2732,
      "mean_token_accuracy": 0.20722768306732178,
      "num_tokens": 59199857.0,
      "step": 29210
    },
    {
      "entropy": 6.026928043365478,
      "epoch": 2.5530892248536223,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043471917591913486,
      "loss": 5.28,
      "mean_token_accuracy": 0.20630669891834258,
      "num_tokens": 59210119.0,
      "step": 29215
    },
    {
      "entropy": 5.939579916000366,
      "epoch": 2.5535261732063272,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043469722638677256,
      "loss": 5.2133,
      "mean_token_accuracy": 0.20543931424617767,
      "num_tokens": 59219258.0,
      "step": 29220
    },
    {
      "entropy": 5.927588701248169,
      "epoch": 2.5539631215590317,
      "grad_norm": 1.640625,
      "learning_rate": 0.00043467527379133377,
      "loss": 5.3067,
      "mean_token_accuracy": 0.20439697802066803,
      "num_tokens": 59229664.0,
      "step": 29225
    },
    {
      "entropy": 6.0564673900604244,
      "epoch": 2.554400069911736,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004346533181332395,
      "loss": 5.2943,
      "mean_token_accuracy": 0.202277335524559,
      "num_tokens": 59239699.0,
      "step": 29230
    },
    {
      "entropy": 6.007443857192993,
      "epoch": 2.554837018264441,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043463135941291105,
      "loss": 5.2535,
      "mean_token_accuracy": 0.2018280431628227,
      "num_tokens": 59249891.0,
      "step": 29235
    },
    {
      "entropy": 5.975586605072022,
      "epoch": 2.555273966617146,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004346093976307694,
      "loss": 5.2927,
      "mean_token_accuracy": 0.19993350952863692,
      "num_tokens": 59260092.0,
      "step": 29240
    },
    {
      "entropy": 5.949393701553345,
      "epoch": 2.5557109149698505,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004345874327872359,
      "loss": 5.3362,
      "mean_token_accuracy": 0.19778950810432433,
      "num_tokens": 59268844.0,
      "step": 29245
    },
    {
      "entropy": 6.025098371505737,
      "epoch": 2.5561478633225554,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004345654648827318,
      "loss": 5.3624,
      "mean_token_accuracy": 0.19194090068340303,
      "num_tokens": 59279275.0,
      "step": 29250
    },
    {
      "entropy": 6.020703363418579,
      "epoch": 2.55658481167526,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004345434939176784,
      "loss": 5.2654,
      "mean_token_accuracy": 0.2080706387758255,
      "num_tokens": 59290596.0,
      "step": 29255
    },
    {
      "entropy": 5.961863279342651,
      "epoch": 2.5570217600279648,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043452151989249727,
      "loss": 5.2027,
      "mean_token_accuracy": 0.20989882797002793,
      "num_tokens": 59300642.0,
      "step": 29260
    },
    {
      "entropy": 5.937086296081543,
      "epoch": 2.5574587083806692,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004344995428076098,
      "loss": 5.3036,
      "mean_token_accuracy": 0.20578316897153853,
      "num_tokens": 59310064.0,
      "step": 29265
    },
    {
      "entropy": 6.077273559570313,
      "epoch": 2.557895656733374,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043447756266343755,
      "loss": 5.2871,
      "mean_token_accuracy": 0.2033427745103836,
      "num_tokens": 59319114.0,
      "step": 29270
    },
    {
      "entropy": 6.038460350036621,
      "epoch": 2.558332605086079,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043445557946040207,
      "loss": 5.343,
      "mean_token_accuracy": 0.20109425187110902,
      "num_tokens": 59329414.0,
      "step": 29275
    },
    {
      "entropy": 6.0287025451660154,
      "epoch": 2.5587695534387835,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043443359319892495,
      "loss": 5.3351,
      "mean_token_accuracy": 0.1984989583492279,
      "num_tokens": 59339980.0,
      "step": 29280
    },
    {
      "entropy": 6.041153287887573,
      "epoch": 2.559206501791488,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000434411603879428,
      "loss": 5.3195,
      "mean_token_accuracy": 0.19172243326902388,
      "num_tokens": 59349754.0,
      "step": 29285
    },
    {
      "entropy": 5.956500673294068,
      "epoch": 2.559643450144193,
      "grad_norm": 1.28125,
      "learning_rate": 0.000434389611502333,
      "loss": 5.1341,
      "mean_token_accuracy": 0.22307960838079452,
      "num_tokens": 59358542.0,
      "step": 29290
    },
    {
      "entropy": 6.032419204711914,
      "epoch": 2.560080398496898,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004343676160680617,
      "loss": 5.2606,
      "mean_token_accuracy": 0.2026614248752594,
      "num_tokens": 59367743.0,
      "step": 29295
    },
    {
      "entropy": 5.975707483291626,
      "epoch": 2.5605173468496023,
      "grad_norm": 1.21875,
      "learning_rate": 0.000434345617577036,
      "loss": 5.29,
      "mean_token_accuracy": 0.19787462204694747,
      "num_tokens": 59377622.0,
      "step": 29300
    },
    {
      "entropy": 5.949283456802368,
      "epoch": 2.560954295202307,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043432361602967787,
      "loss": 5.1942,
      "mean_token_accuracy": 0.20752978175878525,
      "num_tokens": 59388661.0,
      "step": 29305
    },
    {
      "entropy": 6.0053386211395265,
      "epoch": 2.5613912435550117,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043430161142640934,
      "loss": 5.3027,
      "mean_token_accuracy": 0.2004766583442688,
      "num_tokens": 59399935.0,
      "step": 29310
    },
    {
      "entropy": 6.002023363113404,
      "epoch": 2.5618281919077166,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004342796037676524,
      "loss": 5.2635,
      "mean_token_accuracy": 0.21276604533195495,
      "num_tokens": 59409513.0,
      "step": 29315
    },
    {
      "entropy": 5.990970087051392,
      "epoch": 2.562265140260421,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004342575930538292,
      "loss": 5.2252,
      "mean_token_accuracy": 0.21342134177684785,
      "num_tokens": 59420158.0,
      "step": 29320
    },
    {
      "entropy": 5.979756784439087,
      "epoch": 2.562702088613126,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043423557928536197,
      "loss": 5.3146,
      "mean_token_accuracy": 0.20457864552736282,
      "num_tokens": 59429440.0,
      "step": 29325
    },
    {
      "entropy": 5.930715036392212,
      "epoch": 2.563139036965831,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004342135624626728,
      "loss": 5.2029,
      "mean_token_accuracy": 0.2079433098435402,
      "num_tokens": 59439014.0,
      "step": 29330
    },
    {
      "entropy": 5.946143054962159,
      "epoch": 2.5635759853185354,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004341915425861841,
      "loss": 5.217,
      "mean_token_accuracy": 0.21103401780128478,
      "num_tokens": 59448714.0,
      "step": 29335
    },
    {
      "entropy": 5.988485670089721,
      "epoch": 2.56401293367124,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004341695196563183,
      "loss": 5.2751,
      "mean_token_accuracy": 0.2051577553153038,
      "num_tokens": 59459577.0,
      "step": 29340
    },
    {
      "entropy": 6.043767881393433,
      "epoch": 2.5644498820239447,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043414749367349755,
      "loss": 5.2595,
      "mean_token_accuracy": 0.2063654914498329,
      "num_tokens": 59470313.0,
      "step": 29345
    },
    {
      "entropy": 6.038123416900635,
      "epoch": 2.5648868303766497,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004341254646381446,
      "loss": 5.3008,
      "mean_token_accuracy": 0.20723589807748793,
      "num_tokens": 59479484.0,
      "step": 29350
    },
    {
      "entropy": 5.966020679473877,
      "epoch": 2.565323778729354,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043410343255068177,
      "loss": 5.2656,
      "mean_token_accuracy": 0.19819685071706772,
      "num_tokens": 59490290.0,
      "step": 29355
    },
    {
      "entropy": 6.000415658950805,
      "epoch": 2.565760727082059,
      "grad_norm": 1.25,
      "learning_rate": 0.00043408139741153175,
      "loss": 5.2927,
      "mean_token_accuracy": 0.20855616629123688,
      "num_tokens": 59501080.0,
      "step": 29360
    },
    {
      "entropy": 5.9697929382324215,
      "epoch": 2.5661976754347635,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043405935922111725,
      "loss": 5.1839,
      "mean_token_accuracy": 0.2156129941344261,
      "num_tokens": 59511453.0,
      "step": 29365
    },
    {
      "entropy": 5.98969669342041,
      "epoch": 2.5666346237874684,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004340373179798608,
      "loss": 5.3005,
      "mean_token_accuracy": 0.2092667818069458,
      "num_tokens": 59521438.0,
      "step": 29370
    },
    {
      "entropy": 5.9829545497894285,
      "epoch": 2.567071572140173,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004340152736881853,
      "loss": 5.3628,
      "mean_token_accuracy": 0.19493361860513686,
      "num_tokens": 59531763.0,
      "step": 29375
    },
    {
      "entropy": 6.0986328125,
      "epoch": 2.567508520492878,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043399322634651345,
      "loss": 5.3955,
      "mean_token_accuracy": 0.19175173044204713,
      "num_tokens": 59542857.0,
      "step": 29380
    },
    {
      "entropy": 6.068562412261963,
      "epoch": 2.5679454688455827,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004339711759552684,
      "loss": 5.2565,
      "mean_token_accuracy": 0.20424750298261643,
      "num_tokens": 59552353.0,
      "step": 29385
    },
    {
      "entropy": 6.003448915481568,
      "epoch": 2.568382417198287,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004339491225148727,
      "loss": 5.3032,
      "mean_token_accuracy": 0.2052283689379692,
      "num_tokens": 59562377.0,
      "step": 29390
    },
    {
      "entropy": 6.009875631332397,
      "epoch": 2.5688193655509917,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043392706602574967,
      "loss": 5.2737,
      "mean_token_accuracy": 0.20178236365318297,
      "num_tokens": 59572141.0,
      "step": 29395
    },
    {
      "entropy": 6.003791666030883,
      "epoch": 2.5692563139036966,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043390500648832214,
      "loss": 5.2377,
      "mean_token_accuracy": 0.20519647002220154,
      "num_tokens": 59581883.0,
      "step": 29400
    },
    {
      "entropy": 6.023861122131348,
      "epoch": 2.5696932622564015,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004338829439030134,
      "loss": 5.3242,
      "mean_token_accuracy": 0.1985514611005783,
      "num_tokens": 59591798.0,
      "step": 29405
    },
    {
      "entropy": 6.021389484405518,
      "epoch": 2.570130210609106,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004338608782702465,
      "loss": 5.1993,
      "mean_token_accuracy": 0.2080230727791786,
      "num_tokens": 59601525.0,
      "step": 29410
    },
    {
      "entropy": 5.984797430038452,
      "epoch": 2.570567158961811,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004338388095904447,
      "loss": 5.3674,
      "mean_token_accuracy": 0.19673453122377396,
      "num_tokens": 59611335.0,
      "step": 29415
    },
    {
      "entropy": 6.001266431808472,
      "epoch": 2.5710041073145153,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004338167378640314,
      "loss": 5.3247,
      "mean_token_accuracy": 0.20345277190208436,
      "num_tokens": 59620962.0,
      "step": 29420
    },
    {
      "entropy": 5.9940910816192625,
      "epoch": 2.5714410556672203,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004337946630914297,
      "loss": 5.2615,
      "mean_token_accuracy": 0.20512312203645705,
      "num_tokens": 59631568.0,
      "step": 29425
    },
    {
      "entropy": 5.95266489982605,
      "epoch": 2.5718780040199247,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043377258527306316,
      "loss": 5.3035,
      "mean_token_accuracy": 0.20198703706264495,
      "num_tokens": 59642249.0,
      "step": 29430
    },
    {
      "entropy": 6.029226303100586,
      "epoch": 2.5723149523726296,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004337505044093554,
      "loss": 5.3017,
      "mean_token_accuracy": 0.19636043310165405,
      "num_tokens": 59652516.0,
      "step": 29435
    },
    {
      "entropy": 5.954821538925171,
      "epoch": 2.572751900725334,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043372842050072965,
      "loss": 5.1453,
      "mean_token_accuracy": 0.21805758029222488,
      "num_tokens": 59662487.0,
      "step": 29440
    },
    {
      "entropy": 5.945771026611328,
      "epoch": 2.573188849078039,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004337063335476097,
      "loss": 5.3127,
      "mean_token_accuracy": 0.19806759804487228,
      "num_tokens": 59674061.0,
      "step": 29445
    },
    {
      "entropy": 5.982300043106079,
      "epoch": 2.5736257974307435,
      "grad_norm": 1.21875,
      "learning_rate": 0.000433684243550419,
      "loss": 5.3169,
      "mean_token_accuracy": 0.20138702392578126,
      "num_tokens": 59684159.0,
      "step": 29450
    },
    {
      "entropy": 6.042658948898316,
      "epoch": 2.5740627457834484,
      "grad_norm": 1.125,
      "learning_rate": 0.0004336621505095815,
      "loss": 5.3071,
      "mean_token_accuracy": 0.19769781827926636,
      "num_tokens": 59693742.0,
      "step": 29455
    },
    {
      "entropy": 6.058872842788697,
      "epoch": 2.5744996941361533,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043364005442552073,
      "loss": 5.3651,
      "mean_token_accuracy": 0.1931651324033737,
      "num_tokens": 59703613.0,
      "step": 29460
    },
    {
      "entropy": 6.003883934020996,
      "epoch": 2.574936642488858,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004336179552986606,
      "loss": 5.2315,
      "mean_token_accuracy": 0.20877766162157058,
      "num_tokens": 59713409.0,
      "step": 29465
    },
    {
      "entropy": 5.928619337081909,
      "epoch": 2.5753735908415623,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043359585312942496,
      "loss": 5.1863,
      "mean_token_accuracy": 0.21233315914869308,
      "num_tokens": 59723502.0,
      "step": 29470
    },
    {
      "entropy": 6.003376579284668,
      "epoch": 2.575810539194267,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043357374791823784,
      "loss": 5.3363,
      "mean_token_accuracy": 0.1997034505009651,
      "num_tokens": 59734069.0,
      "step": 29475
    },
    {
      "entropy": 5.9886151313781735,
      "epoch": 2.576247487546972,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043355163966552303,
      "loss": 5.2998,
      "mean_token_accuracy": 0.19594279527664185,
      "num_tokens": 59743793.0,
      "step": 29480
    },
    {
      "entropy": 5.9469132900238035,
      "epoch": 2.5766844358996766,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043352952837170467,
      "loss": 5.3359,
      "mean_token_accuracy": 0.19548074454069136,
      "num_tokens": 59754524.0,
      "step": 29485
    },
    {
      "entropy": 6.016112613677978,
      "epoch": 2.5771213842523815,
      "grad_norm": 1.171875,
      "learning_rate": 0.000433507414037207,
      "loss": 5.2927,
      "mean_token_accuracy": 0.20084260702133178,
      "num_tokens": 59765437.0,
      "step": 29490
    },
    {
      "entropy": 6.011226081848145,
      "epoch": 2.577558332605086,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000433485296662454,
      "loss": 5.2999,
      "mean_token_accuracy": 0.19798173904418945,
      "num_tokens": 59776136.0,
      "step": 29495
    },
    {
      "entropy": 5.939509582519531,
      "epoch": 2.577995280957791,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043346317624787007,
      "loss": 5.2465,
      "mean_token_accuracy": 0.2059334859251976,
      "num_tokens": 59786375.0,
      "step": 29500
    },
    {
      "entropy": 5.916600942611694,
      "epoch": 2.5784322293104953,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004334410527938793,
      "loss": 5.1865,
      "mean_token_accuracy": 0.2102493092417717,
      "num_tokens": 59796937.0,
      "step": 29505
    },
    {
      "entropy": 6.02919716835022,
      "epoch": 2.5788691776632002,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043341892630090607,
      "loss": 5.2917,
      "mean_token_accuracy": 0.20567143261432647,
      "num_tokens": 59806625.0,
      "step": 29510
    },
    {
      "entropy": 5.986478853225708,
      "epoch": 2.579306126015905,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004333967967693749,
      "loss": 5.3127,
      "mean_token_accuracy": 0.201907417178154,
      "num_tokens": 59816618.0,
      "step": 29515
    },
    {
      "entropy": 6.011000871658325,
      "epoch": 2.5797430743686096,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004333746641997102,
      "loss": 5.2866,
      "mean_token_accuracy": 0.20190373808145523,
      "num_tokens": 59827666.0,
      "step": 29520
    },
    {
      "entropy": 5.994614315032959,
      "epoch": 2.580180022721314,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004333525285923363,
      "loss": 5.2617,
      "mean_token_accuracy": 0.20901424884796144,
      "num_tokens": 59837704.0,
      "step": 29525
    },
    {
      "entropy": 5.999507331848145,
      "epoch": 2.580616971074019,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004333303899476781,
      "loss": 5.2209,
      "mean_token_accuracy": 0.20735707879066467,
      "num_tokens": 59848186.0,
      "step": 29530
    },
    {
      "entropy": 5.9974462509155275,
      "epoch": 2.581053919426724,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043330824826616,
      "loss": 5.2395,
      "mean_token_accuracy": 0.20033717155456543,
      "num_tokens": 59858453.0,
      "step": 29535
    },
    {
      "entropy": 5.967131280899048,
      "epoch": 2.5814908677794284,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004332861035482067,
      "loss": 5.2786,
      "mean_token_accuracy": 0.19754456877708435,
      "num_tokens": 59867863.0,
      "step": 29540
    },
    {
      "entropy": 5.949313402175903,
      "epoch": 2.5819278161321333,
      "grad_norm": 1.078125,
      "learning_rate": 0.000433263955794243,
      "loss": 5.2499,
      "mean_token_accuracy": 0.20701816976070403,
      "num_tokens": 59879932.0,
      "step": 29545
    },
    {
      "entropy": 6.072998952865601,
      "epoch": 2.5823647644848378,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004332418050046937,
      "loss": 5.2669,
      "mean_token_accuracy": 0.19964490979909896,
      "num_tokens": 59889586.0,
      "step": 29550
    },
    {
      "entropy": 5.9743204593658445,
      "epoch": 2.5828017128375427,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004332196511799837,
      "loss": 5.2114,
      "mean_token_accuracy": 0.20999657213687897,
      "num_tokens": 59898789.0,
      "step": 29555
    },
    {
      "entropy": 5.969680738449097,
      "epoch": 2.583238661190247,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004331974943205379,
      "loss": 5.31,
      "mean_token_accuracy": 0.19887832403182984,
      "num_tokens": 59909346.0,
      "step": 29560
    },
    {
      "entropy": 5.994802236557007,
      "epoch": 2.583675609542952,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043317533442678124,
      "loss": 5.2067,
      "mean_token_accuracy": 0.21044906377792358,
      "num_tokens": 59919979.0,
      "step": 29565
    },
    {
      "entropy": 6.033013439178466,
      "epoch": 2.584112557895657,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043315317149913876,
      "loss": 5.2587,
      "mean_token_accuracy": 0.20417871922254563,
      "num_tokens": 59929855.0,
      "step": 29570
    },
    {
      "entropy": 5.980901050567627,
      "epoch": 2.5845495062483614,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043313100553803564,
      "loss": 5.2728,
      "mean_token_accuracy": 0.2025963231921196,
      "num_tokens": 59939148.0,
      "step": 29575
    },
    {
      "entropy": 5.8769745349884035,
      "epoch": 2.584986454601066,
      "grad_norm": 1.125,
      "learning_rate": 0.0004331088365438969,
      "loss": 5.1872,
      "mean_token_accuracy": 0.2081003174185753,
      "num_tokens": 59948645.0,
      "step": 29580
    },
    {
      "entropy": 5.964238834381104,
      "epoch": 2.585423402953771,
      "grad_norm": 1.125,
      "learning_rate": 0.0004330866645171479,
      "loss": 5.249,
      "mean_token_accuracy": 0.20898983925580977,
      "num_tokens": 59958777.0,
      "step": 29585
    },
    {
      "entropy": 5.980253744125366,
      "epoch": 2.5858603513064757,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004330644894582138,
      "loss": 5.2004,
      "mean_token_accuracy": 0.20983406752347947,
      "num_tokens": 59969176.0,
      "step": 29590
    },
    {
      "entropy": 5.990629053115844,
      "epoch": 2.58629729965918,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043304231136751995,
      "loss": 5.3033,
      "mean_token_accuracy": 0.19338317662477494,
      "num_tokens": 59979263.0,
      "step": 29595
    },
    {
      "entropy": 5.989648103713989,
      "epoch": 2.586734248011885,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043302013024549174,
      "loss": 5.2228,
      "mean_token_accuracy": 0.21167619824409484,
      "num_tokens": 59989313.0,
      "step": 29600
    },
    {
      "entropy": 6.013743352890015,
      "epoch": 2.5871711963645896,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043299794609255474,
      "loss": 5.2766,
      "mean_token_accuracy": 0.2067575827240944,
      "num_tokens": 59999410.0,
      "step": 29605
    },
    {
      "entropy": 5.931461906433105,
      "epoch": 2.5876081447172945,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004329757589091342,
      "loss": 5.1969,
      "mean_token_accuracy": 0.21077795773744584,
      "num_tokens": 60010206.0,
      "step": 29610
    },
    {
      "entropy": 5.865432262420654,
      "epoch": 2.588045093069999,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043295356869565586,
      "loss": 5.1204,
      "mean_token_accuracy": 0.21924712210893632,
      "num_tokens": 60018692.0,
      "step": 29615
    },
    {
      "entropy": 5.986083555221557,
      "epoch": 2.588482041422704,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043293137545254536,
      "loss": 5.3095,
      "mean_token_accuracy": 0.19474272280931473,
      "num_tokens": 60028486.0,
      "step": 29620
    },
    {
      "entropy": 5.989610719680786,
      "epoch": 2.588918989775409,
      "grad_norm": 1.125,
      "learning_rate": 0.0004329091791802283,
      "loss": 5.2955,
      "mean_token_accuracy": 0.20122560262680053,
      "num_tokens": 60039347.0,
      "step": 29625
    },
    {
      "entropy": 5.95909013748169,
      "epoch": 2.5893559381281133,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004328869798791305,
      "loss": 5.1816,
      "mean_token_accuracy": 0.21610940098762513,
      "num_tokens": 60049891.0,
      "step": 29630
    },
    {
      "entropy": 6.016399145126343,
      "epoch": 2.5897928864808177,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043286477754967757,
      "loss": 5.2646,
      "mean_token_accuracy": 0.20059829205274582,
      "num_tokens": 60061264.0,
      "step": 29635
    },
    {
      "entropy": 5.967016172409058,
      "epoch": 2.5902298348335226,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043284257219229557,
      "loss": 5.2856,
      "mean_token_accuracy": 0.201368847489357,
      "num_tokens": 60072453.0,
      "step": 29640
    },
    {
      "entropy": 5.908429050445557,
      "epoch": 2.5906667831862276,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043282036380741035,
      "loss": 5.2843,
      "mean_token_accuracy": 0.19954046308994294,
      "num_tokens": 60083619.0,
      "step": 29645
    },
    {
      "entropy": 6.075131177902222,
      "epoch": 2.591103731538932,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004327981523954478,
      "loss": 5.2154,
      "mean_token_accuracy": 0.20713642984628677,
      "num_tokens": 60093231.0,
      "step": 29650
    },
    {
      "entropy": 5.989723587036133,
      "epoch": 2.591540679891637,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000432775937956834,
      "loss": 5.1969,
      "mean_token_accuracy": 0.21027322709560395,
      "num_tokens": 60103017.0,
      "step": 29655
    },
    {
      "entropy": 5.957818651199341,
      "epoch": 2.5919776282443414,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004327537204919951,
      "loss": 5.2577,
      "mean_token_accuracy": 0.20042052268981933,
      "num_tokens": 60113095.0,
      "step": 29660
    },
    {
      "entropy": 6.0113331317901615,
      "epoch": 2.5924145765970463,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004327315000013571,
      "loss": 5.327,
      "mean_token_accuracy": 0.2007779911160469,
      "num_tokens": 60123506.0,
      "step": 29665
    },
    {
      "entropy": 6.016018629074097,
      "epoch": 2.592851524949751,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004327092764853463,
      "loss": 5.3394,
      "mean_token_accuracy": 0.1930519759654999,
      "num_tokens": 60134349.0,
      "step": 29670
    },
    {
      "entropy": 5.998328733444214,
      "epoch": 2.5932884733024557,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043268704994438897,
      "loss": 5.2538,
      "mean_token_accuracy": 0.1960247501730919,
      "num_tokens": 60143970.0,
      "step": 29675
    },
    {
      "entropy": 5.9474598407745365,
      "epoch": 2.5937254216551606,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004326648203789113,
      "loss": 5.2542,
      "mean_token_accuracy": 0.20331452637910843,
      "num_tokens": 60154904.0,
      "step": 29680
    },
    {
      "entropy": 5.975431632995606,
      "epoch": 2.594162370007865,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004326425877893399,
      "loss": 5.2351,
      "mean_token_accuracy": 0.2068651631474495,
      "num_tokens": 60164747.0,
      "step": 29685
    },
    {
      "entropy": 6.019322395324707,
      "epoch": 2.5945993183605696,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043262035217610104,
      "loss": 5.2552,
      "mean_token_accuracy": 0.2055989310145378,
      "num_tokens": 60175487.0,
      "step": 29690
    },
    {
      "entropy": 6.0010751247406,
      "epoch": 2.5950362667132745,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043259811353962114,
      "loss": 5.2693,
      "mean_token_accuracy": 0.19817678034305572,
      "num_tokens": 60185216.0,
      "step": 29695
    },
    {
      "entropy": 5.987769317626953,
      "epoch": 2.5954732150659794,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004325758718803269,
      "loss": 5.2604,
      "mean_token_accuracy": 0.20205181390047072,
      "num_tokens": 60195495.0,
      "step": 29700
    },
    {
      "entropy": 6.060576295852661,
      "epoch": 2.595910163418684,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043255362719864487,
      "loss": 5.3277,
      "mean_token_accuracy": 0.19519117325544358,
      "num_tokens": 60205556.0,
      "step": 29705
    },
    {
      "entropy": 5.992948055267334,
      "epoch": 2.5963471117713888,
      "grad_norm": 1.25,
      "learning_rate": 0.00043253137949500166,
      "loss": 5.2392,
      "mean_token_accuracy": 0.20532365143299103,
      "num_tokens": 60215904.0,
      "step": 29710
    },
    {
      "entropy": 5.910337352752686,
      "epoch": 2.5967840601240932,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043250912876982415,
      "loss": 5.2142,
      "mean_token_accuracy": 0.21101882606744765,
      "num_tokens": 60224762.0,
      "step": 29715
    },
    {
      "entropy": 5.90726227760315,
      "epoch": 2.597221008476798,
      "grad_norm": 1.171875,
      "learning_rate": 0.000432486875023539,
      "loss": 5.2362,
      "mean_token_accuracy": 0.20779408067464827,
      "num_tokens": 60235284.0,
      "step": 29720
    },
    {
      "entropy": 6.020011520385742,
      "epoch": 2.5976579568295026,
      "grad_norm": 1.109375,
      "learning_rate": 0.000432464618256573,
      "loss": 5.2982,
      "mean_token_accuracy": 0.20411110520362855,
      "num_tokens": 60245939.0,
      "step": 29725
    },
    {
      "entropy": 6.062942886352539,
      "epoch": 2.5980949051822075,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004324423584693531,
      "loss": 5.2825,
      "mean_token_accuracy": 0.20177592486143112,
      "num_tokens": 60256903.0,
      "step": 29730
    },
    {
      "entropy": 6.003367519378662,
      "epoch": 2.5985318535349125,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004324200956623063,
      "loss": 5.2417,
      "mean_token_accuracy": 0.20512323528528215,
      "num_tokens": 60267046.0,
      "step": 29735
    },
    {
      "entropy": 5.966442012786866,
      "epoch": 2.598968801887617,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043239782983585964,
      "loss": 5.3036,
      "mean_token_accuracy": 0.2029082328081131,
      "num_tokens": 60276858.0,
      "step": 29740
    },
    {
      "entropy": 6.003828239440918,
      "epoch": 2.5994057502403214,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043237556099044,
      "loss": 5.2814,
      "mean_token_accuracy": 0.2047649696469307,
      "num_tokens": 60285874.0,
      "step": 29745
    },
    {
      "entropy": 6.012302827835083,
      "epoch": 2.5998426985930263,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004323532891264748,
      "loss": 5.319,
      "mean_token_accuracy": 0.2014700651168823,
      "num_tokens": 60295580.0,
      "step": 29750
    },
    {
      "entropy": 6.004932975769043,
      "epoch": 2.600279646945731,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004323310142443909,
      "loss": 5.2811,
      "mean_token_accuracy": 0.20439355075359344,
      "num_tokens": 60305425.0,
      "step": 29755
    },
    {
      "entropy": 5.991164016723633,
      "epoch": 2.6007165952984357,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043230873634461577,
      "loss": 5.2412,
      "mean_token_accuracy": 0.2035730093717575,
      "num_tokens": 60315578.0,
      "step": 29760
    },
    {
      "entropy": 6.005900192260742,
      "epoch": 2.6011535436511406,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004322864554275767,
      "loss": 5.2634,
      "mean_token_accuracy": 0.20808282792568206,
      "num_tokens": 60325345.0,
      "step": 29765
    },
    {
      "entropy": 6.010300254821777,
      "epoch": 2.601590492003845,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043226417149370093,
      "loss": 5.2332,
      "mean_token_accuracy": 0.20508717000484467,
      "num_tokens": 60334608.0,
      "step": 29770
    },
    {
      "entropy": 5.957543659210205,
      "epoch": 2.60202744035655,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043224188454341593,
      "loss": 5.2267,
      "mean_token_accuracy": 0.20953309535980225,
      "num_tokens": 60346909.0,
      "step": 29775
    },
    {
      "entropy": 5.917585706710815,
      "epoch": 2.6024643887092545,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004322195945771493,
      "loss": 5.2151,
      "mean_token_accuracy": 0.2156725123524666,
      "num_tokens": 60355829.0,
      "step": 29780
    },
    {
      "entropy": 5.951312923431397,
      "epoch": 2.6029013370619594,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004321973015953284,
      "loss": 5.2445,
      "mean_token_accuracy": 0.20271357148885727,
      "num_tokens": 60364954.0,
      "step": 29785
    },
    {
      "entropy": 6.065616607666016,
      "epoch": 2.603338285414664,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043217500559838084,
      "loss": 5.3704,
      "mean_token_accuracy": 0.19582470059394835,
      "num_tokens": 60375575.0,
      "step": 29790
    },
    {
      "entropy": 6.0078693389892575,
      "epoch": 2.6037752337673687,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043215270658673437,
      "loss": 5.2963,
      "mean_token_accuracy": 0.19767942428588867,
      "num_tokens": 60385534.0,
      "step": 29795
    },
    {
      "entropy": 5.977994823455811,
      "epoch": 2.604212182120073,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004321304045608166,
      "loss": 5.283,
      "mean_token_accuracy": 0.20269595682621003,
      "num_tokens": 60395517.0,
      "step": 29800
    },
    {
      "entropy": 6.055715608596802,
      "epoch": 2.604649130472778,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004321080995210554,
      "loss": 5.3412,
      "mean_token_accuracy": 0.20485816448926925,
      "num_tokens": 60404781.0,
      "step": 29805
    },
    {
      "entropy": 5.997811937332154,
      "epoch": 2.605086078825483,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004320857914678785,
      "loss": 5.3112,
      "mean_token_accuracy": 0.20501095801591873,
      "num_tokens": 60415323.0,
      "step": 29810
    },
    {
      "entropy": 6.066546058654785,
      "epoch": 2.6055230271781875,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004320634804017138,
      "loss": 5.4011,
      "mean_token_accuracy": 0.1929475411772728,
      "num_tokens": 60425328.0,
      "step": 29815
    },
    {
      "entropy": 5.96528992652893,
      "epoch": 2.605959975530892,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004320411663229892,
      "loss": 5.2656,
      "mean_token_accuracy": 0.20192069709300994,
      "num_tokens": 60435391.0,
      "step": 29820
    },
    {
      "entropy": 6.0425948143005375,
      "epoch": 2.606396923883597,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004320188492321328,
      "loss": 5.3185,
      "mean_token_accuracy": 0.19814502894878389,
      "num_tokens": 60445155.0,
      "step": 29825
    },
    {
      "entropy": 6.036775732040406,
      "epoch": 2.606833872236302,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004319965291295726,
      "loss": 5.237,
      "mean_token_accuracy": 0.2043925166130066,
      "num_tokens": 60455655.0,
      "step": 29830
    },
    {
      "entropy": 5.974779844284058,
      "epoch": 2.6072708205890063,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043197420601573665,
      "loss": 5.1422,
      "mean_token_accuracy": 0.20836012214422225,
      "num_tokens": 60465080.0,
      "step": 29835
    },
    {
      "entropy": 5.992461442947388,
      "epoch": 2.607707768941711,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043195187989105327,
      "loss": 5.2692,
      "mean_token_accuracy": 0.20698061883449553,
      "num_tokens": 60474970.0,
      "step": 29840
    },
    {
      "entropy": 5.988231658935547,
      "epoch": 2.6081447172944157,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043192955075595045,
      "loss": 5.2158,
      "mean_token_accuracy": 0.20710617899894715,
      "num_tokens": 60486599.0,
      "step": 29845
    },
    {
      "entropy": 6.006523847579956,
      "epoch": 2.6085816656471206,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004319072186108567,
      "loss": 5.3139,
      "mean_token_accuracy": 0.20474245101213456,
      "num_tokens": 60496849.0,
      "step": 29850
    },
    {
      "entropy": 6.024062538146973,
      "epoch": 2.609018613999825,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004318848834562003,
      "loss": 5.3018,
      "mean_token_accuracy": 0.1969187319278717,
      "num_tokens": 60506708.0,
      "step": 29855
    },
    {
      "entropy": 5.860132932662964,
      "epoch": 2.60945556235253,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043186254529240957,
      "loss": 5.1669,
      "mean_token_accuracy": 0.22061666399240493,
      "num_tokens": 60516538.0,
      "step": 29860
    },
    {
      "entropy": 5.984154176712036,
      "epoch": 2.609892510705235,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043184020411991315,
      "loss": 5.2118,
      "mean_token_accuracy": 0.21376788467168809,
      "num_tokens": 60526206.0,
      "step": 29865
    },
    {
      "entropy": 6.011817455291748,
      "epoch": 2.6103294590579393,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004318178599391393,
      "loss": 5.3684,
      "mean_token_accuracy": 0.19204456955194474,
      "num_tokens": 60536878.0,
      "step": 29870
    },
    {
      "entropy": 6.0068131446838375,
      "epoch": 2.610766407410644,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043179551275051675,
      "loss": 5.4341,
      "mean_token_accuracy": 0.19216297566890717,
      "num_tokens": 60547709.0,
      "step": 29875
    },
    {
      "entropy": 6.0360386848449705,
      "epoch": 2.6112033557633487,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004317731625544742,
      "loss": 5.1952,
      "mean_token_accuracy": 0.21161205917596818,
      "num_tokens": 60557257.0,
      "step": 29880
    },
    {
      "entropy": 6.0229404926300045,
      "epoch": 2.6116403041160536,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043175080935144015,
      "loss": 5.2903,
      "mean_token_accuracy": 0.2140835091471672,
      "num_tokens": 60568663.0,
      "step": 29885
    },
    {
      "entropy": 5.9677649974823,
      "epoch": 2.612077252468758,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004317284531418434,
      "loss": 5.2392,
      "mean_token_accuracy": 0.20934347659349442,
      "num_tokens": 60579168.0,
      "step": 29890
    },
    {
      "entropy": 6.034711790084839,
      "epoch": 2.612514200821463,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043170609392611294,
      "loss": 5.32,
      "mean_token_accuracy": 0.19650643616914748,
      "num_tokens": 60589893.0,
      "step": 29895
    },
    {
      "entropy": 5.979324865341186,
      "epoch": 2.6129511491741675,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043168373170467735,
      "loss": 5.2352,
      "mean_token_accuracy": 0.2059606209397316,
      "num_tokens": 60599406.0,
      "step": 29900
    },
    {
      "entropy": 5.933356809616089,
      "epoch": 2.6133880975268724,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043166136647796577,
      "loss": 5.3112,
      "mean_token_accuracy": 0.21183384358882903,
      "num_tokens": 60610435.0,
      "step": 29905
    },
    {
      "entropy": 6.010152769088745,
      "epoch": 2.613825045879577,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000431638998246407,
      "loss": 5.2624,
      "mean_token_accuracy": 0.207366381585598,
      "num_tokens": 60621405.0,
      "step": 29910
    },
    {
      "entropy": 5.97645206451416,
      "epoch": 2.614261994232282,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004316166270104303,
      "loss": 5.2493,
      "mean_token_accuracy": 0.2052520900964737,
      "num_tokens": 60632119.0,
      "step": 29915
    },
    {
      "entropy": 6.020912933349609,
      "epoch": 2.6146989425849867,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043159425277046454,
      "loss": 5.2288,
      "mean_token_accuracy": 0.20653294175863265,
      "num_tokens": 60642163.0,
      "step": 29920
    },
    {
      "entropy": 5.96805362701416,
      "epoch": 2.615135890937691,
      "grad_norm": 1.1875,
      "learning_rate": 0.000431571875526939,
      "loss": 5.2392,
      "mean_token_accuracy": 0.20456581860780715,
      "num_tokens": 60652050.0,
      "step": 29925
    },
    {
      "entropy": 5.964357948303222,
      "epoch": 2.6155728392903956,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004315494952802828,
      "loss": 5.2383,
      "mean_token_accuracy": 0.20387909710407257,
      "num_tokens": 60661843.0,
      "step": 29930
    },
    {
      "entropy": 5.945692729949951,
      "epoch": 2.6160097876431005,
      "grad_norm": 1.25,
      "learning_rate": 0.00043152711203092524,
      "loss": 5.1814,
      "mean_token_accuracy": 0.21233538538217545,
      "num_tokens": 60671618.0,
      "step": 29935
    },
    {
      "entropy": 6.0333864212036135,
      "epoch": 2.6164467359958055,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004315047257792957,
      "loss": 5.381,
      "mean_token_accuracy": 0.1981920674443245,
      "num_tokens": 60682930.0,
      "step": 29940
    },
    {
      "entropy": 6.075390625,
      "epoch": 2.61688368434851,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004314823365258235,
      "loss": 5.2898,
      "mean_token_accuracy": 0.20509008169174195,
      "num_tokens": 60692864.0,
      "step": 29945
    },
    {
      "entropy": 6.051225185394287,
      "epoch": 2.617320632701215,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043145994427093806,
      "loss": 5.2871,
      "mean_token_accuracy": 0.2017506867647171,
      "num_tokens": 60702934.0,
      "step": 29950
    },
    {
      "entropy": 6.013912725448608,
      "epoch": 2.6177575810539193,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043143754901506893,
      "loss": 5.2387,
      "mean_token_accuracy": 0.19795203059911728,
      "num_tokens": 60712289.0,
      "step": 29955
    },
    {
      "entropy": 5.90576205253601,
      "epoch": 2.6181945294066242,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043141515075864563,
      "loss": 5.2168,
      "mean_token_accuracy": 0.20961919873952867,
      "num_tokens": 60723046.0,
      "step": 29960
    },
    {
      "entropy": 6.049145936965942,
      "epoch": 2.6186314777593287,
      "grad_norm": 1.34375,
      "learning_rate": 0.00043139274950209776,
      "loss": 5.3007,
      "mean_token_accuracy": 0.19559989720582963,
      "num_tokens": 60731969.0,
      "step": 29965
    },
    {
      "entropy": 6.0701991558074955,
      "epoch": 2.6190684261120336,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043137034524585506,
      "loss": 5.3659,
      "mean_token_accuracy": 0.20242003798484803,
      "num_tokens": 60743021.0,
      "step": 29970
    },
    {
      "entropy": 6.000274991989135,
      "epoch": 2.6195053744647385,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004313479379903472,
      "loss": 5.301,
      "mean_token_accuracy": 0.20510208159685134,
      "num_tokens": 60751854.0,
      "step": 29975
    },
    {
      "entropy": 6.0455179691314695,
      "epoch": 2.619942322817443,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004313255277360039,
      "loss": 5.2629,
      "mean_token_accuracy": 0.20036872178316117,
      "num_tokens": 60761572.0,
      "step": 29980
    },
    {
      "entropy": 5.942549419403076,
      "epoch": 2.6203792711701475,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004313031144832551,
      "loss": 5.2205,
      "mean_token_accuracy": 0.20243251025676728,
      "num_tokens": 60771618.0,
      "step": 29985
    },
    {
      "entropy": 5.997735500335693,
      "epoch": 2.6208162195228524,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043128069823253064,
      "loss": 5.2236,
      "mean_token_accuracy": 0.20853943079710008,
      "num_tokens": 60781038.0,
      "step": 29990
    },
    {
      "entropy": 6.00559368133545,
      "epoch": 2.6212531678755573,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043125827898426056,
      "loss": 5.2808,
      "mean_token_accuracy": 0.20152007937431335,
      "num_tokens": 60792433.0,
      "step": 29995
    },
    {
      "entropy": 5.996362543106079,
      "epoch": 2.6216901162282618,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043123585673887475,
      "loss": 5.2425,
      "mean_token_accuracy": 0.20130956619977952,
      "num_tokens": 60801995.0,
      "step": 30000
    },
    {
      "epoch": 2.6216901162282618,
      "eval_entropy": 5.75268603989185,
      "eval_loss": 5.309795379638672,
      "eval_mean_token_accuracy": 0.21032517760515418,
      "eval_num_tokens": 60801995.0,
      "eval_runtime": 21.5849,
      "eval_samples_per_second": 1714.809,
      "eval_steps_per_second": 214.363,
      "step": 30000
    },
    {
      "entropy": 6.022671175003052,
      "epoch": 2.6221270645809667,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004312134314968034,
      "loss": 5.2865,
      "mean_token_accuracy": 0.19703701436519622,
      "num_tokens": 60811784.0,
      "step": 30005
    },
    {
      "entropy": 5.985787439346313,
      "epoch": 2.622564012933671,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043119100325847654,
      "loss": 5.2236,
      "mean_token_accuracy": 0.213279789686203,
      "num_tokens": 60822018.0,
      "step": 30010
    },
    {
      "entropy": 5.934683179855346,
      "epoch": 2.623000961286376,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004311685720243243,
      "loss": 5.2727,
      "mean_token_accuracy": 0.2049030065536499,
      "num_tokens": 60832204.0,
      "step": 30015
    },
    {
      "entropy": 5.941218280792237,
      "epoch": 2.6234379096390805,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004311461377947772,
      "loss": 5.2438,
      "mean_token_accuracy": 0.2068067893385887,
      "num_tokens": 60843293.0,
      "step": 30020
    },
    {
      "entropy": 6.04713659286499,
      "epoch": 2.6238748579917854,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004311237005702652,
      "loss": 5.257,
      "mean_token_accuracy": 0.1982850417494774,
      "num_tokens": 60852323.0,
      "step": 30025
    },
    {
      "entropy": 5.985924100875854,
      "epoch": 2.6243118063444903,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004311012603512189,
      "loss": 5.2533,
      "mean_token_accuracy": 0.20734197348356248,
      "num_tokens": 60862788.0,
      "step": 30030
    },
    {
      "entropy": 6.007774448394775,
      "epoch": 2.624748754697195,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043107881713806863,
      "loss": 5.2896,
      "mean_token_accuracy": 0.20142518877983093,
      "num_tokens": 60872793.0,
      "step": 30035
    },
    {
      "entropy": 5.976532697677612,
      "epoch": 2.6251857030498993,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043105637093124484,
      "loss": 5.2291,
      "mean_token_accuracy": 0.20707747489213943,
      "num_tokens": 60882712.0,
      "step": 30040
    },
    {
      "entropy": 5.9891143321990965,
      "epoch": 2.625622651402604,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043103392173117814,
      "loss": 5.2928,
      "mean_token_accuracy": 0.20533307194709777,
      "num_tokens": 60892400.0,
      "step": 30045
    },
    {
      "entropy": 5.985976266860962,
      "epoch": 2.626059599755309,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043101146953829895,
      "loss": 5.22,
      "mean_token_accuracy": 0.21106933653354645,
      "num_tokens": 60902480.0,
      "step": 30050
    },
    {
      "entropy": 6.001287841796875,
      "epoch": 2.6264965481080136,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004309890143530382,
      "loss": 5.2991,
      "mean_token_accuracy": 0.20380872040987014,
      "num_tokens": 60912183.0,
      "step": 30055
    },
    {
      "entropy": 6.039124965667725,
      "epoch": 2.6269334964607185,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004309665561758262,
      "loss": 5.2802,
      "mean_token_accuracy": 0.20284735709428786,
      "num_tokens": 60922166.0,
      "step": 30060
    },
    {
      "entropy": 6.004079723358155,
      "epoch": 2.627370444813423,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043094409500709404,
      "loss": 5.2135,
      "mean_token_accuracy": 0.20730700045824052,
      "num_tokens": 60931980.0,
      "step": 30065
    },
    {
      "entropy": 5.962682247161865,
      "epoch": 2.627807393166128,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004309216308472725,
      "loss": 5.4066,
      "mean_token_accuracy": 0.19726070612668992,
      "num_tokens": 60943382.0,
      "step": 30070
    },
    {
      "entropy": 6.007565212249756,
      "epoch": 2.6282443415188323,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004308991636967923,
      "loss": 5.282,
      "mean_token_accuracy": 0.20113292336463928,
      "num_tokens": 60953505.0,
      "step": 30075
    },
    {
      "entropy": 6.056906795501709,
      "epoch": 2.6286812898715373,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004308766935560844,
      "loss": 5.2917,
      "mean_token_accuracy": 0.20222438126802444,
      "num_tokens": 60963851.0,
      "step": 30080
    },
    {
      "entropy": 6.022710466384888,
      "epoch": 2.6291182382242417,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043085422042557985,
      "loss": 5.3084,
      "mean_token_accuracy": 0.202333901822567,
      "num_tokens": 60974919.0,
      "step": 30085
    },
    {
      "entropy": 6.035843420028686,
      "epoch": 2.6295551865769466,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043083174430570974,
      "loss": 5.3203,
      "mean_token_accuracy": 0.19841823130846023,
      "num_tokens": 60983866.0,
      "step": 30090
    },
    {
      "entropy": 5.985694742202758,
      "epoch": 2.629992134929651,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043080926519690517,
      "loss": 5.239,
      "mean_token_accuracy": 0.20702464133501053,
      "num_tokens": 60994393.0,
      "step": 30095
    },
    {
      "entropy": 5.994475555419922,
      "epoch": 2.630429083282356,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043078678309959716,
      "loss": 5.3078,
      "mean_token_accuracy": 0.20332684963941575,
      "num_tokens": 61004634.0,
      "step": 30100
    },
    {
      "entropy": 5.949233293533325,
      "epoch": 2.630866031635061,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004307642980142171,
      "loss": 5.2626,
      "mean_token_accuracy": 0.20523544549942016,
      "num_tokens": 61014771.0,
      "step": 30105
    },
    {
      "entropy": 5.989298963546753,
      "epoch": 2.6313029799877654,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043074180994119604,
      "loss": 5.2884,
      "mean_token_accuracy": 0.20533948093652726,
      "num_tokens": 61025213.0,
      "step": 30110
    },
    {
      "entropy": 6.006053829193116,
      "epoch": 2.63173992834047,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004307193188809656,
      "loss": 5.1789,
      "mean_token_accuracy": 0.21508674919605256,
      "num_tokens": 61035601.0,
      "step": 30115
    },
    {
      "entropy": 5.972050428390503,
      "epoch": 2.632176876693175,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043069682483395694,
      "loss": 5.2476,
      "mean_token_accuracy": 0.2047293722629547,
      "num_tokens": 61045862.0,
      "step": 30120
    },
    {
      "entropy": 5.986014080047608,
      "epoch": 2.6326138250458797,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043067432780060163,
      "loss": 5.2623,
      "mean_token_accuracy": 0.20745094865560532,
      "num_tokens": 61056702.0,
      "step": 30125
    },
    {
      "entropy": 6.003835773468017,
      "epoch": 2.633050773398584,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043065182778133116,
      "loss": 5.3401,
      "mean_token_accuracy": 0.19378331452608108,
      "num_tokens": 61066353.0,
      "step": 30130
    },
    {
      "entropy": 5.990417957305908,
      "epoch": 2.633487721751289,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000430629324776577,
      "loss": 5.2645,
      "mean_token_accuracy": 0.20370913445949554,
      "num_tokens": 61076534.0,
      "step": 30135
    },
    {
      "entropy": 5.97571210861206,
      "epoch": 2.6339246701039936,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043060681878677096,
      "loss": 5.1731,
      "mean_token_accuracy": 0.21491122394800186,
      "num_tokens": 61086635.0,
      "step": 30140
    },
    {
      "entropy": 5.984228086471558,
      "epoch": 2.6343616184566985,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043058430981234453,
      "loss": 5.2976,
      "mean_token_accuracy": 0.19308728575706482,
      "num_tokens": 61096605.0,
      "step": 30145
    },
    {
      "entropy": 5.968532705307007,
      "epoch": 2.634798566809403,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004305617978537295,
      "loss": 5.2354,
      "mean_token_accuracy": 0.21071612238883972,
      "num_tokens": 61106465.0,
      "step": 30150
    },
    {
      "entropy": 5.997684049606323,
      "epoch": 2.635235515162108,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004305392829113577,
      "loss": 5.3283,
      "mean_token_accuracy": 0.20405336022377013,
      "num_tokens": 61116063.0,
      "step": 30155
    },
    {
      "entropy": 6.04552149772644,
      "epoch": 2.6356724635148128,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043051676498566095,
      "loss": 5.314,
      "mean_token_accuracy": 0.1956547200679779,
      "num_tokens": 61125794.0,
      "step": 30160
    },
    {
      "entropy": 5.996600341796875,
      "epoch": 2.6361094118675172,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004304942440770712,
      "loss": 5.2942,
      "mean_token_accuracy": 0.20461156070232392,
      "num_tokens": 61135734.0,
      "step": 30165
    },
    {
      "entropy": 5.95318636894226,
      "epoch": 2.6365463602202217,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004304717201860203,
      "loss": 5.2553,
      "mean_token_accuracy": 0.21013389825820922,
      "num_tokens": 61146071.0,
      "step": 30170
    },
    {
      "entropy": 5.985926723480224,
      "epoch": 2.6369833085729266,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004304491933129404,
      "loss": 5.2291,
      "mean_token_accuracy": 0.20796307921409607,
      "num_tokens": 61154931.0,
      "step": 30175
    },
    {
      "entropy": 5.989266061782837,
      "epoch": 2.6374202569256315,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004304266634582634,
      "loss": 5.2735,
      "mean_token_accuracy": 0.2060681864619255,
      "num_tokens": 61164464.0,
      "step": 30180
    },
    {
      "entropy": 6.015170621871948,
      "epoch": 2.637857205278336,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004304041306224217,
      "loss": 5.2962,
      "mean_token_accuracy": 0.2028801456093788,
      "num_tokens": 61175379.0,
      "step": 30185
    },
    {
      "entropy": 5.95460877418518,
      "epoch": 2.638294153631041,
      "grad_norm": 1.125,
      "learning_rate": 0.00043038159480584723,
      "loss": 5.2099,
      "mean_token_accuracy": 0.2129680797457695,
      "num_tokens": 61186340.0,
      "step": 30190
    },
    {
      "entropy": 5.950805902481079,
      "epoch": 2.6387311019837454,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043035905600897237,
      "loss": 5.2258,
      "mean_token_accuracy": 0.20847396105527877,
      "num_tokens": 61196495.0,
      "step": 30195
    },
    {
      "entropy": 6.018413400650024,
      "epoch": 2.6391680503364503,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043033651423222947,
      "loss": 5.3035,
      "mean_token_accuracy": 0.20046763569116594,
      "num_tokens": 61206555.0,
      "step": 30200
    },
    {
      "entropy": 6.063429832458496,
      "epoch": 2.6396049986891548,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004303139694760508,
      "loss": 5.3617,
      "mean_token_accuracy": 0.19969621300697327,
      "num_tokens": 61215818.0,
      "step": 30205
    },
    {
      "entropy": 5.998223829269409,
      "epoch": 2.6400419470418597,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043029142174086875,
      "loss": 5.2408,
      "mean_token_accuracy": 0.2026909127831459,
      "num_tokens": 61226325.0,
      "step": 30210
    },
    {
      "entropy": 6.029062795639038,
      "epoch": 2.6404788953945646,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004302688710271159,
      "loss": 5.2703,
      "mean_token_accuracy": 0.20249908566474914,
      "num_tokens": 61236179.0,
      "step": 30215
    },
    {
      "entropy": 5.982855176925659,
      "epoch": 2.640915843747269,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043024631733522484,
      "loss": 5.2199,
      "mean_token_accuracy": 0.2076778843998909,
      "num_tokens": 61245680.0,
      "step": 30220
    },
    {
      "entropy": 5.988448858261108,
      "epoch": 2.6413527920999735,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000430223760665628,
      "loss": 5.3349,
      "mean_token_accuracy": 0.19849657416343688,
      "num_tokens": 61256369.0,
      "step": 30225
    },
    {
      "entropy": 5.944233703613281,
      "epoch": 2.6417897404526784,
      "grad_norm": 1.125,
      "learning_rate": 0.00043020120101875807,
      "loss": 5.19,
      "mean_token_accuracy": 0.2105144217610359,
      "num_tokens": 61266395.0,
      "step": 30230
    },
    {
      "entropy": 6.020794486999511,
      "epoch": 2.6422266888053834,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043017863839504784,
      "loss": 5.3217,
      "mean_token_accuracy": 0.20346090495586394,
      "num_tokens": 61276442.0,
      "step": 30235
    },
    {
      "entropy": 6.041897249221802,
      "epoch": 2.642663637158088,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043015607279493,
      "loss": 5.3598,
      "mean_token_accuracy": 0.19522668719291686,
      "num_tokens": 61286230.0,
      "step": 30240
    },
    {
      "entropy": 5.964523029327393,
      "epoch": 2.6431005855107927,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004301335042188373,
      "loss": 5.2237,
      "mean_token_accuracy": 0.20947593003511428,
      "num_tokens": 61297853.0,
      "step": 30245
    },
    {
      "entropy": 5.960389709472656,
      "epoch": 2.643537533863497,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043011093266720273,
      "loss": 5.184,
      "mean_token_accuracy": 0.20916053354740144,
      "num_tokens": 61307976.0,
      "step": 30250
    },
    {
      "entropy": 6.001199197769165,
      "epoch": 2.643974482216202,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043008835814045926,
      "loss": 5.2903,
      "mean_token_accuracy": 0.20050296932458878,
      "num_tokens": 61319101.0,
      "step": 30255
    },
    {
      "entropy": 5.9935211658477785,
      "epoch": 2.6444114305689066,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004300657806390398,
      "loss": 5.275,
      "mean_token_accuracy": 0.203927443921566,
      "num_tokens": 61329492.0,
      "step": 30260
    },
    {
      "entropy": 6.022943544387817,
      "epoch": 2.6448483789216115,
      "grad_norm": 1.125,
      "learning_rate": 0.00043004320016337737,
      "loss": 5.3298,
      "mean_token_accuracy": 0.19901198744773865,
      "num_tokens": 61340480.0,
      "step": 30265
    },
    {
      "entropy": 6.023687696456909,
      "epoch": 2.6452853272743164,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043002061671390513,
      "loss": 5.2403,
      "mean_token_accuracy": 0.19940153360366822,
      "num_tokens": 61350572.0,
      "step": 30270
    },
    {
      "entropy": 6.01480450630188,
      "epoch": 2.645722275627021,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004299980302910563,
      "loss": 5.274,
      "mean_token_accuracy": 0.20189918130636214,
      "num_tokens": 61361329.0,
      "step": 30275
    },
    {
      "entropy": 6.0501524925231935,
      "epoch": 2.6461592239797254,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004299754408952639,
      "loss": 5.3337,
      "mean_token_accuracy": 0.20053446739912034,
      "num_tokens": 61372618.0,
      "step": 30280
    },
    {
      "entropy": 6.088239336013794,
      "epoch": 2.6465961723324303,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004299528485269614,
      "loss": 5.3793,
      "mean_token_accuracy": 0.19118010401725768,
      "num_tokens": 61383831.0,
      "step": 30285
    },
    {
      "entropy": 5.970826768875122,
      "epoch": 2.647033120685135,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004299302531865821,
      "loss": 5.2547,
      "mean_token_accuracy": 0.2017740160226822,
      "num_tokens": 61394527.0,
      "step": 30290
    },
    {
      "entropy": 6.019202709197998,
      "epoch": 2.6474700690378397,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042990765487455935,
      "loss": 5.3586,
      "mean_token_accuracy": 0.1969352275133133,
      "num_tokens": 61404787.0,
      "step": 30295
    },
    {
      "entropy": 5.989014434814453,
      "epoch": 2.6479070173905446,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004298850535913266,
      "loss": 5.1649,
      "mean_token_accuracy": 0.21965281665325165,
      "num_tokens": 61413922.0,
      "step": 30300
    },
    {
      "entropy": 6.004567337036133,
      "epoch": 2.648343965743249,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004298624493373174,
      "loss": 5.2699,
      "mean_token_accuracy": 0.20036544352769853,
      "num_tokens": 61424750.0,
      "step": 30305
    },
    {
      "entropy": 5.90907998085022,
      "epoch": 2.648780914095954,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004298398421129653,
      "loss": 5.2302,
      "mean_token_accuracy": 0.21257929503917694,
      "num_tokens": 61435222.0,
      "step": 30310
    },
    {
      "entropy": 5.994287061691284,
      "epoch": 2.6492178624486584,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004298172319187038,
      "loss": 5.2795,
      "mean_token_accuracy": 0.21483869403600692,
      "num_tokens": 61444230.0,
      "step": 30315
    },
    {
      "entropy": 5.997687816619873,
      "epoch": 2.6496548108013633,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004297946187549667,
      "loss": 5.2922,
      "mean_token_accuracy": 0.19589485824108124,
      "num_tokens": 61454975.0,
      "step": 30320
    },
    {
      "entropy": 6.007258415222168,
      "epoch": 2.6500917591540682,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042977200262218776,
      "loss": 5.363,
      "mean_token_accuracy": 0.1975494757294655,
      "num_tokens": 61465563.0,
      "step": 30325
    },
    {
      "entropy": 6.010755634307861,
      "epoch": 2.6505287075067727,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004297493835208006,
      "loss": 5.2907,
      "mean_token_accuracy": 0.20119556337594985,
      "num_tokens": 61475330.0,
      "step": 30330
    },
    {
      "entropy": 6.046370315551758,
      "epoch": 2.650965655859477,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042972676145123926,
      "loss": 5.3123,
      "mean_token_accuracy": 0.2005407303571701,
      "num_tokens": 61485577.0,
      "step": 30335
    },
    {
      "entropy": 6.024390649795532,
      "epoch": 2.651402604212182,
      "grad_norm": 1.125,
      "learning_rate": 0.00042970413641393754,
      "loss": 5.2556,
      "mean_token_accuracy": 0.20510361045598985,
      "num_tokens": 61495502.0,
      "step": 30340
    },
    {
      "entropy": 5.980573511123657,
      "epoch": 2.651839552564887,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004296815084093294,
      "loss": 5.2903,
      "mean_token_accuracy": 0.19975464940071105,
      "num_tokens": 61505383.0,
      "step": 30345
    },
    {
      "entropy": 5.966918230056763,
      "epoch": 2.6522765009175915,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042965887743784886,
      "loss": 5.2726,
      "mean_token_accuracy": 0.20504204630851747,
      "num_tokens": 61515052.0,
      "step": 30350
    },
    {
      "entropy": 6.025878429412842,
      "epoch": 2.6527134492702964,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004296362434999301,
      "loss": 5.2263,
      "mean_token_accuracy": 0.20270873606204987,
      "num_tokens": 61525139.0,
      "step": 30355
    },
    {
      "entropy": 5.9577501773834225,
      "epoch": 2.653150397623001,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042961360659600706,
      "loss": 5.1699,
      "mean_token_accuracy": 0.21336994618177413,
      "num_tokens": 61534901.0,
      "step": 30360
    },
    {
      "entropy": 5.981212091445923,
      "epoch": 2.653587345975706,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004295909667265141,
      "loss": 5.2756,
      "mean_token_accuracy": 0.1961616650223732,
      "num_tokens": 61543258.0,
      "step": 30365
    },
    {
      "entropy": 5.942918682098389,
      "epoch": 2.6540242943284102,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042956832389188534,
      "loss": 5.2563,
      "mean_token_accuracy": 0.200726617872715,
      "num_tokens": 61553294.0,
      "step": 30370
    },
    {
      "entropy": 6.0337090492248535,
      "epoch": 2.654461242681115,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004295456780925551,
      "loss": 5.2467,
      "mean_token_accuracy": 0.2074637234210968,
      "num_tokens": 61562986.0,
      "step": 30375
    },
    {
      "entropy": 5.951260232925415,
      "epoch": 2.65489819103382,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042952302932895784,
      "loss": 5.3057,
      "mean_token_accuracy": 0.2068344920873642,
      "num_tokens": 61573994.0,
      "step": 30380
    },
    {
      "entropy": 5.929577445983886,
      "epoch": 2.6553351393865245,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042950037760152793,
      "loss": 5.2414,
      "mean_token_accuracy": 0.20889557749032975,
      "num_tokens": 61584279.0,
      "step": 30385
    },
    {
      "entropy": 6.06705493927002,
      "epoch": 2.655772087739229,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004294777229106998,
      "loss": 5.325,
      "mean_token_accuracy": 0.1988336056470871,
      "num_tokens": 61595188.0,
      "step": 30390
    },
    {
      "entropy": 6.001477527618408,
      "epoch": 2.656209036091934,
      "grad_norm": 1.171875,
      "learning_rate": 0.000429455065256908,
      "loss": 5.3464,
      "mean_token_accuracy": 0.19811661094427108,
      "num_tokens": 61605321.0,
      "step": 30395
    },
    {
      "entropy": 6.083032608032227,
      "epoch": 2.656645984444639,
      "grad_norm": 1.125,
      "learning_rate": 0.0004294324046405871,
      "loss": 5.2967,
      "mean_token_accuracy": 0.20071753859519958,
      "num_tokens": 61615137.0,
      "step": 30400
    },
    {
      "entropy": 6.017293119430542,
      "epoch": 2.6570829327973433,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042940974106217176,
      "loss": 5.1832,
      "mean_token_accuracy": 0.2074673891067505,
      "num_tokens": 61625912.0,
      "step": 30405
    },
    {
      "entropy": 5.977476406097412,
      "epoch": 2.6575198811500482,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004293870745220967,
      "loss": 5.2269,
      "mean_token_accuracy": 0.20321110486984253,
      "num_tokens": 61635836.0,
      "step": 30410
    },
    {
      "entropy": 5.9648254871368405,
      "epoch": 2.6579568295027527,
      "grad_norm": 1.25,
      "learning_rate": 0.0004293644050207966,
      "loss": 5.2827,
      "mean_token_accuracy": 0.19955832362174988,
      "num_tokens": 61645151.0,
      "step": 30415
    },
    {
      "entropy": 6.0809258937835695,
      "epoch": 2.6583937778554576,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004293417325587064,
      "loss": 5.3192,
      "mean_token_accuracy": 0.19933933019638062,
      "num_tokens": 61654266.0,
      "step": 30420
    },
    {
      "entropy": 6.081716918945313,
      "epoch": 2.658830726208162,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004293190571362608,
      "loss": 5.4124,
      "mean_token_accuracy": 0.19283279031515121,
      "num_tokens": 61664355.0,
      "step": 30425
    },
    {
      "entropy": 6.048876142501831,
      "epoch": 2.659267674560867,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00042929637875389486,
      "loss": 5.3265,
      "mean_token_accuracy": 0.20243245214223862,
      "num_tokens": 61674130.0,
      "step": 30430
    },
    {
      "entropy": 6.077041959762573,
      "epoch": 2.6597046229135715,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042927369741204357,
      "loss": 5.3448,
      "mean_token_accuracy": 0.19857798218727113,
      "num_tokens": 61683612.0,
      "step": 30435
    },
    {
      "entropy": 5.975652885437012,
      "epoch": 2.6601415712662764,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004292510131111419,
      "loss": 5.3101,
      "mean_token_accuracy": 0.20455690771341323,
      "num_tokens": 61693156.0,
      "step": 30440
    },
    {
      "entropy": 5.973151826858521,
      "epoch": 2.660578519618981,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042922832585162496,
      "loss": 5.1926,
      "mean_token_accuracy": 0.20551849007606507,
      "num_tokens": 61701925.0,
      "step": 30445
    },
    {
      "entropy": 6.021953296661377,
      "epoch": 2.6610154679716858,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042920563563392803,
      "loss": 5.242,
      "mean_token_accuracy": 0.20999474972486495,
      "num_tokens": 61712388.0,
      "step": 30450
    },
    {
      "entropy": 5.973141813278199,
      "epoch": 2.6614524163243907,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042918294245848605,
      "loss": 5.2331,
      "mean_token_accuracy": 0.20247334837913514,
      "num_tokens": 61722402.0,
      "step": 30455
    },
    {
      "entropy": 6.0527183532714846,
      "epoch": 2.661889364677095,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042916024632573454,
      "loss": 5.3336,
      "mean_token_accuracy": 0.1998656466603279,
      "num_tokens": 61733160.0,
      "step": 30460
    },
    {
      "entropy": 6.044924020767212,
      "epoch": 2.6623263130297996,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042913754723610865,
      "loss": 5.2493,
      "mean_token_accuracy": 0.20771287977695466,
      "num_tokens": 61742536.0,
      "step": 30465
    },
    {
      "entropy": 6.006866598129273,
      "epoch": 2.6627632613825045,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004291148451900439,
      "loss": 5.2871,
      "mean_token_accuracy": 0.20239067673683167,
      "num_tokens": 61753439.0,
      "step": 30470
    },
    {
      "entropy": 5.958768320083618,
      "epoch": 2.6632002097352094,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042909214018797565,
      "loss": 5.2663,
      "mean_token_accuracy": 0.20867546200752257,
      "num_tokens": 61763994.0,
      "step": 30475
    },
    {
      "entropy": 6.048877811431884,
      "epoch": 2.663637158087914,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004290694322303394,
      "loss": 5.2984,
      "mean_token_accuracy": 0.20281277000904083,
      "num_tokens": 61774082.0,
      "step": 30480
    },
    {
      "entropy": 5.970942974090576,
      "epoch": 2.664074106440619,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042904672131757065,
      "loss": 5.2384,
      "mean_token_accuracy": 0.21124094873666763,
      "num_tokens": 61784528.0,
      "step": 30485
    },
    {
      "entropy": 5.998379898071289,
      "epoch": 2.6645110547933233,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004290240074501051,
      "loss": 5.2611,
      "mean_token_accuracy": 0.20443346351385117,
      "num_tokens": 61794267.0,
      "step": 30490
    },
    {
      "entropy": 6.032269048690796,
      "epoch": 2.664948003146028,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004290012906283784,
      "loss": 5.2642,
      "mean_token_accuracy": 0.20596137195825576,
      "num_tokens": 61804842.0,
      "step": 30495
    },
    {
      "entropy": 6.035627222061157,
      "epoch": 2.6653849514987327,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042897857085282627,
      "loss": 5.2831,
      "mean_token_accuracy": 0.2025743007659912,
      "num_tokens": 61814316.0,
      "step": 30500
    },
    {
      "entropy": 6.007596254348755,
      "epoch": 2.6658218998514376,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042895584812388445,
      "loss": 5.2907,
      "mean_token_accuracy": 0.2017189681529999,
      "num_tokens": 61823828.0,
      "step": 30505
    },
    {
      "entropy": 6.0255414009094235,
      "epoch": 2.6662588482041425,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004289331224419888,
      "loss": 5.2882,
      "mean_token_accuracy": 0.20144977420568466,
      "num_tokens": 61833878.0,
      "step": 30510
    },
    {
      "entropy": 5.9682090282440186,
      "epoch": 2.666695796556847,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042891039380757516,
      "loss": 5.2064,
      "mean_token_accuracy": 0.2065055251121521,
      "num_tokens": 61844591.0,
      "step": 30515
    },
    {
      "entropy": 6.021211624145508,
      "epoch": 2.6671327449095514,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042888766222107954,
      "loss": 5.2897,
      "mean_token_accuracy": 0.20583913922309877,
      "num_tokens": 61854791.0,
      "step": 30520
    },
    {
      "entropy": 5.97507152557373,
      "epoch": 2.6675696932622563,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000428864927682938,
      "loss": 5.3139,
      "mean_token_accuracy": 0.2020081326365471,
      "num_tokens": 61865194.0,
      "step": 30525
    },
    {
      "entropy": 5.97273530960083,
      "epoch": 2.6680066416149613,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042884219019358644,
      "loss": 5.22,
      "mean_token_accuracy": 0.20226111710071565,
      "num_tokens": 61874827.0,
      "step": 30530
    },
    {
      "entropy": 6.047497177124024,
      "epoch": 2.6684435899676657,
      "grad_norm": 1.125,
      "learning_rate": 0.0004288194497534611,
      "loss": 5.2921,
      "mean_token_accuracy": 0.1991311341524124,
      "num_tokens": 61885596.0,
      "step": 30535
    },
    {
      "entropy": 5.967178773880005,
      "epoch": 2.6688805383203706,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042879670636299805,
      "loss": 5.1546,
      "mean_token_accuracy": 0.21285105049610137,
      "num_tokens": 61895732.0,
      "step": 30540
    },
    {
      "entropy": 5.976064205169678,
      "epoch": 2.669317486673075,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042877396002263357,
      "loss": 5.2614,
      "mean_token_accuracy": 0.19978971630334855,
      "num_tokens": 61905581.0,
      "step": 30545
    },
    {
      "entropy": 5.969105339050293,
      "epoch": 2.66975443502578,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042875121073280404,
      "loss": 5.2098,
      "mean_token_accuracy": 0.21016037911176683,
      "num_tokens": 61914926.0,
      "step": 30550
    },
    {
      "entropy": 5.9371928691864015,
      "epoch": 2.6701913833784845,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042872845849394563,
      "loss": 5.2587,
      "mean_token_accuracy": 0.20497529655694963,
      "num_tokens": 61926501.0,
      "step": 30555
    },
    {
      "entropy": 6.0585442066192625,
      "epoch": 2.6706283317311894,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042870570330649495,
      "loss": 5.3267,
      "mean_token_accuracy": 0.1980424180626869,
      "num_tokens": 61937161.0,
      "step": 30560
    },
    {
      "entropy": 5.995780897140503,
      "epoch": 2.6710652800838943,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004286829451708883,
      "loss": 5.2363,
      "mean_token_accuracy": 0.20484184175729753,
      "num_tokens": 61949125.0,
      "step": 30565
    },
    {
      "entropy": 5.971923112869263,
      "epoch": 2.671502228436599,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004286601840875622,
      "loss": 5.1596,
      "mean_token_accuracy": 0.20863838195800782,
      "num_tokens": 61958110.0,
      "step": 30570
    },
    {
      "entropy": 5.975925159454346,
      "epoch": 2.6719391767893033,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004286374200569533,
      "loss": 5.2492,
      "mean_token_accuracy": 0.2046648845076561,
      "num_tokens": 61967805.0,
      "step": 30575
    },
    {
      "entropy": 5.999409246444702,
      "epoch": 2.672376125142008,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042861465307949815,
      "loss": 5.2738,
      "mean_token_accuracy": 0.2093235194683075,
      "num_tokens": 61977189.0,
      "step": 30580
    },
    {
      "entropy": 6.0235284805297855,
      "epoch": 2.672813073494713,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042859188315563345,
      "loss": 5.3724,
      "mean_token_accuracy": 0.19953493922948837,
      "num_tokens": 61988568.0,
      "step": 30585
    },
    {
      "entropy": 5.995709943771362,
      "epoch": 2.6732500218474176,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042856911028579597,
      "loss": 5.2086,
      "mean_token_accuracy": 0.20679156333208085,
      "num_tokens": 61999316.0,
      "step": 30590
    },
    {
      "entropy": 6.019359922409057,
      "epoch": 2.6736869702001225,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042854633447042247,
      "loss": 5.294,
      "mean_token_accuracy": 0.20554877519607545,
      "num_tokens": 62009345.0,
      "step": 30595
    },
    {
      "entropy": 5.985444355010986,
      "epoch": 2.674123918552827,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042852355570994983,
      "loss": 5.2376,
      "mean_token_accuracy": 0.20671343803405762,
      "num_tokens": 62020847.0,
      "step": 30600
    },
    {
      "entropy": 5.972698593139649,
      "epoch": 2.674560866905532,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042850077400481495,
      "loss": 5.2162,
      "mean_token_accuracy": 0.21411569863557817,
      "num_tokens": 62030506.0,
      "step": 30605
    },
    {
      "entropy": 6.046198320388794,
      "epoch": 2.6749978152582363,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042847798935545484,
      "loss": 5.3629,
      "mean_token_accuracy": 0.19642360061407088,
      "num_tokens": 62040238.0,
      "step": 30610
    },
    {
      "entropy": 5.946278619766235,
      "epoch": 2.6754347636109412,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004284552017623064,
      "loss": 5.2548,
      "mean_token_accuracy": 0.20573389679193496,
      "num_tokens": 62050192.0,
      "step": 30615
    },
    {
      "entropy": 5.994769716262818,
      "epoch": 2.675871711963646,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004284324112258068,
      "loss": 5.27,
      "mean_token_accuracy": 0.20571813285350798,
      "num_tokens": 62059475.0,
      "step": 30620
    },
    {
      "entropy": 6.021439743041992,
      "epoch": 2.6763086603163506,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042840961774639314,
      "loss": 5.2535,
      "mean_token_accuracy": 0.19955746680498124,
      "num_tokens": 62068799.0,
      "step": 30625
    },
    {
      "entropy": 6.016920757293701,
      "epoch": 2.676745608669055,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004283868213245026,
      "loss": 5.3275,
      "mean_token_accuracy": 0.19712096005678176,
      "num_tokens": 62078205.0,
      "step": 30630
    },
    {
      "entropy": 5.965047979354859,
      "epoch": 2.67718255702176,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042836402196057255,
      "loss": 5.1877,
      "mean_token_accuracy": 0.2079334557056427,
      "num_tokens": 62088084.0,
      "step": 30635
    },
    {
      "entropy": 5.936508560180664,
      "epoch": 2.677619505374465,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042834121965504014,
      "loss": 5.2326,
      "mean_token_accuracy": 0.20651101022958757,
      "num_tokens": 62098709.0,
      "step": 30640
    },
    {
      "entropy": 5.89131269454956,
      "epoch": 2.6780564537271694,
      "grad_norm": 1.25,
      "learning_rate": 0.0004283184144083428,
      "loss": 5.1935,
      "mean_token_accuracy": 0.20750370472669602,
      "num_tokens": 62108949.0,
      "step": 30645
    },
    {
      "entropy": 5.976825761795044,
      "epoch": 2.6784934020798743,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042829560622091795,
      "loss": 5.2363,
      "mean_token_accuracy": 0.21065573394298553,
      "num_tokens": 62118922.0,
      "step": 30650
    },
    {
      "entropy": 6.0060337543487545,
      "epoch": 2.6789303504325788,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042827279509320297,
      "loss": 5.2411,
      "mean_token_accuracy": 0.20473716855049134,
      "num_tokens": 62128396.0,
      "step": 30655
    },
    {
      "entropy": 6.047131061553955,
      "epoch": 2.6793672987852837,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004282499810256356,
      "loss": 5.332,
      "mean_token_accuracy": 0.19041860699653626,
      "num_tokens": 62139223.0,
      "step": 30660
    },
    {
      "entropy": 5.959854936599731,
      "epoch": 2.679804247137988,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042822716401865314,
      "loss": 5.2327,
      "mean_token_accuracy": 0.20661506056785583,
      "num_tokens": 62149432.0,
      "step": 30665
    },
    {
      "entropy": 5.981575059890747,
      "epoch": 2.680241195490693,
      "grad_norm": 1.25,
      "learning_rate": 0.00042820434407269347,
      "loss": 5.2676,
      "mean_token_accuracy": 0.20603243708610536,
      "num_tokens": 62160945.0,
      "step": 30670
    },
    {
      "entropy": 6.026442813873291,
      "epoch": 2.680678143843398,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004281815211881942,
      "loss": 5.2668,
      "mean_token_accuracy": 0.19729045629501343,
      "num_tokens": 62171514.0,
      "step": 30675
    },
    {
      "entropy": 6.009144878387451,
      "epoch": 2.6811150921961024,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004281586953655931,
      "loss": 5.3166,
      "mean_token_accuracy": 0.1981555461883545,
      "num_tokens": 62181011.0,
      "step": 30680
    },
    {
      "entropy": 5.955629062652588,
      "epoch": 2.681552040548807,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004281358666053279,
      "loss": 5.2451,
      "mean_token_accuracy": 0.20831608027219772,
      "num_tokens": 62190915.0,
      "step": 30685
    },
    {
      "entropy": 5.9936872005462645,
      "epoch": 2.681988988901512,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004281130349078366,
      "loss": 5.2434,
      "mean_token_accuracy": 0.2041707307100296,
      "num_tokens": 62200660.0,
      "step": 30690
    },
    {
      "entropy": 5.969823360443115,
      "epoch": 2.6824259372542167,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000428090200273557,
      "loss": 5.268,
      "mean_token_accuracy": 0.20499007552862167,
      "num_tokens": 62210425.0,
      "step": 30695
    },
    {
      "entropy": 5.975924682617188,
      "epoch": 2.682862885606921,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004280673627029272,
      "loss": 5.2222,
      "mean_token_accuracy": 0.20503124892711638,
      "num_tokens": 62220921.0,
      "step": 30700
    },
    {
      "entropy": 5.988649654388428,
      "epoch": 2.683299833959626,
      "grad_norm": 1.25,
      "learning_rate": 0.0004280445221963852,
      "loss": 5.2714,
      "mean_token_accuracy": 0.20116584599018097,
      "num_tokens": 62230561.0,
      "step": 30705
    },
    {
      "entropy": 6.030050468444824,
      "epoch": 2.6837367823123306,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000428021678754369,
      "loss": 5.3104,
      "mean_token_accuracy": 0.19878402054309846,
      "num_tokens": 62240744.0,
      "step": 30710
    },
    {
      "entropy": 5.960582447052002,
      "epoch": 2.6841737306650355,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042799883237731683,
      "loss": 5.2083,
      "mean_token_accuracy": 0.21080002784729004,
      "num_tokens": 62250902.0,
      "step": 30715
    },
    {
      "entropy": 6.043517923355102,
      "epoch": 2.68461067901774,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042797598306566694,
      "loss": 5.3101,
      "mean_token_accuracy": 0.20353104174137115,
      "num_tokens": 62260459.0,
      "step": 30720
    },
    {
      "entropy": 6.005472040176391,
      "epoch": 2.685047627370445,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004279531308198575,
      "loss": 5.1755,
      "mean_token_accuracy": 0.20696819871664046,
      "num_tokens": 62269001.0,
      "step": 30725
    },
    {
      "entropy": 5.938387060165406,
      "epoch": 2.6854845757231494,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042793027564032684,
      "loss": 5.212,
      "mean_token_accuracy": 0.204866398870945,
      "num_tokens": 62279197.0,
      "step": 30730
    },
    {
      "entropy": 5.958546495437622,
      "epoch": 2.6859215240758543,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042790741752751337,
      "loss": 5.2289,
      "mean_token_accuracy": 0.20261396318674088,
      "num_tokens": 62289229.0,
      "step": 30735
    },
    {
      "entropy": 5.963184547424317,
      "epoch": 2.6863584724285587,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042788455648185556,
      "loss": 5.2378,
      "mean_token_accuracy": 0.2092027932405472,
      "num_tokens": 62298819.0,
      "step": 30740
    },
    {
      "entropy": 6.035699796676636,
      "epoch": 2.6867954207812637,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042786169250379175,
      "loss": 5.2309,
      "mean_token_accuracy": 0.20560273677110671,
      "num_tokens": 62308567.0,
      "step": 30745
    },
    {
      "entropy": 6.015159606933594,
      "epoch": 2.6872323691339686,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042783882559376067,
      "loss": 5.2941,
      "mean_token_accuracy": 0.20158660411834717,
      "num_tokens": 62318552.0,
      "step": 30750
    },
    {
      "entropy": 5.971013736724854,
      "epoch": 2.687669317486673,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042781595575220086,
      "loss": 5.2086,
      "mean_token_accuracy": 0.2128511533141136,
      "num_tokens": 62328430.0,
      "step": 30755
    },
    {
      "entropy": 6.03854284286499,
      "epoch": 2.688106265839378,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042779308297955084,
      "loss": 5.3306,
      "mean_token_accuracy": 0.19839993715286255,
      "num_tokens": 62338540.0,
      "step": 30760
    },
    {
      "entropy": 5.9577531814575195,
      "epoch": 2.6885432141920824,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042777020727624945,
      "loss": 5.2646,
      "mean_token_accuracy": 0.2046046420931816,
      "num_tokens": 62348161.0,
      "step": 30765
    },
    {
      "entropy": 6.046537065505982,
      "epoch": 2.6889801625447873,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004277473286427354,
      "loss": 5.3459,
      "mean_token_accuracy": 0.19577657133340837,
      "num_tokens": 62358217.0,
      "step": 30770
    },
    {
      "entropy": 5.981743621826172,
      "epoch": 2.689417110897492,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042772444707944767,
      "loss": 5.2266,
      "mean_token_accuracy": 0.20598673224449157,
      "num_tokens": 62368801.0,
      "step": 30775
    },
    {
      "entropy": 6.002650356292724,
      "epoch": 2.6898540592501967,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004277015625868249,
      "loss": 5.2605,
      "mean_token_accuracy": 0.20280737429857254,
      "num_tokens": 62377632.0,
      "step": 30780
    },
    {
      "entropy": 6.036296081542969,
      "epoch": 2.690291007602901,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004276786751653062,
      "loss": 5.2959,
      "mean_token_accuracy": 0.20082113593816758,
      "num_tokens": 62388056.0,
      "step": 30785
    },
    {
      "entropy": 6.023630809783936,
      "epoch": 2.690727955955606,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004276557848153305,
      "loss": 5.238,
      "mean_token_accuracy": 0.20916089415550232,
      "num_tokens": 62398015.0,
      "step": 30790
    },
    {
      "entropy": 6.02565279006958,
      "epoch": 2.6911649043083106,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042763289153733685,
      "loss": 5.3082,
      "mean_token_accuracy": 0.2001193106174469,
      "num_tokens": 62408174.0,
      "step": 30795
    },
    {
      "entropy": 6.034869003295898,
      "epoch": 2.6916018526610155,
      "grad_norm": 1.25,
      "learning_rate": 0.00042760999533176435,
      "loss": 5.2671,
      "mean_token_accuracy": 0.20059133619070052,
      "num_tokens": 62417571.0,
      "step": 30800
    },
    {
      "entropy": 5.975448417663574,
      "epoch": 2.6920388010137204,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004275870961990523,
      "loss": 5.2924,
      "mean_token_accuracy": 0.2089310973882675,
      "num_tokens": 62427734.0,
      "step": 30805
    },
    {
      "entropy": 6.040390729904175,
      "epoch": 2.692475749366425,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042756419413963965,
      "loss": 5.2658,
      "mean_token_accuracy": 0.19927252680063248,
      "num_tokens": 62437109.0,
      "step": 30810
    },
    {
      "entropy": 5.979140949249268,
      "epoch": 2.6929126977191293,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042754128915396583,
      "loss": 5.2473,
      "mean_token_accuracy": 0.2019638240337372,
      "num_tokens": 62448163.0,
      "step": 30815
    },
    {
      "entropy": 5.972826385498047,
      "epoch": 2.6933496460718342,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042751838124247015,
      "loss": 5.2998,
      "mean_token_accuracy": 0.19755506962537767,
      "num_tokens": 62457993.0,
      "step": 30820
    },
    {
      "entropy": 6.032448625564575,
      "epoch": 2.693786594424539,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042749547040559204,
      "loss": 5.3426,
      "mean_token_accuracy": 0.20465001612901687,
      "num_tokens": 62468554.0,
      "step": 30825
    },
    {
      "entropy": 5.989367389678955,
      "epoch": 2.6942235427772436,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004274725566437709,
      "loss": 5.2408,
      "mean_token_accuracy": 0.20548056215047836,
      "num_tokens": 62479204.0,
      "step": 30830
    },
    {
      "entropy": 5.98424654006958,
      "epoch": 2.6946604911299485,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004274496399574463,
      "loss": 5.2409,
      "mean_token_accuracy": 0.20476736426353453,
      "num_tokens": 62491391.0,
      "step": 30835
    },
    {
      "entropy": 5.950775623321533,
      "epoch": 2.695097439482653,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004274267203470576,
      "loss": 5.2062,
      "mean_token_accuracy": 0.2131299778819084,
      "num_tokens": 62500796.0,
      "step": 30840
    },
    {
      "entropy": 5.973508262634278,
      "epoch": 2.695534387835358,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042740379781304455,
      "loss": 5.2322,
      "mean_token_accuracy": 0.20483021885156633,
      "num_tokens": 62509996.0,
      "step": 30845
    },
    {
      "entropy": 6.024729490280151,
      "epoch": 2.6959713361880624,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042738087235584684,
      "loss": 5.2522,
      "mean_token_accuracy": 0.20182168185710908,
      "num_tokens": 62519676.0,
      "step": 30850
    },
    {
      "entropy": 6.018858242034912,
      "epoch": 2.6964082845407673,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042735794397590407,
      "loss": 5.2507,
      "mean_token_accuracy": 0.20502467304468155,
      "num_tokens": 62530775.0,
      "step": 30855
    },
    {
      "entropy": 5.9694756984710695,
      "epoch": 2.696845232893472,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004273350126736562,
      "loss": 5.2183,
      "mean_token_accuracy": 0.2127676397562027,
      "num_tokens": 62540643.0,
      "step": 30860
    },
    {
      "entropy": 6.013828277587891,
      "epoch": 2.6972821812461767,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004273120784495429,
      "loss": 5.2139,
      "mean_token_accuracy": 0.21180282831192015,
      "num_tokens": 62549766.0,
      "step": 30865
    },
    {
      "entropy": 5.980956649780273,
      "epoch": 2.697719129598881,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004272891413040042,
      "loss": 5.2017,
      "mean_token_accuracy": 0.2077423170208931,
      "num_tokens": 62560130.0,
      "step": 30870
    },
    {
      "entropy": 5.999184942245483,
      "epoch": 2.698156077951586,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004272662012374799,
      "loss": 5.2848,
      "mean_token_accuracy": 0.20413743555545807,
      "num_tokens": 62571256.0,
      "step": 30875
    },
    {
      "entropy": 5.970711994171142,
      "epoch": 2.698593026304291,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004272432582504101,
      "loss": 5.1818,
      "mean_token_accuracy": 0.2155964970588684,
      "num_tokens": 62582269.0,
      "step": 30880
    },
    {
      "entropy": 5.942597150802612,
      "epoch": 2.6990299746569955,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042722031234323485,
      "loss": 5.2275,
      "mean_token_accuracy": 0.21093344390392305,
      "num_tokens": 62591875.0,
      "step": 30885
    },
    {
      "entropy": 6.058519458770752,
      "epoch": 2.6994669230097004,
      "grad_norm": 1.328125,
      "learning_rate": 0.00042719736351639416,
      "loss": 5.3355,
      "mean_token_accuracy": 0.200768318772316,
      "num_tokens": 62602467.0,
      "step": 30890
    },
    {
      "entropy": 6.031662273406982,
      "epoch": 2.699903871362405,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004271744117703284,
      "loss": 5.2135,
      "mean_token_accuracy": 0.20406518876552582,
      "num_tokens": 62611191.0,
      "step": 30895
    },
    {
      "entropy": 5.9395208835601805,
      "epoch": 2.7003408197151098,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042715145710547767,
      "loss": 5.2968,
      "mean_token_accuracy": 0.2076843723654747,
      "num_tokens": 62620961.0,
      "step": 30900
    },
    {
      "entropy": 5.9793219566345215,
      "epoch": 2.700777768067814,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004271284995222822,
      "loss": 5.2381,
      "mean_token_accuracy": 0.20898657888174058,
      "num_tokens": 62631972.0,
      "step": 30905
    },
    {
      "entropy": 6.0103020668029785,
      "epoch": 2.701214716420519,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004271055390211824,
      "loss": 5.1995,
      "mean_token_accuracy": 0.21502260863780975,
      "num_tokens": 62641073.0,
      "step": 30910
    },
    {
      "entropy": 5.977571249008179,
      "epoch": 2.701651664773224,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004270825756026187,
      "loss": 5.2492,
      "mean_token_accuracy": 0.20233619958162308,
      "num_tokens": 62650364.0,
      "step": 30915
    },
    {
      "entropy": 6.001585245132446,
      "epoch": 2.7020886131259285,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004270596092670315,
      "loss": 5.2856,
      "mean_token_accuracy": 0.20714316815137862,
      "num_tokens": 62659846.0,
      "step": 30920
    },
    {
      "entropy": 6.056235313415527,
      "epoch": 2.702525561478633,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004270366400148613,
      "loss": 5.2793,
      "mean_token_accuracy": 0.2095634713768959,
      "num_tokens": 62670466.0,
      "step": 30925
    },
    {
      "entropy": 5.974641132354736,
      "epoch": 2.702962509831338,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004270136678465487,
      "loss": 5.1987,
      "mean_token_accuracy": 0.21052664816379546,
      "num_tokens": 62680330.0,
      "step": 30930
    },
    {
      "entropy": 6.052443981170654,
      "epoch": 2.703399458184043,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042699069276253427,
      "loss": 5.3247,
      "mean_token_accuracy": 0.19620506912469865,
      "num_tokens": 62690177.0,
      "step": 30935
    },
    {
      "entropy": 6.020646715164185,
      "epoch": 2.7038364065367473,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004269677147632587,
      "loss": 5.2607,
      "mean_token_accuracy": 0.20501824021339415,
      "num_tokens": 62699365.0,
      "step": 30940
    },
    {
      "entropy": 5.878935384750366,
      "epoch": 2.704273354889452,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004269447338491628,
      "loss": 5.2138,
      "mean_token_accuracy": 0.21099217236042023,
      "num_tokens": 62709825.0,
      "step": 30945
    },
    {
      "entropy": 6.045932197570801,
      "epoch": 2.7047103032421567,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004269217500206873,
      "loss": 5.301,
      "mean_token_accuracy": 0.20744674801826476,
      "num_tokens": 62719848.0,
      "step": 30950
    },
    {
      "entropy": 6.047064781188965,
      "epoch": 2.7051472515948616,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042689876327827293,
      "loss": 5.3322,
      "mean_token_accuracy": 0.2000260904431343,
      "num_tokens": 62730405.0,
      "step": 30955
    },
    {
      "entropy": 5.983474969863892,
      "epoch": 2.705584199947566,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042687577362236073,
      "loss": 5.2928,
      "mean_token_accuracy": 0.199974524974823,
      "num_tokens": 62740275.0,
      "step": 30960
    },
    {
      "entropy": 6.001554536819458,
      "epoch": 2.706021148300271,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004268527810533917,
      "loss": 5.2511,
      "mean_token_accuracy": 0.2093322157859802,
      "num_tokens": 62750847.0,
      "step": 30965
    },
    {
      "entropy": 6.0227189540863035,
      "epoch": 2.706458096652976,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042682978557180665,
      "loss": 5.3252,
      "mean_token_accuracy": 0.19507597982883454,
      "num_tokens": 62761688.0,
      "step": 30970
    },
    {
      "entropy": 6.021255254745483,
      "epoch": 2.7068950450056803,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00042680678717804676,
      "loss": 5.3115,
      "mean_token_accuracy": 0.19861642718315126,
      "num_tokens": 62771168.0,
      "step": 30975
    },
    {
      "entropy": 5.924331998825073,
      "epoch": 2.707331993358385,
      "grad_norm": 1.25,
      "learning_rate": 0.00042678378587255315,
      "loss": 5.1635,
      "mean_token_accuracy": 0.21693459004163743,
      "num_tokens": 62780412.0,
      "step": 30980
    },
    {
      "entropy": 6.006033611297608,
      "epoch": 2.7077689417110897,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042676078165576706,
      "loss": 5.2782,
      "mean_token_accuracy": 0.2066545605659485,
      "num_tokens": 62791673.0,
      "step": 30985
    },
    {
      "entropy": 6.0455926895141605,
      "epoch": 2.7082058900637946,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004267377745281296,
      "loss": 5.2885,
      "mean_token_accuracy": 0.19716450721025466,
      "num_tokens": 62800980.0,
      "step": 30990
    },
    {
      "entropy": 5.96173677444458,
      "epoch": 2.708642838416499,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042671476449008205,
      "loss": 5.2353,
      "mean_token_accuracy": 0.2081724926829338,
      "num_tokens": 62813183.0,
      "step": 30995
    },
    {
      "entropy": 5.989572906494141,
      "epoch": 2.709079786769204,
      "grad_norm": 1.1875,
      "learning_rate": 0.000426691751542066,
      "loss": 5.2657,
      "mean_token_accuracy": 0.20166302621364593,
      "num_tokens": 62823170.0,
      "step": 31000
    },
    {
      "entropy": 6.002878427505493,
      "epoch": 2.7095167351219085,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004266687356845225,
      "loss": 5.2465,
      "mean_token_accuracy": 0.203484146296978,
      "num_tokens": 62833024.0,
      "step": 31005
    },
    {
      "entropy": 5.98090124130249,
      "epoch": 2.7099536834746134,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042664571691789323,
      "loss": 5.2712,
      "mean_token_accuracy": 0.20427127778530121,
      "num_tokens": 62842533.0,
      "step": 31010
    },
    {
      "entropy": 6.035297632217407,
      "epoch": 2.710390631827318,
      "grad_norm": 1.125,
      "learning_rate": 0.0004266226952426197,
      "loss": 5.2968,
      "mean_token_accuracy": 0.2003532886505127,
      "num_tokens": 62852984.0,
      "step": 31015
    },
    {
      "entropy": 6.0343852043151855,
      "epoch": 2.710827580180023,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004265996706591434,
      "loss": 5.3081,
      "mean_token_accuracy": 0.1964734748005867,
      "num_tokens": 62861926.0,
      "step": 31020
    },
    {
      "entropy": 5.993601560592651,
      "epoch": 2.7112645285327277,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042657664316790594,
      "loss": 5.1895,
      "mean_token_accuracy": 0.21239721328020095,
      "num_tokens": 62872194.0,
      "step": 31025
    },
    {
      "entropy": 6.001134157180786,
      "epoch": 2.711701476885432,
      "grad_norm": 1.25,
      "learning_rate": 0.000426553612769349,
      "loss": 5.1902,
      "mean_token_accuracy": 0.2073300898075104,
      "num_tokens": 62882887.0,
      "step": 31030
    },
    {
      "entropy": 5.96626877784729,
      "epoch": 2.7121384252381366,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042653057946391446,
      "loss": 5.2573,
      "mean_token_accuracy": 0.20491715967655183,
      "num_tokens": 62893563.0,
      "step": 31035
    },
    {
      "entropy": 6.0208433151245115,
      "epoch": 2.7125753735908416,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042650754325204395,
      "loss": 5.2576,
      "mean_token_accuracy": 0.20591710656881332,
      "num_tokens": 62902691.0,
      "step": 31040
    },
    {
      "entropy": 5.967905902862549,
      "epoch": 2.7130123219435465,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042648450413417937,
      "loss": 5.1996,
      "mean_token_accuracy": 0.2092528909444809,
      "num_tokens": 62912261.0,
      "step": 31045
    },
    {
      "entropy": 6.060820722579956,
      "epoch": 2.713449270296251,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042646146211076263,
      "loss": 5.3208,
      "mean_token_accuracy": 0.1995721012353897,
      "num_tokens": 62922438.0,
      "step": 31050
    },
    {
      "entropy": 6.011176204681396,
      "epoch": 2.713886218648956,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004264384171822357,
      "loss": 5.1897,
      "mean_token_accuracy": 0.20802783071994782,
      "num_tokens": 62932378.0,
      "step": 31055
    },
    {
      "entropy": 6.048171138763427,
      "epoch": 2.7143231670016603,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00042641536934904064,
      "loss": 5.3655,
      "mean_token_accuracy": 0.19247663468122483,
      "num_tokens": 62943590.0,
      "step": 31060
    },
    {
      "entropy": 5.982727193832398,
      "epoch": 2.7147601153543652,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004263923186116193,
      "loss": 5.2549,
      "mean_token_accuracy": 0.20487824231386184,
      "num_tokens": 62953393.0,
      "step": 31065
    },
    {
      "entropy": 6.030145740509033,
      "epoch": 2.7151970637070697,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042636926497041406,
      "loss": 5.2557,
      "mean_token_accuracy": 0.2068746104836464,
      "num_tokens": 62964407.0,
      "step": 31070
    },
    {
      "entropy": 5.9965616226196286,
      "epoch": 2.7156340120597746,
      "grad_norm": 1.171875,
      "learning_rate": 0.000426346208425867,
      "loss": 5.3047,
      "mean_token_accuracy": 0.19905711710453033,
      "num_tokens": 62974968.0,
      "step": 31075
    },
    {
      "entropy": 6.031717443466187,
      "epoch": 2.716070960412479,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004263231489784203,
      "loss": 5.2195,
      "mean_token_accuracy": 0.20322959423065184,
      "num_tokens": 62984978.0,
      "step": 31080
    },
    {
      "entropy": 5.981384944915772,
      "epoch": 2.716507908765184,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004263000866285164,
      "loss": 5.3025,
      "mean_token_accuracy": 0.20316310822963715,
      "num_tokens": 62995139.0,
      "step": 31085
    },
    {
      "entropy": 5.978621006011963,
      "epoch": 2.7169448571178885,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042627702137659746,
      "loss": 5.3068,
      "mean_token_accuracy": 0.20541045516729356,
      "num_tokens": 63006127.0,
      "step": 31090
    },
    {
      "entropy": 6.0684020042419435,
      "epoch": 2.7173818054705934,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042625395322310605,
      "loss": 5.3216,
      "mean_token_accuracy": 0.19973017871379853,
      "num_tokens": 63015844.0,
      "step": 31095
    },
    {
      "entropy": 6.048851776123047,
      "epoch": 2.7178187538232983,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042623088216848446,
      "loss": 5.263,
      "mean_token_accuracy": 0.20630283355712892,
      "num_tokens": 63025815.0,
      "step": 31100
    },
    {
      "entropy": 6.028672361373902,
      "epoch": 2.7182557021760028,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042620780821317543,
      "loss": 5.3412,
      "mean_token_accuracy": 0.19784766137599946,
      "num_tokens": 63036108.0,
      "step": 31105
    },
    {
      "entropy": 5.938451147079467,
      "epoch": 2.7186926505287072,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004261847313576214,
      "loss": 5.2235,
      "mean_token_accuracy": 0.20956489741802214,
      "num_tokens": 63046305.0,
      "step": 31110
    },
    {
      "entropy": 6.059383869171143,
      "epoch": 2.719129598881412,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000426161651602265,
      "loss": 5.378,
      "mean_token_accuracy": 0.1928788274526596,
      "num_tokens": 63057115.0,
      "step": 31115
    },
    {
      "entropy": 6.0357483386993405,
      "epoch": 2.719566547234117,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042613856894754886,
      "loss": 5.2354,
      "mean_token_accuracy": 0.21007314920425416,
      "num_tokens": 63067730.0,
      "step": 31120
    },
    {
      "entropy": 5.980817031860352,
      "epoch": 2.7200034955868215,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042611548339391574,
      "loss": 5.2391,
      "mean_token_accuracy": 0.20610693097114563,
      "num_tokens": 63077510.0,
      "step": 31125
    },
    {
      "entropy": 5.930734014511108,
      "epoch": 2.7204404439395264,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042609239494180857,
      "loss": 5.204,
      "mean_token_accuracy": 0.20962383449077607,
      "num_tokens": 63087160.0,
      "step": 31130
    },
    {
      "entropy": 6.024066591262818,
      "epoch": 2.720877392292231,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004260693035916701,
      "loss": 5.2751,
      "mean_token_accuracy": 0.20536193400621414,
      "num_tokens": 63097284.0,
      "step": 31135
    },
    {
      "entropy": 6.05815806388855,
      "epoch": 2.721314340644936,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042604620934394313,
      "loss": 5.2865,
      "mean_token_accuracy": 0.19453266263008118,
      "num_tokens": 63106910.0,
      "step": 31140
    },
    {
      "entropy": 6.000786066055298,
      "epoch": 2.7217512889976403,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042602311219907076,
      "loss": 5.1833,
      "mean_token_accuracy": 0.20629042387008667,
      "num_tokens": 63116129.0,
      "step": 31145
    },
    {
      "entropy": 5.92711763381958,
      "epoch": 2.722188237350345,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042600001215749593,
      "loss": 5.2753,
      "mean_token_accuracy": 0.20214060842990875,
      "num_tokens": 63125602.0,
      "step": 31150
    },
    {
      "entropy": 6.019848346710205,
      "epoch": 2.72262518570305,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004259769092196618,
      "loss": 5.2351,
      "mean_token_accuracy": 0.20840574204921722,
      "num_tokens": 63136592.0,
      "step": 31155
    },
    {
      "entropy": 6.077630519866943,
      "epoch": 2.7230621340557546,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042595380338601144,
      "loss": 5.2525,
      "mean_token_accuracy": 0.20695895552635193,
      "num_tokens": 63147040.0,
      "step": 31160
    },
    {
      "entropy": 5.9540855407714846,
      "epoch": 2.723499082408459,
      "grad_norm": 1.203125,
      "learning_rate": 0.000425930694656988,
      "loss": 5.2499,
      "mean_token_accuracy": 0.1988942563533783,
      "num_tokens": 63156766.0,
      "step": 31165
    },
    {
      "entropy": 5.995650148391723,
      "epoch": 2.723936030761164,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004259075830330348,
      "loss": 5.2534,
      "mean_token_accuracy": 0.2061636596918106,
      "num_tokens": 63166253.0,
      "step": 31170
    },
    {
      "entropy": 6.00070538520813,
      "epoch": 2.724372979113869,
      "grad_norm": 1.265625,
      "learning_rate": 0.000425884468514595,
      "loss": 5.334,
      "mean_token_accuracy": 0.19932988733053209,
      "num_tokens": 63177352.0,
      "step": 31175
    },
    {
      "entropy": 6.033847379684448,
      "epoch": 2.7248099274665734,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004258613511021121,
      "loss": 5.3466,
      "mean_token_accuracy": 0.1965990588068962,
      "num_tokens": 63188092.0,
      "step": 31180
    },
    {
      "entropy": 5.982688474655151,
      "epoch": 2.7252468758192783,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004258382307960294,
      "loss": 5.2004,
      "mean_token_accuracy": 0.21387678533792495,
      "num_tokens": 63197677.0,
      "step": 31185
    },
    {
      "entropy": 6.024665355682373,
      "epoch": 2.7256838241719827,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004258151075967904,
      "loss": 5.3765,
      "mean_token_accuracy": 0.19715280830860138,
      "num_tokens": 63208536.0,
      "step": 31190
    },
    {
      "entropy": 5.9799864292144775,
      "epoch": 2.7261207725246877,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004257919815048387,
      "loss": 5.2216,
      "mean_token_accuracy": 0.20634587556123735,
      "num_tokens": 63217966.0,
      "step": 31195
    },
    {
      "entropy": 5.9948241233825685,
      "epoch": 2.726557720877392,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004257688525206176,
      "loss": 5.2579,
      "mean_token_accuracy": 0.2023555487394333,
      "num_tokens": 63227420.0,
      "step": 31200
    },
    {
      "entropy": 6.05230975151062,
      "epoch": 2.726994669230097,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004257457206445711,
      "loss": 5.3331,
      "mean_token_accuracy": 0.19869424849748613,
      "num_tokens": 63237887.0,
      "step": 31205
    },
    {
      "entropy": 6.000870037078857,
      "epoch": 2.727431617582802,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042572258587714255,
      "loss": 5.2443,
      "mean_token_accuracy": 0.2073824882507324,
      "num_tokens": 63247539.0,
      "step": 31210
    },
    {
      "entropy": 6.007664203643799,
      "epoch": 2.7278685659355064,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004256994482187759,
      "loss": 5.2022,
      "mean_token_accuracy": 0.2124006748199463,
      "num_tokens": 63257839.0,
      "step": 31215
    },
    {
      "entropy": 5.955444097518921,
      "epoch": 2.728305514288211,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042567630766991477,
      "loss": 5.3118,
      "mean_token_accuracy": 0.1992315024137497,
      "num_tokens": 63268667.0,
      "step": 31220
    },
    {
      "entropy": 6.046867752075196,
      "epoch": 2.728742462640916,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004256531642310032,
      "loss": 5.3007,
      "mean_token_accuracy": 0.2062206357717514,
      "num_tokens": 63277413.0,
      "step": 31225
    },
    {
      "entropy": 6.046495723724365,
      "epoch": 2.7291794109936207,
      "grad_norm": 1.234375,
      "learning_rate": 0.000425630017902485,
      "loss": 5.2924,
      "mean_token_accuracy": 0.20547138452529906,
      "num_tokens": 63287225.0,
      "step": 31230
    },
    {
      "entropy": 5.9786561012268065,
      "epoch": 2.729616359346325,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004256068686848041,
      "loss": 5.2667,
      "mean_token_accuracy": 0.20870791524648666,
      "num_tokens": 63297211.0,
      "step": 31235
    },
    {
      "entropy": 5.937107706069947,
      "epoch": 2.73005330769903,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004255837165784045,
      "loss": 5.2496,
      "mean_token_accuracy": 0.2088109716773033,
      "num_tokens": 63307443.0,
      "step": 31240
    },
    {
      "entropy": 5.996672439575195,
      "epoch": 2.7304902560517346,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004255605615837304,
      "loss": 5.315,
      "mean_token_accuracy": 0.20485783368349075,
      "num_tokens": 63317951.0,
      "step": 31245
    },
    {
      "entropy": 5.995319604873657,
      "epoch": 2.7309272044044395,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042553740370122574,
      "loss": 5.2352,
      "mean_token_accuracy": 0.20808180123567582,
      "num_tokens": 63328138.0,
      "step": 31250
    },
    {
      "entropy": 5.968439435958862,
      "epoch": 2.731364152757144,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004255142429313348,
      "loss": 5.1561,
      "mean_token_accuracy": 0.21707241237163544,
      "num_tokens": 63338229.0,
      "step": 31255
    },
    {
      "entropy": 6.048698091506958,
      "epoch": 2.731801101109849,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004254910792745019,
      "loss": 5.3097,
      "mean_token_accuracy": 0.20277788490056992,
      "num_tokens": 63348454.0,
      "step": 31260
    },
    {
      "entropy": 5.98923282623291,
      "epoch": 2.7322380494625538,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042546791273117125,
      "loss": 5.1426,
      "mean_token_accuracy": 0.21176755875349046,
      "num_tokens": 63357346.0,
      "step": 31265
    },
    {
      "entropy": 5.94808440208435,
      "epoch": 2.7326749978152582,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004254447433017871,
      "loss": 5.3144,
      "mean_token_accuracy": 0.2009742096066475,
      "num_tokens": 63368329.0,
      "step": 31270
    },
    {
      "entropy": 6.051198291778564,
      "epoch": 2.7331119461679627,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042542157098679397,
      "loss": 5.3386,
      "mean_token_accuracy": 0.19868330508470536,
      "num_tokens": 63378815.0,
      "step": 31275
    },
    {
      "entropy": 5.985528945922852,
      "epoch": 2.7335488945206676,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004253983957866364,
      "loss": 5.2307,
      "mean_token_accuracy": 0.2054128497838974,
      "num_tokens": 63390099.0,
      "step": 31280
    },
    {
      "entropy": 5.98143196105957,
      "epoch": 2.7339858428733725,
      "grad_norm": 1.421875,
      "learning_rate": 0.00042537521770175856,
      "loss": 5.3096,
      "mean_token_accuracy": 0.20192105919122696,
      "num_tokens": 63401244.0,
      "step": 31285
    },
    {
      "entropy": 5.984152317047119,
      "epoch": 2.734422791226077,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042535203673260544,
      "loss": 5.2397,
      "mean_token_accuracy": 0.20644793659448624,
      "num_tokens": 63410750.0,
      "step": 31290
    },
    {
      "entropy": 5.976120185852051,
      "epoch": 2.734859739578782,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042532885287962146,
      "loss": 5.2429,
      "mean_token_accuracy": 0.20619104355573653,
      "num_tokens": 63420936.0,
      "step": 31295
    },
    {
      "entropy": 5.948454475402832,
      "epoch": 2.7352966879314864,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042530566614325125,
      "loss": 5.2601,
      "mean_token_accuracy": 0.21047924011945723,
      "num_tokens": 63431614.0,
      "step": 31300
    },
    {
      "entropy": 5.965546751022339,
      "epoch": 2.7357336362841913,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042528247652393964,
      "loss": 5.2264,
      "mean_token_accuracy": 0.20380927622318268,
      "num_tokens": 63441298.0,
      "step": 31305
    },
    {
      "entropy": 5.987734270095825,
      "epoch": 2.7361705846368958,
      "grad_norm": 1.25,
      "learning_rate": 0.00042525928402213127,
      "loss": 5.2304,
      "mean_token_accuracy": 0.20684618800878524,
      "num_tokens": 63451542.0,
      "step": 31310
    },
    {
      "entropy": 6.038769340515136,
      "epoch": 2.7366075329896007,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004252360886382712,
      "loss": 5.2906,
      "mean_token_accuracy": 0.20360316038131715,
      "num_tokens": 63460863.0,
      "step": 31315
    },
    {
      "entropy": 6.028654384613037,
      "epoch": 2.7370444813423056,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042521289037280425,
      "loss": 5.3092,
      "mean_token_accuracy": 0.19681877344846727,
      "num_tokens": 63470499.0,
      "step": 31320
    },
    {
      "entropy": 6.085239315032959,
      "epoch": 2.73748142969501,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042518968922617534,
      "loss": 5.3,
      "mean_token_accuracy": 0.20195231437683106,
      "num_tokens": 63480803.0,
      "step": 31325
    },
    {
      "entropy": 5.965778017044068,
      "epoch": 2.7379183780477145,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004251664851988295,
      "loss": 5.2089,
      "mean_token_accuracy": 0.2095950350165367,
      "num_tokens": 63491544.0,
      "step": 31330
    },
    {
      "entropy": 5.946390438079834,
      "epoch": 2.7383553264004195,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042514327829121175,
      "loss": 5.1257,
      "mean_token_accuracy": 0.21613197177648544,
      "num_tokens": 63501086.0,
      "step": 31335
    },
    {
      "entropy": 6.000783681869507,
      "epoch": 2.7387922747531244,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042512006850376724,
      "loss": 5.2611,
      "mean_token_accuracy": 0.2033819004893303,
      "num_tokens": 63512408.0,
      "step": 31340
    },
    {
      "entropy": 6.046214294433594,
      "epoch": 2.739229223105829,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042509685583694116,
      "loss": 5.2399,
      "mean_token_accuracy": 0.20302235186100007,
      "num_tokens": 63521735.0,
      "step": 31345
    },
    {
      "entropy": 5.99093804359436,
      "epoch": 2.7396661714585337,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004250736402911788,
      "loss": 5.2255,
      "mean_token_accuracy": 0.21594954580068587,
      "num_tokens": 63532045.0,
      "step": 31350
    },
    {
      "entropy": 5.947933578491211,
      "epoch": 2.740103119811238,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004250504218669253,
      "loss": 5.2181,
      "mean_token_accuracy": 0.2101554825901985,
      "num_tokens": 63541286.0,
      "step": 31355
    },
    {
      "entropy": 5.97402024269104,
      "epoch": 2.740540068163943,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004250272005646261,
      "loss": 5.2114,
      "mean_token_accuracy": 0.2086111858487129,
      "num_tokens": 63552195.0,
      "step": 31360
    },
    {
      "entropy": 6.01168384552002,
      "epoch": 2.7409770165166476,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004250039763847266,
      "loss": 5.2308,
      "mean_token_accuracy": 0.2107400819659233,
      "num_tokens": 63561318.0,
      "step": 31365
    },
    {
      "entropy": 5.942185306549073,
      "epoch": 2.7414139648693525,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042498074932767227,
      "loss": 5.2666,
      "mean_token_accuracy": 0.2040909692645073,
      "num_tokens": 63571637.0,
      "step": 31370
    },
    {
      "entropy": 6.043250370025635,
      "epoch": 2.7418509132220574,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042495751939390853,
      "loss": 5.3117,
      "mean_token_accuracy": 0.20204173475503923,
      "num_tokens": 63581618.0,
      "step": 31375
    },
    {
      "entropy": 5.9368799209594725,
      "epoch": 2.742287861574762,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042493428658388094,
      "loss": 5.1478,
      "mean_token_accuracy": 0.22075667828321457,
      "num_tokens": 63592129.0,
      "step": 31380
    },
    {
      "entropy": 6.0214841842651365,
      "epoch": 2.7427248099274664,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004249110508980353,
      "loss": 5.3278,
      "mean_token_accuracy": 0.2032761961221695,
      "num_tokens": 63602861.0,
      "step": 31385
    },
    {
      "entropy": 6.0017331600189205,
      "epoch": 2.7431617582801713,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042488781233681705,
      "loss": 5.2273,
      "mean_token_accuracy": 0.20559435784816743,
      "num_tokens": 63612423.0,
      "step": 31390
    },
    {
      "entropy": 5.984505224227905,
      "epoch": 2.743598706632876,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042486457090067206,
      "loss": 5.2885,
      "mean_token_accuracy": 0.20160361528396606,
      "num_tokens": 63622351.0,
      "step": 31395
    },
    {
      "entropy": 5.987055397033691,
      "epoch": 2.7440356549855807,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042484132659004604,
      "loss": 5.2028,
      "mean_token_accuracy": 0.21036819964647294,
      "num_tokens": 63632024.0,
      "step": 31400
    },
    {
      "entropy": 5.98797721862793,
      "epoch": 2.7444726033382856,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004248180794053849,
      "loss": 5.2455,
      "mean_token_accuracy": 0.20055660754442214,
      "num_tokens": 63641962.0,
      "step": 31405
    },
    {
      "entropy": 5.9599169254302975,
      "epoch": 2.74490955169099,
      "grad_norm": 1.25,
      "learning_rate": 0.00042479482934713446,
      "loss": 5.2817,
      "mean_token_accuracy": 0.2037967175245285,
      "num_tokens": 63652095.0,
      "step": 31410
    },
    {
      "entropy": 6.047541999816895,
      "epoch": 2.745346500043695,
      "grad_norm": 1.25,
      "learning_rate": 0.0004247715764157407,
      "loss": 5.3391,
      "mean_token_accuracy": 0.20229894667863846,
      "num_tokens": 63661392.0,
      "step": 31415
    },
    {
      "entropy": 6.016101360321045,
      "epoch": 2.7457834483963994,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042474832061164966,
      "loss": 5.2358,
      "mean_token_accuracy": 0.20241538435220718,
      "num_tokens": 63672383.0,
      "step": 31420
    },
    {
      "entropy": 5.960604667663574,
      "epoch": 2.7462203967491043,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042472506193530734,
      "loss": 5.2676,
      "mean_token_accuracy": 0.20422085076570512,
      "num_tokens": 63682522.0,
      "step": 31425
    },
    {
      "entropy": 5.974162483215332,
      "epoch": 2.746657345101809,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004247018003871599,
      "loss": 5.2259,
      "mean_token_accuracy": 0.20641184449195862,
      "num_tokens": 63692443.0,
      "step": 31430
    },
    {
      "entropy": 5.997109842300415,
      "epoch": 2.7470942934545137,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042467853596765345,
      "loss": 5.2413,
      "mean_token_accuracy": 0.2051355227828026,
      "num_tokens": 63702268.0,
      "step": 31435
    },
    {
      "entropy": 5.915981817245483,
      "epoch": 2.747531241807218,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042465526867723423,
      "loss": 5.1495,
      "mean_token_accuracy": 0.21406208723783493,
      "num_tokens": 63712621.0,
      "step": 31440
    },
    {
      "entropy": 6.021644449234008,
      "epoch": 2.747968190159923,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004246319985163487,
      "loss": 5.3197,
      "mean_token_accuracy": 0.20284022092819215,
      "num_tokens": 63722319.0,
      "step": 31445
    },
    {
      "entropy": 6.014077091217041,
      "epoch": 2.748405138512628,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004246087254854428,
      "loss": 5.3066,
      "mean_token_accuracy": 0.2070568948984146,
      "num_tokens": 63732726.0,
      "step": 31450
    },
    {
      "entropy": 5.955705308914185,
      "epoch": 2.7488420868653325,
      "grad_norm": 1.25,
      "learning_rate": 0.00042458544958496333,
      "loss": 5.1993,
      "mean_token_accuracy": 0.21088216155767442,
      "num_tokens": 63741197.0,
      "step": 31455
    },
    {
      "entropy": 5.986575794219971,
      "epoch": 2.749279035218037,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042456217081535645,
      "loss": 5.2176,
      "mean_token_accuracy": 0.21132999509572983,
      "num_tokens": 63751316.0,
      "step": 31460
    },
    {
      "entropy": 6.050425910949707,
      "epoch": 2.749715983570742,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042453888917706885,
      "loss": 5.2828,
      "mean_token_accuracy": 0.20063281804323196,
      "num_tokens": 63761246.0,
      "step": 31465
    },
    {
      "entropy": 6.017696332931519,
      "epoch": 2.750152931923447,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042451560467054696,
      "loss": 5.3001,
      "mean_token_accuracy": 0.19924434572458266,
      "num_tokens": 63771442.0,
      "step": 31470
    },
    {
      "entropy": 6.013802576065063,
      "epoch": 2.7505898802761513,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042449231729623735,
      "loss": 5.3406,
      "mean_token_accuracy": 0.20260000228881836,
      "num_tokens": 63782840.0,
      "step": 31475
    },
    {
      "entropy": 5.992890119552612,
      "epoch": 2.751026828628856,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042446902705458683,
      "loss": 5.203,
      "mean_token_accuracy": 0.20698740780353547,
      "num_tokens": 63793089.0,
      "step": 31480
    },
    {
      "entropy": 6.026852226257324,
      "epoch": 2.7514637769815606,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004244457339460421,
      "loss": 5.3061,
      "mean_token_accuracy": 0.19583598226308824,
      "num_tokens": 63804125.0,
      "step": 31485
    },
    {
      "entropy": 5.9935304641723635,
      "epoch": 2.7519007253342656,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042442243797104976,
      "loss": 5.2778,
      "mean_token_accuracy": 0.20278451442718506,
      "num_tokens": 63814713.0,
      "step": 31490
    },
    {
      "entropy": 6.051537227630615,
      "epoch": 2.75233767368697,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004243991391300569,
      "loss": 5.3444,
      "mean_token_accuracy": 0.19276710599660873,
      "num_tokens": 63824403.0,
      "step": 31495
    },
    {
      "entropy": 6.0451555252075195,
      "epoch": 2.752774622039675,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004243758374235102,
      "loss": 5.2912,
      "mean_token_accuracy": 0.20274795293807985,
      "num_tokens": 63834182.0,
      "step": 31500
    },
    {
      "entropy": 6.019663953781128,
      "epoch": 2.75321157039238,
      "grad_norm": 1.125,
      "learning_rate": 0.0004243525328518567,
      "loss": 5.2797,
      "mean_token_accuracy": 0.20270183980464934,
      "num_tokens": 63844282.0,
      "step": 31505
    },
    {
      "entropy": 6.010911178588867,
      "epoch": 2.7536485187450843,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004243292254155433,
      "loss": 5.2544,
      "mean_token_accuracy": 0.20316483825445175,
      "num_tokens": 63854898.0,
      "step": 31510
    },
    {
      "entropy": 5.9557201862335205,
      "epoch": 2.754085467097789,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042430591511501717,
      "loss": 5.1824,
      "mean_token_accuracy": 0.21358041167259217,
      "num_tokens": 63864501.0,
      "step": 31515
    },
    {
      "entropy": 5.960829019546509,
      "epoch": 2.7545224154504937,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042428260195072527,
      "loss": 5.2706,
      "mean_token_accuracy": 0.20422741621732712,
      "num_tokens": 63874812.0,
      "step": 31520
    },
    {
      "entropy": 6.013330745697021,
      "epoch": 2.7549593638031986,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004242592859231149,
      "loss": 5.2526,
      "mean_token_accuracy": 0.20506078749895096,
      "num_tokens": 63884028.0,
      "step": 31525
    },
    {
      "entropy": 5.99479455947876,
      "epoch": 2.755396312155903,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042423596703263324,
      "loss": 5.2563,
      "mean_token_accuracy": 0.20639749765396118,
      "num_tokens": 63893991.0,
      "step": 31530
    },
    {
      "entropy": 5.998426246643066,
      "epoch": 2.755833260508608,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004242126452797275,
      "loss": 5.319,
      "mean_token_accuracy": 0.2005082830786705,
      "num_tokens": 63904168.0,
      "step": 31535
    },
    {
      "entropy": 5.979735612869263,
      "epoch": 2.7562702088613125,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042418932066484505,
      "loss": 5.3034,
      "mean_token_accuracy": 0.19507982581853867,
      "num_tokens": 63913881.0,
      "step": 31540
    },
    {
      "entropy": 6.004946660995484,
      "epoch": 2.7567071572140174,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042416599318843324,
      "loss": 5.3609,
      "mean_token_accuracy": 0.19600010067224502,
      "num_tokens": 63924521.0,
      "step": 31545
    },
    {
      "entropy": 6.044146728515625,
      "epoch": 2.757144105566722,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004241426628509395,
      "loss": 5.2988,
      "mean_token_accuracy": 0.19526590257883072,
      "num_tokens": 63935046.0,
      "step": 31550
    },
    {
      "entropy": 6.014778184890747,
      "epoch": 2.7575810539194268,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042411932965281143,
      "loss": 5.2776,
      "mean_token_accuracy": 0.19671535342931748,
      "num_tokens": 63944048.0,
      "step": 31555
    },
    {
      "entropy": 6.018464279174805,
      "epoch": 2.7580180022721317,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004240959935944964,
      "loss": 5.2523,
      "mean_token_accuracy": 0.20676687508821487,
      "num_tokens": 63954943.0,
      "step": 31560
    },
    {
      "entropy": 5.962192630767822,
      "epoch": 2.758454950624836,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004240726546764421,
      "loss": 5.2441,
      "mean_token_accuracy": 0.20456784814596177,
      "num_tokens": 63965929.0,
      "step": 31565
    },
    {
      "entropy": 5.968528699874878,
      "epoch": 2.7588918989775406,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042404931289909614,
      "loss": 5.2889,
      "mean_token_accuracy": 0.19908424019813536,
      "num_tokens": 63976353.0,
      "step": 31570
    },
    {
      "entropy": 6.067388296127319,
      "epoch": 2.7593288473302455,
      "grad_norm": 1.125,
      "learning_rate": 0.00042402596826290634,
      "loss": 5.2833,
      "mean_token_accuracy": 0.20131775885820388,
      "num_tokens": 63987073.0,
      "step": 31575
    },
    {
      "entropy": 6.003280878067017,
      "epoch": 2.7597657956829504,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004240026207683203,
      "loss": 5.2765,
      "mean_token_accuracy": 0.20997366309165955,
      "num_tokens": 63996631.0,
      "step": 31580
    },
    {
      "entropy": 5.957438564300537,
      "epoch": 2.760202744035655,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042397927041578596,
      "loss": 5.2862,
      "mean_token_accuracy": 0.19956034868955613,
      "num_tokens": 64007662.0,
      "step": 31585
    },
    {
      "entropy": 5.973447704315186,
      "epoch": 2.76063969238836,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042395591720575113,
      "loss": 5.2297,
      "mean_token_accuracy": 0.20775575637817384,
      "num_tokens": 64017868.0,
      "step": 31590
    },
    {
      "entropy": 6.072250890731811,
      "epoch": 2.7610766407410643,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042393256113866377,
      "loss": 5.2912,
      "mean_token_accuracy": 0.19187693893909455,
      "num_tokens": 64027464.0,
      "step": 31595
    },
    {
      "entropy": 5.988804483413697,
      "epoch": 2.761513589093769,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004239092022149718,
      "loss": 5.2822,
      "mean_token_accuracy": 0.2046971172094345,
      "num_tokens": 64037795.0,
      "step": 31600
    },
    {
      "entropy": 6.059532642364502,
      "epoch": 2.7619505374464737,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042388584043512333,
      "loss": 5.3111,
      "mean_token_accuracy": 0.19968790411949158,
      "num_tokens": 64048055.0,
      "step": 31605
    },
    {
      "entropy": 6.015092945098877,
      "epoch": 2.7623874857991786,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004238624757995664,
      "loss": 5.1893,
      "mean_token_accuracy": 0.2094716966152191,
      "num_tokens": 64057531.0,
      "step": 31610
    },
    {
      "entropy": 5.9751428127288815,
      "epoch": 2.7628244341518835,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004238391083087493,
      "loss": 5.2478,
      "mean_token_accuracy": 0.20625563263893126,
      "num_tokens": 64066476.0,
      "step": 31615
    },
    {
      "entropy": 5.971633625030518,
      "epoch": 2.763261382504588,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042381573796312,
      "loss": 5.2026,
      "mean_token_accuracy": 0.21987358182668687,
      "num_tokens": 64077069.0,
      "step": 31620
    },
    {
      "entropy": 5.963371992111206,
      "epoch": 2.7636983308572924,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004237923647631269,
      "loss": 5.266,
      "mean_token_accuracy": 0.20768419951200484,
      "num_tokens": 64087325.0,
      "step": 31625
    },
    {
      "entropy": 5.971025609970093,
      "epoch": 2.7641352792099974,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004237689887092183,
      "loss": 5.2672,
      "mean_token_accuracy": 0.20868650078773499,
      "num_tokens": 64097040.0,
      "step": 31630
    },
    {
      "entropy": 5.971757793426514,
      "epoch": 2.7645722275627023,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042374560980184254,
      "loss": 5.2531,
      "mean_token_accuracy": 0.20973075479269027,
      "num_tokens": 64107798.0,
      "step": 31635
    },
    {
      "entropy": 6.045603847503662,
      "epoch": 2.7650091759154067,
      "grad_norm": 1.125,
      "learning_rate": 0.000423722228041448,
      "loss": 5.3056,
      "mean_token_accuracy": 0.20447219610214235,
      "num_tokens": 64118395.0,
      "step": 31640
    },
    {
      "entropy": 6.027225303649902,
      "epoch": 2.7654461242681116,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042369884342848327,
      "loss": 5.3362,
      "mean_token_accuracy": 0.19945759773254396,
      "num_tokens": 64128040.0,
      "step": 31645
    },
    {
      "entropy": 5.9984269618988035,
      "epoch": 2.765883072620816,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042367545596339674,
      "loss": 5.2506,
      "mean_token_accuracy": 0.2074636146426201,
      "num_tokens": 64137660.0,
      "step": 31650
    },
    {
      "entropy": 5.982512283325195,
      "epoch": 2.766320020973521,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042365206564663705,
      "loss": 5.2623,
      "mean_token_accuracy": 0.2075825273990631,
      "num_tokens": 64147632.0,
      "step": 31655
    },
    {
      "entropy": 5.971044063568115,
      "epoch": 2.7667569693262255,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004236286724786529,
      "loss": 5.251,
      "mean_token_accuracy": 0.20625532120466233,
      "num_tokens": 64157604.0,
      "step": 31660
    },
    {
      "entropy": 6.051620960235596,
      "epoch": 2.7671939176789304,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042360527645989287,
      "loss": 5.3307,
      "mean_token_accuracy": 0.19548899233341216,
      "num_tokens": 64167626.0,
      "step": 31665
    },
    {
      "entropy": 6.066037178039551,
      "epoch": 2.7676308660316353,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004235818775908059,
      "loss": 5.309,
      "mean_token_accuracy": 0.20214545875787734,
      "num_tokens": 64178436.0,
      "step": 31670
    },
    {
      "entropy": 5.913223695755005,
      "epoch": 2.76806781438434,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042355847587184056,
      "loss": 5.2104,
      "mean_token_accuracy": 0.21319956183433533,
      "num_tokens": 64188911.0,
      "step": 31675
    },
    {
      "entropy": 5.9525621891021725,
      "epoch": 2.7685047627370443,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042353507130344584,
      "loss": 5.1939,
      "mean_token_accuracy": 0.21535699665546418,
      "num_tokens": 64198539.0,
      "step": 31680
    },
    {
      "entropy": 6.040860366821289,
      "epoch": 2.768941711089749,
      "grad_norm": 1.375,
      "learning_rate": 0.0004235116638860705,
      "loss": 5.2746,
      "mean_token_accuracy": 0.20746220797300338,
      "num_tokens": 64209121.0,
      "step": 31685
    },
    {
      "entropy": 6.064619445800782,
      "epoch": 2.769378659442454,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004234882536201638,
      "loss": 5.2732,
      "mean_token_accuracy": 0.2013103574514389,
      "num_tokens": 64218610.0,
      "step": 31690
    },
    {
      "entropy": 5.958262872695923,
      "epoch": 2.7698156077951586,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004234648405061745,
      "loss": 5.2549,
      "mean_token_accuracy": 0.2045914575457573,
      "num_tokens": 64228626.0,
      "step": 31695
    },
    {
      "entropy": 5.9128941059112545,
      "epoch": 2.7702525561478635,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042344142454455174,
      "loss": 5.2333,
      "mean_token_accuracy": 0.20526354312896727,
      "num_tokens": 64238575.0,
      "step": 31700
    },
    {
      "entropy": 6.002455615997315,
      "epoch": 2.770689504500568,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004234180057357448,
      "loss": 5.276,
      "mean_token_accuracy": 0.19853490293025972,
      "num_tokens": 64248551.0,
      "step": 31705
    },
    {
      "entropy": 6.030704975128174,
      "epoch": 2.771126452853273,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042339458408020267,
      "loss": 5.2694,
      "mean_token_accuracy": 0.20056725591421126,
      "num_tokens": 64259315.0,
      "step": 31710
    },
    {
      "entropy": 6.055716180801392,
      "epoch": 2.7715634012059773,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004233711595783746,
      "loss": 5.2781,
      "mean_token_accuracy": 0.20151465088129045,
      "num_tokens": 64269742.0,
      "step": 31715
    },
    {
      "entropy": 6.039543867111206,
      "epoch": 2.7720003495586822,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00042334773223071,
      "loss": 5.2644,
      "mean_token_accuracy": 0.20601593255996703,
      "num_tokens": 64279266.0,
      "step": 31720
    },
    {
      "entropy": 5.991690683364868,
      "epoch": 2.7724372979113867,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042332430203765816,
      "loss": 5.2567,
      "mean_token_accuracy": 0.20025346726179122,
      "num_tokens": 64289567.0,
      "step": 31725
    },
    {
      "entropy": 5.976130247116089,
      "epoch": 2.7728742462640916,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004233008689996685,
      "loss": 5.3294,
      "mean_token_accuracy": 0.2055324822664261,
      "num_tokens": 64299126.0,
      "step": 31730
    },
    {
      "entropy": 5.961445665359497,
      "epoch": 2.773311194616796,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004232774331171904,
      "loss": 5.2265,
      "mean_token_accuracy": 0.2124423161149025,
      "num_tokens": 64308964.0,
      "step": 31735
    },
    {
      "entropy": 6.066354560852051,
      "epoch": 2.773748142969501,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042325399439067354,
      "loss": 5.2887,
      "mean_token_accuracy": 0.20595939755439757,
      "num_tokens": 64318463.0,
      "step": 31740
    },
    {
      "entropy": 6.0894615173339846,
      "epoch": 2.774185091322206,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004232305528205673,
      "loss": 5.4106,
      "mean_token_accuracy": 0.19378147572278975,
      "num_tokens": 64328841.0,
      "step": 31745
    },
    {
      "entropy": 6.015965843200684,
      "epoch": 2.7746220396749104,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004232071084073214,
      "loss": 5.2737,
      "mean_token_accuracy": 0.21053926944732665,
      "num_tokens": 64339037.0,
      "step": 31750
    },
    {
      "entropy": 6.042277050018311,
      "epoch": 2.775058988027615,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042318366115138553,
      "loss": 5.4028,
      "mean_token_accuracy": 0.2002808064222336,
      "num_tokens": 64350048.0,
      "step": 31755
    },
    {
      "entropy": 6.043776130676269,
      "epoch": 2.7754959363803198,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042316021105320937,
      "loss": 5.3719,
      "mean_token_accuracy": 0.19873762875795364,
      "num_tokens": 64361066.0,
      "step": 31760
    },
    {
      "entropy": 6.071073913574219,
      "epoch": 2.7759328847330247,
      "grad_norm": 1.125,
      "learning_rate": 0.00042313675811324276,
      "loss": 5.3185,
      "mean_token_accuracy": 0.19745370447635652,
      "num_tokens": 64371064.0,
      "step": 31765
    },
    {
      "entropy": 5.993376159667969,
      "epoch": 2.776369833085729,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00042311330233193545,
      "loss": 5.1944,
      "mean_token_accuracy": 0.21687153428792955,
      "num_tokens": 64380820.0,
      "step": 31770
    },
    {
      "entropy": 5.975958919525146,
      "epoch": 2.776806781438434,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004230898437097374,
      "loss": 5.2069,
      "mean_token_accuracy": 0.20462548285722731,
      "num_tokens": 64389777.0,
      "step": 31775
    },
    {
      "entropy": 6.025047731399536,
      "epoch": 2.7772437297911385,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004230663822470986,
      "loss": 5.3499,
      "mean_token_accuracy": 0.19820813983678817,
      "num_tokens": 64399542.0,
      "step": 31780
    },
    {
      "entropy": 6.005260848999024,
      "epoch": 2.7776806781438435,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004230429179444689,
      "loss": 5.2684,
      "mean_token_accuracy": 0.19797811657190323,
      "num_tokens": 64410254.0,
      "step": 31785
    },
    {
      "entropy": 5.968990659713745,
      "epoch": 2.778117626496548,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004230194508022986,
      "loss": 5.2557,
      "mean_token_accuracy": 0.20720195174217224,
      "num_tokens": 64420487.0,
      "step": 31790
    },
    {
      "entropy": 6.011153984069824,
      "epoch": 2.778554574849253,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042299598082103757,
      "loss": 5.2877,
      "mean_token_accuracy": 0.20702100098133086,
      "num_tokens": 64430171.0,
      "step": 31795
    },
    {
      "entropy": 6.039728116989136,
      "epoch": 2.7789915232019577,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042297250800113614,
      "loss": 5.3381,
      "mean_token_accuracy": 0.19551147669553756,
      "num_tokens": 64439960.0,
      "step": 31800
    },
    {
      "entropy": 6.063127851486206,
      "epoch": 2.779428471554662,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004229490323430444,
      "loss": 5.287,
      "mean_token_accuracy": 0.20205519050359727,
      "num_tokens": 64450238.0,
      "step": 31805
    },
    {
      "entropy": 5.9996661186218265,
      "epoch": 2.7798654199073667,
      "grad_norm": 1.25,
      "learning_rate": 0.00042292555384721277,
      "loss": 5.2733,
      "mean_token_accuracy": 0.20418355017900466,
      "num_tokens": 64460137.0,
      "step": 31810
    },
    {
      "entropy": 5.979858160018921,
      "epoch": 2.7803023682600716,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004229020725140915,
      "loss": 5.2188,
      "mean_token_accuracy": 0.21386984139680862,
      "num_tokens": 64470138.0,
      "step": 31815
    },
    {
      "entropy": 5.982955360412598,
      "epoch": 2.7807393166127765,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004228785883441309,
      "loss": 5.3344,
      "mean_token_accuracy": 0.1977784350514412,
      "num_tokens": 64480536.0,
      "step": 31820
    },
    {
      "entropy": 6.040992546081543,
      "epoch": 2.781176264965481,
      "grad_norm": 1.25,
      "learning_rate": 0.0004228551013377816,
      "loss": 5.2409,
      "mean_token_accuracy": 0.21269832998514177,
      "num_tokens": 64489899.0,
      "step": 31825
    },
    {
      "entropy": 5.9800519943237305,
      "epoch": 2.781613213318186,
      "grad_norm": 1.125,
      "learning_rate": 0.00042283161149549387,
      "loss": 5.2114,
      "mean_token_accuracy": 0.21624509990215302,
      "num_tokens": 64500272.0,
      "step": 31830
    },
    {
      "entropy": 6.006553316116333,
      "epoch": 2.7820501616708904,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00042280811881771846,
      "loss": 5.2979,
      "mean_token_accuracy": 0.20425993353128433,
      "num_tokens": 64510507.0,
      "step": 31835
    },
    {
      "entropy": 5.930188083648682,
      "epoch": 2.7824871100235953,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042278462330490584,
      "loss": 5.2028,
      "mean_token_accuracy": 0.21304200142621993,
      "num_tokens": 64521105.0,
      "step": 31840
    },
    {
      "entropy": 5.980663633346557,
      "epoch": 2.7829240583762997,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042276112495750676,
      "loss": 5.2776,
      "mean_token_accuracy": 0.20521688610315322,
      "num_tokens": 64531828.0,
      "step": 31845
    },
    {
      "entropy": 6.005830097198486,
      "epoch": 2.7833610067290047,
      "grad_norm": 1.25,
      "learning_rate": 0.0004227376237759718,
      "loss": 5.1431,
      "mean_token_accuracy": 0.21090665459632874,
      "num_tokens": 64540783.0,
      "step": 31850
    },
    {
      "entropy": 6.013379001617432,
      "epoch": 2.7837979550817096,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004227141197607518,
      "loss": 5.252,
      "mean_token_accuracy": 0.20506582409143448,
      "num_tokens": 64550958.0,
      "step": 31855
    },
    {
      "entropy": 5.957888126373291,
      "epoch": 2.784234903434414,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042269061291229766,
      "loss": 5.2348,
      "mean_token_accuracy": 0.21178335398435594,
      "num_tokens": 64561666.0,
      "step": 31860
    },
    {
      "entropy": 6.040019130706787,
      "epoch": 2.7846718517871185,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042266710323106004,
      "loss": 5.3019,
      "mean_token_accuracy": 0.19544614106416702,
      "num_tokens": 64570908.0,
      "step": 31865
    },
    {
      "entropy": 5.977200651168824,
      "epoch": 2.7851088001398234,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042264359071749017,
      "loss": 5.2117,
      "mean_token_accuracy": 0.20756401717662812,
      "num_tokens": 64580981.0,
      "step": 31870
    },
    {
      "entropy": 6.024312591552734,
      "epoch": 2.7855457484925283,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042262007537203876,
      "loss": 5.2881,
      "mean_token_accuracy": 0.20110594481229782,
      "num_tokens": 64591776.0,
      "step": 31875
    },
    {
      "entropy": 6.02110104560852,
      "epoch": 2.785982696845233,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042259655719515706,
      "loss": 5.2149,
      "mean_token_accuracy": 0.20794848054647447,
      "num_tokens": 64601201.0,
      "step": 31880
    },
    {
      "entropy": 6.02463002204895,
      "epoch": 2.7864196451979377,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004225730361872959,
      "loss": 5.2469,
      "mean_token_accuracy": 0.21201557517051697,
      "num_tokens": 64610998.0,
      "step": 31885
    },
    {
      "entropy": 6.07282395362854,
      "epoch": 2.786856593550642,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004225495123489067,
      "loss": 5.364,
      "mean_token_accuracy": 0.19349345415830613,
      "num_tokens": 64620503.0,
      "step": 31890
    },
    {
      "entropy": 6.007886981964111,
      "epoch": 2.787293541903347,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004225259856804405,
      "loss": 5.2157,
      "mean_token_accuracy": 0.2088899075984955,
      "num_tokens": 64630172.0,
      "step": 31895
    },
    {
      "entropy": 5.980282545089722,
      "epoch": 2.7877304902560516,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004225024561823486,
      "loss": 5.2336,
      "mean_token_accuracy": 0.20611557215452195,
      "num_tokens": 64639737.0,
      "step": 31900
    },
    {
      "entropy": 5.974046802520752,
      "epoch": 2.7881674386087565,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004224789238550823,
      "loss": 5.1432,
      "mean_token_accuracy": 0.2147744581103325,
      "num_tokens": 64650324.0,
      "step": 31905
    },
    {
      "entropy": 6.0540618896484375,
      "epoch": 2.7886043869614614,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004224553886990929,
      "loss": 5.3419,
      "mean_token_accuracy": 0.19805721193552017,
      "num_tokens": 64661938.0,
      "step": 31910
    },
    {
      "entropy": 6.008874225616455,
      "epoch": 2.789041335314166,
      "grad_norm": 1.25,
      "learning_rate": 0.00042243185071483193,
      "loss": 5.2164,
      "mean_token_accuracy": 0.20996900647878647,
      "num_tokens": 64672042.0,
      "step": 31915
    },
    {
      "entropy": 5.960060882568359,
      "epoch": 2.7894782836668703,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042240830990275085,
      "loss": 5.2676,
      "mean_token_accuracy": 0.2052510127425194,
      "num_tokens": 64681798.0,
      "step": 31920
    },
    {
      "entropy": 6.007536268234253,
      "epoch": 2.7899152320195753,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000422384766263301,
      "loss": 5.2952,
      "mean_token_accuracy": 0.2040834292769432,
      "num_tokens": 64692329.0,
      "step": 31925
    },
    {
      "entropy": 6.037142705917359,
      "epoch": 2.79035218037228,
      "grad_norm": 1.25,
      "learning_rate": 0.00042236121979693424,
      "loss": 5.2451,
      "mean_token_accuracy": 0.20841434895992278,
      "num_tokens": 64701605.0,
      "step": 31930
    },
    {
      "entropy": 6.0667518138885494,
      "epoch": 2.7907891287249846,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042233767050410206,
      "loss": 5.3324,
      "mean_token_accuracy": 0.20343460887670517,
      "num_tokens": 64711649.0,
      "step": 31935
    },
    {
      "entropy": 5.98931131362915,
      "epoch": 2.7912260770776895,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004223141183852562,
      "loss": 5.2778,
      "mean_token_accuracy": 0.2008076310157776,
      "num_tokens": 64721451.0,
      "step": 31940
    },
    {
      "entropy": 6.012391376495361,
      "epoch": 2.791663025430394,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004222905634408482,
      "loss": 5.2422,
      "mean_token_accuracy": 0.20509547293186187,
      "num_tokens": 64731725.0,
      "step": 31945
    },
    {
      "entropy": 6.014767742156982,
      "epoch": 2.792099973783099,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042226700567133015,
      "loss": 5.1682,
      "mean_token_accuracy": 0.21114119589328767,
      "num_tokens": 64741731.0,
      "step": 31950
    },
    {
      "entropy": 5.974693250656128,
      "epoch": 2.7925369221358034,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004222434450771537,
      "loss": 5.305,
      "mean_token_accuracy": 0.20027866810560227,
      "num_tokens": 64753324.0,
      "step": 31955
    },
    {
      "entropy": 5.942831182479859,
      "epoch": 2.7929738704885083,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042221988165877083,
      "loss": 5.2171,
      "mean_token_accuracy": 0.21054156124591827,
      "num_tokens": 64763209.0,
      "step": 31960
    },
    {
      "entropy": 6.012267589569092,
      "epoch": 2.7934108188412132,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004221963154166335,
      "loss": 5.2772,
      "mean_token_accuracy": 0.20321819484233855,
      "num_tokens": 64773867.0,
      "step": 31965
    },
    {
      "entropy": 6.035675048828125,
      "epoch": 2.7938477671939177,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004221727463511937,
      "loss": 5.2461,
      "mean_token_accuracy": 0.20100697726011277,
      "num_tokens": 64785242.0,
      "step": 31970
    },
    {
      "entropy": 5.917937517166138,
      "epoch": 2.794284715546622,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004221491744629035,
      "loss": 5.2248,
      "mean_token_accuracy": 0.20323044955730438,
      "num_tokens": 64795852.0,
      "step": 31975
    },
    {
      "entropy": 5.9965616226196286,
      "epoch": 2.794721663899327,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042212559975221504,
      "loss": 5.1564,
      "mean_token_accuracy": 0.21911224722862244,
      "num_tokens": 64805982.0,
      "step": 31980
    },
    {
      "entropy": 6.000169277191162,
      "epoch": 2.795158612252032,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042210202221958055,
      "loss": 5.2795,
      "mean_token_accuracy": 0.2061985045671463,
      "num_tokens": 64816997.0,
      "step": 31985
    },
    {
      "entropy": 5.991363954544068,
      "epoch": 2.7955955606047365,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004220784418654521,
      "loss": 5.2436,
      "mean_token_accuracy": 0.19767025411128997,
      "num_tokens": 64826597.0,
      "step": 31990
    },
    {
      "entropy": 5.960865640640259,
      "epoch": 2.7960325089574414,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042205485869028207,
      "loss": 5.2239,
      "mean_token_accuracy": 0.20726161301136017,
      "num_tokens": 64835917.0,
      "step": 31995
    },
    {
      "entropy": 6.003108024597168,
      "epoch": 2.796469457310146,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00042203127269452284,
      "loss": 5.2268,
      "mean_token_accuracy": 0.20458547323942183,
      "num_tokens": 64846889.0,
      "step": 32000
    },
    {
      "entropy": 6.039382362365723,
      "epoch": 2.7969064056628508,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042200768387862676,
      "loss": 5.331,
      "mean_token_accuracy": 0.20057173520326615,
      "num_tokens": 64857321.0,
      "step": 32005
    },
    {
      "entropy": 5.963539409637451,
      "epoch": 2.7973433540155552,
      "grad_norm": 1.25,
      "learning_rate": 0.00042198409224304624,
      "loss": 5.2836,
      "mean_token_accuracy": 0.20276326537132264,
      "num_tokens": 64867255.0,
      "step": 32010
    },
    {
      "entropy": 6.004115581512451,
      "epoch": 2.79778030236826,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004219604977882338,
      "loss": 5.2467,
      "mean_token_accuracy": 0.20401121973991393,
      "num_tokens": 64876179.0,
      "step": 32015
    },
    {
      "entropy": 6.068100881576538,
      "epoch": 2.798217250720965,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042193690051464206,
      "loss": 5.3458,
      "mean_token_accuracy": 0.20111594796180726,
      "num_tokens": 64886387.0,
      "step": 32020
    },
    {
      "entropy": 6.023219728469849,
      "epoch": 2.7986541990736695,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042191330042272353,
      "loss": 5.2843,
      "mean_token_accuracy": 0.20720139741897584,
      "num_tokens": 64896816.0,
      "step": 32025
    },
    {
      "entropy": 5.938279867172241,
      "epoch": 2.799091147426374,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042188969751293093,
      "loss": 5.2295,
      "mean_token_accuracy": 0.21136112362146378,
      "num_tokens": 64907747.0,
      "step": 32030
    },
    {
      "entropy": 5.96572265625,
      "epoch": 2.799528095779079,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042186609178571696,
      "loss": 5.2749,
      "mean_token_accuracy": 0.20042705833911895,
      "num_tokens": 64918066.0,
      "step": 32035
    },
    {
      "entropy": 6.047101068496704,
      "epoch": 2.799965044131784,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004218424832415344,
      "loss": 5.3316,
      "mean_token_accuracy": 0.20124776363372804,
      "num_tokens": 64927021.0,
      "step": 32040
    },
    {
      "entropy": 5.9884013652801515,
      "epoch": 2.8004019924844883,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042181887188083605,
      "loss": 5.1292,
      "mean_token_accuracy": 0.21337211281061172,
      "num_tokens": 64937477.0,
      "step": 32045
    },
    {
      "entropy": 5.935149812698365,
      "epoch": 2.800838940837193,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042179525770407486,
      "loss": 5.2332,
      "mean_token_accuracy": 0.20775631219148635,
      "num_tokens": 64947178.0,
      "step": 32050
    },
    {
      "entropy": 5.985747528076172,
      "epoch": 2.8012758891898977,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042177164071170373,
      "loss": 5.3118,
      "mean_token_accuracy": 0.198447285592556,
      "num_tokens": 64956921.0,
      "step": 32055
    },
    {
      "entropy": 5.989143562316895,
      "epoch": 2.8017128375426026,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004217480209041755,
      "loss": 5.2084,
      "mean_token_accuracy": 0.2041856124997139,
      "num_tokens": 64966688.0,
      "step": 32060
    },
    {
      "entropy": 6.0230231285095215,
      "epoch": 2.802149785895307,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004217243982819434,
      "loss": 5.2102,
      "mean_token_accuracy": 0.2088591620326042,
      "num_tokens": 64975577.0,
      "step": 32065
    },
    {
      "entropy": 5.97547492980957,
      "epoch": 2.802586734248012,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004217007728454605,
      "loss": 5.2828,
      "mean_token_accuracy": 0.20180410742759705,
      "num_tokens": 64986121.0,
      "step": 32070
    },
    {
      "entropy": 6.065494394302368,
      "epoch": 2.8030236826007164,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004216771445951798,
      "loss": 5.2969,
      "mean_token_accuracy": 0.19720084518194197,
      "num_tokens": 64996623.0,
      "step": 32075
    },
    {
      "entropy": 5.990990495681762,
      "epoch": 2.8034606309534214,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042165351353155476,
      "loss": 5.2302,
      "mean_token_accuracy": 0.19989530742168427,
      "num_tokens": 65006756.0,
      "step": 32080
    },
    {
      "entropy": 6.014942979812622,
      "epoch": 2.803897579306126,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042162987965503834,
      "loss": 5.2773,
      "mean_token_accuracy": 0.20316673517227174,
      "num_tokens": 65017453.0,
      "step": 32085
    },
    {
      "entropy": 6.062999820709228,
      "epoch": 2.8043345276588307,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042160624296608403,
      "loss": 5.2979,
      "mean_token_accuracy": 0.19418917447328568,
      "num_tokens": 65028090.0,
      "step": 32090
    },
    {
      "entropy": 6.032772827148437,
      "epoch": 2.8047714760115356,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004215826034651452,
      "loss": 5.3865,
      "mean_token_accuracy": 0.19277853816747664,
      "num_tokens": 65038422.0,
      "step": 32095
    },
    {
      "entropy": 5.988374805450439,
      "epoch": 2.80520842436424,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004215589611526752,
      "loss": 5.1712,
      "mean_token_accuracy": 0.2132576361298561,
      "num_tokens": 65048244.0,
      "step": 32100
    },
    {
      "entropy": 6.014710378646851,
      "epoch": 2.8056453727169446,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004215353160291275,
      "loss": 5.3095,
      "mean_token_accuracy": 0.20494887381792068,
      "num_tokens": 65058411.0,
      "step": 32105
    },
    {
      "entropy": 5.955136823654175,
      "epoch": 2.8060823210696495,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004215116680949558,
      "loss": 5.2276,
      "mean_token_accuracy": 0.2060987502336502,
      "num_tokens": 65068148.0,
      "step": 32110
    },
    {
      "entropy": 6.034640693664551,
      "epoch": 2.8065192694223544,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004214880173506133,
      "loss": 5.2608,
      "mean_token_accuracy": 0.1990837886929512,
      "num_tokens": 65078212.0,
      "step": 32115
    },
    {
      "entropy": 5.9960081577301025,
      "epoch": 2.806956217775059,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042146436379655405,
      "loss": 5.2385,
      "mean_token_accuracy": 0.21152850091457367,
      "num_tokens": 65088133.0,
      "step": 32120
    },
    {
      "entropy": 6.003298807144165,
      "epoch": 2.807393166127764,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042144070743323155,
      "loss": 5.2304,
      "mean_token_accuracy": 0.20046185404062272,
      "num_tokens": 65097671.0,
      "step": 32125
    },
    {
      "entropy": 6.0241865634918215,
      "epoch": 2.8078301144804683,
      "grad_norm": 1.25,
      "learning_rate": 0.0004214170482610995,
      "loss": 5.3563,
      "mean_token_accuracy": 0.19464507400989534,
      "num_tokens": 65107833.0,
      "step": 32130
    },
    {
      "entropy": 5.99145917892456,
      "epoch": 2.808267062833173,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042139338628061177,
      "loss": 5.224,
      "mean_token_accuracy": 0.20714160948991775,
      "num_tokens": 65118414.0,
      "step": 32135
    },
    {
      "entropy": 6.048810911178589,
      "epoch": 2.8087040111858776,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042136972149222203,
      "loss": 5.3609,
      "mean_token_accuracy": 0.19523254930973052,
      "num_tokens": 65128920.0,
      "step": 32140
    },
    {
      "entropy": 5.974699687957764,
      "epoch": 2.8091409595385826,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004213460538963845,
      "loss": 5.2187,
      "mean_token_accuracy": 0.21100672334432602,
      "num_tokens": 65138869.0,
      "step": 32145
    },
    {
      "entropy": 6.010856628417969,
      "epoch": 2.8095779078912875,
      "grad_norm": 1.125,
      "learning_rate": 0.0004213223834935529,
      "loss": 5.2799,
      "mean_token_accuracy": 0.20579004883766175,
      "num_tokens": 65148737.0,
      "step": 32150
    },
    {
      "entropy": 6.039405441284179,
      "epoch": 2.810014856243992,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042129871028418137,
      "loss": 5.324,
      "mean_token_accuracy": 0.19602653980255128,
      "num_tokens": 65158075.0,
      "step": 32155
    },
    {
      "entropy": 6.037004232406616,
      "epoch": 2.8104518045966964,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004212750342687239,
      "loss": 5.3472,
      "mean_token_accuracy": 0.1938310295343399,
      "num_tokens": 65167788.0,
      "step": 32160
    },
    {
      "entropy": 5.989744853973389,
      "epoch": 2.8108887529494013,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004212513554476346,
      "loss": 5.2839,
      "mean_token_accuracy": 0.20737276822328568,
      "num_tokens": 65177581.0,
      "step": 32165
    },
    {
      "entropy": 6.047765636444092,
      "epoch": 2.8113257013021062,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042122767382136766,
      "loss": 5.246,
      "mean_token_accuracy": 0.21596966236829757,
      "num_tokens": 65186546.0,
      "step": 32170
    },
    {
      "entropy": 5.9174055576324465,
      "epoch": 2.8117626496548107,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004212039893903774,
      "loss": 5.174,
      "mean_token_accuracy": 0.21625203043222427,
      "num_tokens": 65195717.0,
      "step": 32175
    },
    {
      "entropy": 5.991904163360596,
      "epoch": 2.8121995980075156,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042118030215511793,
      "loss": 5.2955,
      "mean_token_accuracy": 0.2020087018609047,
      "num_tokens": 65206090.0,
      "step": 32180
    },
    {
      "entropy": 5.99215612411499,
      "epoch": 2.81263654636022,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042115661211604373,
      "loss": 5.3145,
      "mean_token_accuracy": 0.20411340296268463,
      "num_tokens": 65217194.0,
      "step": 32185
    },
    {
      "entropy": 6.120053434371949,
      "epoch": 2.813073494712925,
      "grad_norm": 1.234375,
      "learning_rate": 0.000421132919273609,
      "loss": 5.3235,
      "mean_token_accuracy": 0.20236674398183824,
      "num_tokens": 65228906.0,
      "step": 32190
    },
    {
      "entropy": 6.060681533813477,
      "epoch": 2.8135104430656295,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004211092236282685,
      "loss": 5.3035,
      "mean_token_accuracy": 0.20163415223360062,
      "num_tokens": 65238606.0,
      "step": 32195
    },
    {
      "entropy": 5.9846351623535154,
      "epoch": 2.8139473914183344,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042108552518047644,
      "loss": 5.2311,
      "mean_token_accuracy": 0.2078977882862091,
      "num_tokens": 65248259.0,
      "step": 32200
    },
    {
      "entropy": 6.022395181655884,
      "epoch": 2.8143843397710393,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004210618239306874,
      "loss": 5.3049,
      "mean_token_accuracy": 0.197019562125206,
      "num_tokens": 65259294.0,
      "step": 32205
    },
    {
      "entropy": 6.125720357894897,
      "epoch": 2.8148212881237438,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004210381198793561,
      "loss": 5.3202,
      "mean_token_accuracy": 0.19847084283828736,
      "num_tokens": 65269498.0,
      "step": 32210
    },
    {
      "entropy": 6.076746034622192,
      "epoch": 2.8152582364764482,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004210144130269371,
      "loss": 5.3619,
      "mean_token_accuracy": 0.1928400218486786,
      "num_tokens": 65280027.0,
      "step": 32215
    },
    {
      "entropy": 5.987857532501221,
      "epoch": 2.815695184829153,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042099070337388517,
      "loss": 5.269,
      "mean_token_accuracy": 0.20357988625764847,
      "num_tokens": 65290451.0,
      "step": 32220
    },
    {
      "entropy": 5.9794364929199215,
      "epoch": 2.816132133181858,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042096699092065504,
      "loss": 5.2357,
      "mean_token_accuracy": 0.20651257485151292,
      "num_tokens": 65299906.0,
      "step": 32225
    },
    {
      "entropy": 5.9994007587432865,
      "epoch": 2.8165690815345625,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004209432756677015,
      "loss": 5.2579,
      "mean_token_accuracy": 0.20595213025808334,
      "num_tokens": 65310197.0,
      "step": 32230
    },
    {
      "entropy": 6.001691913604736,
      "epoch": 2.8170060298872674,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042091955761547963,
      "loss": 5.3109,
      "mean_token_accuracy": 0.20133645236492156,
      "num_tokens": 65320779.0,
      "step": 32235
    },
    {
      "entropy": 5.927051544189453,
      "epoch": 2.817442978239972,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000420895836764444,
      "loss": 5.1911,
      "mean_token_accuracy": 0.20895151048898697,
      "num_tokens": 65330578.0,
      "step": 32240
    },
    {
      "entropy": 5.990714693069458,
      "epoch": 2.817879926592677,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042087211311504986,
      "loss": 5.2868,
      "mean_token_accuracy": 0.20089613944292067,
      "num_tokens": 65341202.0,
      "step": 32245
    },
    {
      "entropy": 6.022494077682495,
      "epoch": 2.8183168749453813,
      "grad_norm": 1.25,
      "learning_rate": 0.0004208483866677521,
      "loss": 5.242,
      "mean_token_accuracy": 0.21001620292663575,
      "num_tokens": 65352584.0,
      "step": 32250
    },
    {
      "entropy": 6.039282655715942,
      "epoch": 2.818753823298086,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004208246574230059,
      "loss": 5.2279,
      "mean_token_accuracy": 0.20663381069898606,
      "num_tokens": 65363191.0,
      "step": 32255
    },
    {
      "entropy": 5.960844564437866,
      "epoch": 2.819190771650791,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042080092538126633,
      "loss": 5.2471,
      "mean_token_accuracy": 0.20391215533018112,
      "num_tokens": 65373880.0,
      "step": 32260
    },
    {
      "entropy": 5.994707775115967,
      "epoch": 2.8196277200034956,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042077719054298867,
      "loss": 5.3245,
      "mean_token_accuracy": 0.20249886959791183,
      "num_tokens": 65383149.0,
      "step": 32265
    },
    {
      "entropy": 6.0227148056030275,
      "epoch": 2.8200646683562,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004207534529086281,
      "loss": 5.2202,
      "mean_token_accuracy": 0.21146385371685028,
      "num_tokens": 65393005.0,
      "step": 32270
    },
    {
      "entropy": 6.037353897094727,
      "epoch": 2.820501616708905,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042072971247863984,
      "loss": 5.2819,
      "mean_token_accuracy": 0.20052288174629213,
      "num_tokens": 65402640.0,
      "step": 32275
    },
    {
      "entropy": 6.041448545455933,
      "epoch": 2.82093856506161,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004207059692534794,
      "loss": 5.3143,
      "mean_token_accuracy": 0.19343896210193634,
      "num_tokens": 65412863.0,
      "step": 32280
    },
    {
      "entropy": 5.940723848342896,
      "epoch": 2.8213755134143144,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042068222323360214,
      "loss": 5.3191,
      "mean_token_accuracy": 0.19922893047332763,
      "num_tokens": 65422581.0,
      "step": 32285
    },
    {
      "entropy": 6.005614185333252,
      "epoch": 2.8218124617670193,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042065847441946354,
      "loss": 5.271,
      "mean_token_accuracy": 0.20834242254495622,
      "num_tokens": 65433604.0,
      "step": 32290
    },
    {
      "entropy": 5.979412937164307,
      "epoch": 2.8222494101197237,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000420634722811519,
      "loss": 5.2084,
      "mean_token_accuracy": 0.2065222829580307,
      "num_tokens": 65442956.0,
      "step": 32295
    },
    {
      "entropy": 6.037739562988281,
      "epoch": 2.8226863584724287,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042061096841022416,
      "loss": 5.2638,
      "mean_token_accuracy": 0.20008471608161926,
      "num_tokens": 65452441.0,
      "step": 32300
    },
    {
      "entropy": 6.0245222568511965,
      "epoch": 2.823123306825133,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042058721121603467,
      "loss": 5.2813,
      "mean_token_accuracy": 0.21123829036951064,
      "num_tokens": 65463628.0,
      "step": 32305
    },
    {
      "entropy": 5.9847252368927,
      "epoch": 2.823560255177838,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004205634512294062,
      "loss": 5.2503,
      "mean_token_accuracy": 0.2087850645184517,
      "num_tokens": 65472771.0,
      "step": 32310
    },
    {
      "entropy": 6.031204319000244,
      "epoch": 2.823997203530543,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042053968845079436,
      "loss": 5.3015,
      "mean_token_accuracy": 0.20153152495622634,
      "num_tokens": 65482678.0,
      "step": 32315
    },
    {
      "entropy": 6.082346487045288,
      "epoch": 2.8244341518832474,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042051592288065514,
      "loss": 5.2977,
      "mean_token_accuracy": 0.19859569817781447,
      "num_tokens": 65493127.0,
      "step": 32320
    },
    {
      "entropy": 6.004867458343506,
      "epoch": 2.824871100235952,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004204921545194443,
      "loss": 5.3173,
      "mean_token_accuracy": 0.1999363750219345,
      "num_tokens": 65502508.0,
      "step": 32325
    },
    {
      "entropy": 6.020262384414673,
      "epoch": 2.825308048588657,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004204683833676176,
      "loss": 5.3336,
      "mean_token_accuracy": 0.19641396999359131,
      "num_tokens": 65512336.0,
      "step": 32330
    },
    {
      "entropy": 6.037469720840454,
      "epoch": 2.8257449969413617,
      "grad_norm": 1.25,
      "learning_rate": 0.0004204446094256312,
      "loss": 5.2521,
      "mean_token_accuracy": 0.19927924424409865,
      "num_tokens": 65523139.0,
      "step": 32335
    },
    {
      "entropy": 5.9716041564941404,
      "epoch": 2.826181945294066,
      "grad_norm": 1.25,
      "learning_rate": 0.00042042083269394083,
      "loss": 5.2451,
      "mean_token_accuracy": 0.20951423048973083,
      "num_tokens": 65532846.0,
      "step": 32340
    },
    {
      "entropy": 6.017780303955078,
      "epoch": 2.826618893646771,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004203970531730028,
      "loss": 5.2393,
      "mean_token_accuracy": 0.20243889540433885,
      "num_tokens": 65544493.0,
      "step": 32345
    },
    {
      "entropy": 6.05112099647522,
      "epoch": 2.8270558419994756,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042037327086327304,
      "loss": 5.3525,
      "mean_token_accuracy": 0.20077478140592575,
      "num_tokens": 65554183.0,
      "step": 32350
    },
    {
      "entropy": 6.0156549453735355,
      "epoch": 2.8274927903521805,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004203494857652078,
      "loss": 5.2451,
      "mean_token_accuracy": 0.2062898874282837,
      "num_tokens": 65564068.0,
      "step": 32355
    },
    {
      "entropy": 5.97075572013855,
      "epoch": 2.827929738704885,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042032569787926326,
      "loss": 5.2026,
      "mean_token_accuracy": 0.21085550785064697,
      "num_tokens": 65574272.0,
      "step": 32360
    },
    {
      "entropy": 5.958322191238404,
      "epoch": 2.82836668705759,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042030190720589573,
      "loss": 5.2536,
      "mean_token_accuracy": 0.20736325532197952,
      "num_tokens": 65584890.0,
      "step": 32365
    },
    {
      "entropy": 6.04754958152771,
      "epoch": 2.8288036354102943,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004202781137455614,
      "loss": 5.2954,
      "mean_token_accuracy": 0.20000335127115249,
      "num_tokens": 65595776.0,
      "step": 32370
    },
    {
      "entropy": 5.973357057571411,
      "epoch": 2.8292405837629993,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004202543174987168,
      "loss": 5.2481,
      "mean_token_accuracy": 0.2077904984354973,
      "num_tokens": 65606234.0,
      "step": 32375
    },
    {
      "entropy": 6.069826602935791,
      "epoch": 2.8296775321157037,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042023051846581824,
      "loss": 5.391,
      "mean_token_accuracy": 0.19796949923038482,
      "num_tokens": 65616576.0,
      "step": 32380
    },
    {
      "entropy": 6.00982027053833,
      "epoch": 2.8301144804684086,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004202067166473222,
      "loss": 5.204,
      "mean_token_accuracy": 0.2124270111322403,
      "num_tokens": 65626792.0,
      "step": 32385
    },
    {
      "entropy": 6.0256185054779055,
      "epoch": 2.8305514288211135,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004201829120436854,
      "loss": 5.233,
      "mean_token_accuracy": 0.2047160565853119,
      "num_tokens": 65637041.0,
      "step": 32390
    },
    {
      "entropy": 5.980459213256836,
      "epoch": 2.830988377173818,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004201591046553641,
      "loss": 5.3266,
      "mean_token_accuracy": 0.19878651350736617,
      "num_tokens": 65647132.0,
      "step": 32395
    },
    {
      "entropy": 6.070791006088257,
      "epoch": 2.831425325526523,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042013529448281517,
      "loss": 5.3325,
      "mean_token_accuracy": 0.199150687456131,
      "num_tokens": 65657372.0,
      "step": 32400
    },
    {
      "entropy": 6.028166389465332,
      "epoch": 2.8318622738792274,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004201114815264952,
      "loss": 5.2756,
      "mean_token_accuracy": 0.20533011555671693,
      "num_tokens": 65667118.0,
      "step": 32405
    },
    {
      "entropy": 6.027500963211059,
      "epoch": 2.8322992222319323,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000420087665786861,
      "loss": 5.3106,
      "mean_token_accuracy": 0.20399150401353836,
      "num_tokens": 65677830.0,
      "step": 32410
    },
    {
      "entropy": 5.929320383071899,
      "epoch": 2.832736170584637,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042006384726436943,
      "loss": 5.261,
      "mean_token_accuracy": 0.2088939964771271,
      "num_tokens": 65687238.0,
      "step": 32415
    },
    {
      "entropy": 5.967250776290894,
      "epoch": 2.8331731189373417,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042004002595947713,
      "loss": 5.2253,
      "mean_token_accuracy": 0.21675384938716888,
      "num_tokens": 65697547.0,
      "step": 32420
    },
    {
      "entropy": 6.016834259033203,
      "epoch": 2.833610067290046,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004200162018726412,
      "loss": 5.3655,
      "mean_token_accuracy": 0.20081379860639573,
      "num_tokens": 65709272.0,
      "step": 32425
    },
    {
      "entropy": 6.001433420181274,
      "epoch": 2.834047015642751,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004199923750043185,
      "loss": 5.3076,
      "mean_token_accuracy": 0.20302882194519042,
      "num_tokens": 65720986.0,
      "step": 32430
    },
    {
      "entropy": 6.080328750610351,
      "epoch": 2.8344839639954555,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041996854535496616,
      "loss": 5.3119,
      "mean_token_accuracy": 0.1997762456536293,
      "num_tokens": 65731339.0,
      "step": 32435
    },
    {
      "entropy": 6.065917539596557,
      "epoch": 2.8349209123481605,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004199447129250411,
      "loss": 5.2924,
      "mean_token_accuracy": 0.2038639098405838,
      "num_tokens": 65741826.0,
      "step": 32440
    },
    {
      "entropy": 5.998392724990845,
      "epoch": 2.8353578607008654,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004199208777150004,
      "loss": 5.321,
      "mean_token_accuracy": 0.19794889986515046,
      "num_tokens": 65752925.0,
      "step": 32445
    },
    {
      "entropy": 6.0477453708648685,
      "epoch": 2.83579480905357,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004198970397253015,
      "loss": 5.2963,
      "mean_token_accuracy": 0.20113494843244553,
      "num_tokens": 65763948.0,
      "step": 32450
    },
    {
      "entropy": 6.033010387420655,
      "epoch": 2.8362317574062743,
      "grad_norm": 1.125,
      "learning_rate": 0.0004198731989564014,
      "loss": 5.2805,
      "mean_token_accuracy": 0.2036979988217354,
      "num_tokens": 65774086.0,
      "step": 32455
    },
    {
      "entropy": 5.962444591522217,
      "epoch": 2.8366687057589792,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004198493554087574,
      "loss": 5.1746,
      "mean_token_accuracy": 0.21264494508504866,
      "num_tokens": 65783630.0,
      "step": 32460
    },
    {
      "entropy": 6.005564403533936,
      "epoch": 2.837105654111684,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041982550908282686,
      "loss": 5.2788,
      "mean_token_accuracy": 0.21095500737428666,
      "num_tokens": 65793903.0,
      "step": 32465
    },
    {
      "entropy": 6.001746654510498,
      "epoch": 2.8375426024643886,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041980165997906714,
      "loss": 5.2112,
      "mean_token_accuracy": 0.2101442039012909,
      "num_tokens": 65803483.0,
      "step": 32470
    },
    {
      "entropy": 6.024435138702392,
      "epoch": 2.8379795508170935,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004197778080979358,
      "loss": 5.2854,
      "mean_token_accuracy": 0.2071791931986809,
      "num_tokens": 65814336.0,
      "step": 32475
    },
    {
      "entropy": 6.036310052871704,
      "epoch": 2.838416499169798,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041975395343989017,
      "loss": 5.3096,
      "mean_token_accuracy": 0.2049396365880966,
      "num_tokens": 65825525.0,
      "step": 32480
    },
    {
      "entropy": 6.000756359100341,
      "epoch": 2.838853447522503,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004197300960053879,
      "loss": 5.2468,
      "mean_token_accuracy": 0.2107456922531128,
      "num_tokens": 65836276.0,
      "step": 32485
    },
    {
      "entropy": 6.024409246444702,
      "epoch": 2.8392903958752074,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004197062357948865,
      "loss": 5.3166,
      "mean_token_accuracy": 0.19813595563173295,
      "num_tokens": 65846164.0,
      "step": 32490
    },
    {
      "entropy": 6.040357398986816,
      "epoch": 2.8397273442279123,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041968237280884364,
      "loss": 5.2796,
      "mean_token_accuracy": 0.2043122574687004,
      "num_tokens": 65855759.0,
      "step": 32495
    },
    {
      "entropy": 5.945798301696778,
      "epoch": 2.840164292580617,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004196585070477172,
      "loss": 5.139,
      "mean_token_accuracy": 0.21811979860067368,
      "num_tokens": 65866009.0,
      "step": 32500
    },
    {
      "entropy": 6.0223864078521725,
      "epoch": 2.8406012409333217,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041963463851196466,
      "loss": 5.2273,
      "mean_token_accuracy": 0.20453447848558426,
      "num_tokens": 65875707.0,
      "step": 32505
    },
    {
      "entropy": 6.02051100730896,
      "epoch": 2.841038189286026,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041961076720204403,
      "loss": 5.2552,
      "mean_token_accuracy": 0.21158922761678695,
      "num_tokens": 65884396.0,
      "step": 32510
    },
    {
      "entropy": 6.013979482650757,
      "epoch": 2.841475137638731,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004195868931184131,
      "loss": 5.2929,
      "mean_token_accuracy": 0.20098719447851182,
      "num_tokens": 65894278.0,
      "step": 32515
    },
    {
      "entropy": 5.983173370361328,
      "epoch": 2.841912085991436,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041956301626152974,
      "loss": 5.2762,
      "mean_token_accuracy": 0.2047438681125641,
      "num_tokens": 65903863.0,
      "step": 32520
    },
    {
      "entropy": 5.958584833145141,
      "epoch": 2.8423490343441404,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041953913663185207,
      "loss": 5.1524,
      "mean_token_accuracy": 0.21144578754901885,
      "num_tokens": 65913996.0,
      "step": 32525
    },
    {
      "entropy": 5.940230798721314,
      "epoch": 2.8427859826968453,
      "grad_norm": 1.375,
      "learning_rate": 0.000419515254229838,
      "loss": 5.1949,
      "mean_token_accuracy": 0.2149798423051834,
      "num_tokens": 65923457.0,
      "step": 32530
    },
    {
      "entropy": 6.053950214385987,
      "epoch": 2.84322293104955,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041949136905594555,
      "loss": 5.2587,
      "mean_token_accuracy": 0.2014596089720726,
      "num_tokens": 65933132.0,
      "step": 32535
    },
    {
      "entropy": 6.009577798843384,
      "epoch": 2.8436598794022547,
      "grad_norm": 1.296875,
      "learning_rate": 0.000419467481110633,
      "loss": 5.2453,
      "mean_token_accuracy": 0.21184816658496858,
      "num_tokens": 65942784.0,
      "step": 32540
    },
    {
      "entropy": 5.997848129272461,
      "epoch": 2.844096827754959,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004194435903943585,
      "loss": 5.2167,
      "mean_token_accuracy": 0.2107362776994705,
      "num_tokens": 65952734.0,
      "step": 32545
    },
    {
      "entropy": 5.939011001586914,
      "epoch": 2.844533776107664,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041941969690758023,
      "loss": 5.1127,
      "mean_token_accuracy": 0.22260621190071106,
      "num_tokens": 65962732.0,
      "step": 32550
    },
    {
      "entropy": 6.010365200042725,
      "epoch": 2.844970724460369,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041939580065075645,
      "loss": 5.3205,
      "mean_token_accuracy": 0.208839613199234,
      "num_tokens": 65973287.0,
      "step": 32555
    },
    {
      "entropy": 6.055050420761108,
      "epoch": 2.8454076728130735,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004193719016243456,
      "loss": 5.2444,
      "mean_token_accuracy": 0.1998657539486885,
      "num_tokens": 65982034.0,
      "step": 32560
    },
    {
      "entropy": 6.040397310256958,
      "epoch": 2.845844621165778,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041934799982880594,
      "loss": 5.2411,
      "mean_token_accuracy": 0.20509952008724214,
      "num_tokens": 65992612.0,
      "step": 32565
    },
    {
      "entropy": 6.007578992843628,
      "epoch": 2.846281569518483,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041932409526459607,
      "loss": 5.2329,
      "mean_token_accuracy": 0.20564584881067277,
      "num_tokens": 66002788.0,
      "step": 32570
    },
    {
      "entropy": 5.9984441757202145,
      "epoch": 2.846718517871188,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041930018793217443,
      "loss": 5.3118,
      "mean_token_accuracy": 0.20306323915719987,
      "num_tokens": 66012559.0,
      "step": 32575
    },
    {
      "entropy": 5.997462224960327,
      "epoch": 2.8471554662238923,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004192762778319996,
      "loss": 5.3177,
      "mean_token_accuracy": 0.1943889781832695,
      "num_tokens": 66023230.0,
      "step": 32580
    },
    {
      "entropy": 6.033141708374023,
      "epoch": 2.847592414576597,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041925236496453006,
      "loss": 5.3141,
      "mean_token_accuracy": 0.20572146624326706,
      "num_tokens": 66033556.0,
      "step": 32585
    },
    {
      "entropy": 6.071158742904663,
      "epoch": 2.8480293629293016,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041922844933022467,
      "loss": 5.3024,
      "mean_token_accuracy": 0.20163604915142058,
      "num_tokens": 66042495.0,
      "step": 32590
    },
    {
      "entropy": 5.978684997558593,
      "epoch": 2.8484663112820066,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041920453092954204,
      "loss": 5.2517,
      "mean_token_accuracy": 0.20757684707641602,
      "num_tokens": 66051933.0,
      "step": 32595
    },
    {
      "entropy": 5.976401281356812,
      "epoch": 2.848903259634711,
      "grad_norm": 1.25,
      "learning_rate": 0.000419180609762941,
      "loss": 5.2116,
      "mean_token_accuracy": 0.2116439998149872,
      "num_tokens": 66060964.0,
      "step": 32600
    },
    {
      "entropy": 6.048172378540039,
      "epoch": 2.849340207987416,
      "grad_norm": 1.328125,
      "learning_rate": 0.00041915668583088016,
      "loss": 5.2608,
      "mean_token_accuracy": 0.20448898226022721,
      "num_tokens": 66070210.0,
      "step": 32605
    },
    {
      "entropy": 6.060002374649048,
      "epoch": 2.849777156340121,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041913275913381864,
      "loss": 5.3193,
      "mean_token_accuracy": 0.2055842638015747,
      "num_tokens": 66081010.0,
      "step": 32610
    },
    {
      "entropy": 6.026166009902954,
      "epoch": 2.8502141046928253,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004191088296722153,
      "loss": 5.2217,
      "mean_token_accuracy": 0.21191585510969163,
      "num_tokens": 66090448.0,
      "step": 32615
    },
    {
      "entropy": 6.028362131118774,
      "epoch": 2.85065105304553,
      "grad_norm": 1.3125,
      "learning_rate": 0.000419084897446529,
      "loss": 5.3158,
      "mean_token_accuracy": 0.20195286422967912,
      "num_tokens": 66100357.0,
      "step": 32620
    },
    {
      "entropy": 6.015615320205688,
      "epoch": 2.8510880013982347,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000419060962457219,
      "loss": 5.2571,
      "mean_token_accuracy": 0.2073473647236824,
      "num_tokens": 66111068.0,
      "step": 32625
    },
    {
      "entropy": 6.044771957397461,
      "epoch": 2.8515249497509396,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041903702470474414,
      "loss": 5.254,
      "mean_token_accuracy": 0.20501651018857955,
      "num_tokens": 66122185.0,
      "step": 32630
    },
    {
      "entropy": 6.029228162765503,
      "epoch": 2.851961898103644,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004190130841895637,
      "loss": 5.247,
      "mean_token_accuracy": 0.2056267887353897,
      "num_tokens": 66133192.0,
      "step": 32635
    },
    {
      "entropy": 6.024877500534058,
      "epoch": 2.852398846456349,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004189891409121369,
      "loss": 5.3332,
      "mean_token_accuracy": 0.2019622653722763,
      "num_tokens": 66143381.0,
      "step": 32640
    },
    {
      "entropy": 6.015320682525635,
      "epoch": 2.8528357948090535,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041896519487292285,
      "loss": 5.2684,
      "mean_token_accuracy": 0.20604032874107361,
      "num_tokens": 66152883.0,
      "step": 32645
    },
    {
      "entropy": 6.03453049659729,
      "epoch": 2.8532727431617584,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041894124607238104,
      "loss": 5.2006,
      "mean_token_accuracy": 0.20699473321437836,
      "num_tokens": 66162725.0,
      "step": 32650
    },
    {
      "entropy": 6.09308090209961,
      "epoch": 2.853709691514463,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004189172945109706,
      "loss": 5.3294,
      "mean_token_accuracy": 0.20370350331068038,
      "num_tokens": 66171478.0,
      "step": 32655
    },
    {
      "entropy": 5.926352405548096,
      "epoch": 2.8541466398671678,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004188933401891511,
      "loss": 5.1614,
      "mean_token_accuracy": 0.21533098369836806,
      "num_tokens": 66181044.0,
      "step": 32660
    },
    {
      "entropy": 5.981067752838134,
      "epoch": 2.8545835882198727,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004188693831073818,
      "loss": 5.2581,
      "mean_token_accuracy": 0.20638370811939238,
      "num_tokens": 66190848.0,
      "step": 32665
    },
    {
      "entropy": 6.023552370071411,
      "epoch": 2.855020536572577,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00041884542326612245,
      "loss": 5.3031,
      "mean_token_accuracy": 0.19695708751678467,
      "num_tokens": 66200756.0,
      "step": 32670
    },
    {
      "entropy": 6.0655430316925045,
      "epoch": 2.8554574849252816,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041882146066583257,
      "loss": 5.3168,
      "mean_token_accuracy": 0.1989654168486595,
      "num_tokens": 66211463.0,
      "step": 32675
    },
    {
      "entropy": 6.074695110321045,
      "epoch": 2.8558944332779865,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041879749530697165,
      "loss": 5.3017,
      "mean_token_accuracy": 0.2043053388595581,
      "num_tokens": 66222522.0,
      "step": 32680
    },
    {
      "entropy": 6.063910913467407,
      "epoch": 2.8563313816306914,
      "grad_norm": 1.125,
      "learning_rate": 0.0004187735271899994,
      "loss": 5.3318,
      "mean_token_accuracy": 0.20017997920513153,
      "num_tokens": 66232741.0,
      "step": 32685
    },
    {
      "entropy": 6.057132720947266,
      "epoch": 2.856768329983396,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041874955631537555,
      "loss": 5.2057,
      "mean_token_accuracy": 0.2036381870508194,
      "num_tokens": 66242845.0,
      "step": 32690
    },
    {
      "entropy": 5.975073766708374,
      "epoch": 2.857205278336101,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041872558268355993,
      "loss": 5.2121,
      "mean_token_accuracy": 0.209047594666481,
      "num_tokens": 66252697.0,
      "step": 32695
    },
    {
      "entropy": 5.972945642471314,
      "epoch": 2.8576422266888053,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041870160629501224,
      "loss": 5.279,
      "mean_token_accuracy": 0.20177379697561265,
      "num_tokens": 66263607.0,
      "step": 32700
    },
    {
      "entropy": 6.01730899810791,
      "epoch": 2.85807917504151,
      "grad_norm": 1.125,
      "learning_rate": 0.00041867762715019257,
      "loss": 5.2186,
      "mean_token_accuracy": 0.2112663984298706,
      "num_tokens": 66274352.0,
      "step": 32705
    },
    {
      "entropy": 6.033155107498169,
      "epoch": 2.8585161233942147,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041865364524956066,
      "loss": 5.2811,
      "mean_token_accuracy": 0.2086113840341568,
      "num_tokens": 66283972.0,
      "step": 32710
    },
    {
      "entropy": 6.032536315917969,
      "epoch": 2.8589530717469196,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041862966059357665,
      "loss": 5.3171,
      "mean_token_accuracy": 0.20252638012170793,
      "num_tokens": 66295179.0,
      "step": 32715
    },
    {
      "entropy": 6.053675937652588,
      "epoch": 2.859390020099624,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004186056731827003,
      "loss": 5.3238,
      "mean_token_accuracy": 0.19883206188678743,
      "num_tokens": 66306065.0,
      "step": 32720
    },
    {
      "entropy": 6.01224365234375,
      "epoch": 2.859826968452329,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000418581683017392,
      "loss": 5.2331,
      "mean_token_accuracy": 0.20346461832523347,
      "num_tokens": 66315491.0,
      "step": 32725
    },
    {
      "entropy": 5.996980285644531,
      "epoch": 2.8602639168050334,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004185576900981119,
      "loss": 5.2159,
      "mean_token_accuracy": 0.20853203386068345,
      "num_tokens": 66324327.0,
      "step": 32730
    },
    {
      "entropy": 6.016183567047119,
      "epoch": 2.8607008651577384,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004185336944253199,
      "loss": 5.2019,
      "mean_token_accuracy": 0.2069592535495758,
      "num_tokens": 66334201.0,
      "step": 32735
    },
    {
      "entropy": 6.031157064437866,
      "epoch": 2.8611378135104433,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041850969599947655,
      "loss": 5.2598,
      "mean_token_accuracy": 0.20723959505558015,
      "num_tokens": 66344745.0,
      "step": 32740
    },
    {
      "entropy": 5.936083555221558,
      "epoch": 2.8615747618631477,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041848569482104197,
      "loss": 5.2397,
      "mean_token_accuracy": 0.2091739758849144,
      "num_tokens": 66354815.0,
      "step": 32745
    },
    {
      "entropy": 6.018384599685669,
      "epoch": 2.862011710215852,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004184616908904765,
      "loss": 5.2717,
      "mean_token_accuracy": 0.20134119540452958,
      "num_tokens": 66364744.0,
      "step": 32750
    },
    {
      "entropy": 6.021817922592163,
      "epoch": 2.862448658568557,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004184376842082408,
      "loss": 5.3316,
      "mean_token_accuracy": 0.2016370713710785,
      "num_tokens": 66376363.0,
      "step": 32755
    },
    {
      "entropy": 6.032803964614868,
      "epoch": 2.862885606921262,
      "grad_norm": 1.40625,
      "learning_rate": 0.00041841367477479503,
      "loss": 5.2925,
      "mean_token_accuracy": 0.20121046900749207,
      "num_tokens": 66386149.0,
      "step": 32760
    },
    {
      "entropy": 6.018321180343628,
      "epoch": 2.8633225552739665,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004183896625905999,
      "loss": 5.2367,
      "mean_token_accuracy": 0.20959027260541915,
      "num_tokens": 66396874.0,
      "step": 32765
    },
    {
      "entropy": 6.007727193832397,
      "epoch": 2.8637595036266714,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041836564765611583,
      "loss": 5.2177,
      "mean_token_accuracy": 0.20888155102729797,
      "num_tokens": 66406761.0,
      "step": 32770
    },
    {
      "entropy": 6.058092880249023,
      "epoch": 2.864196451979376,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041834162997180364,
      "loss": 5.3371,
      "mean_token_accuracy": 0.20237759202718736,
      "num_tokens": 66416947.0,
      "step": 32775
    },
    {
      "entropy": 5.941180849075318,
      "epoch": 2.864633400332081,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004183176095381238,
      "loss": 5.2521,
      "mean_token_accuracy": 0.20978668779134751,
      "num_tokens": 66427980.0,
      "step": 32780
    },
    {
      "entropy": 5.941051387786866,
      "epoch": 2.8650703486847853,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041829358635553707,
      "loss": 5.1525,
      "mean_token_accuracy": 0.21431707143783568,
      "num_tokens": 66437226.0,
      "step": 32785
    },
    {
      "entropy": 5.985772275924683,
      "epoch": 2.86550729703749,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041826956042450436,
      "loss": 5.3066,
      "mean_token_accuracy": 0.2010796546936035,
      "num_tokens": 66447820.0,
      "step": 32790
    },
    {
      "entropy": 6.012532663345337,
      "epoch": 2.865944245390195,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004182455317454863,
      "loss": 5.2822,
      "mean_token_accuracy": 0.20089313238859177,
      "num_tokens": 66459727.0,
      "step": 32795
    },
    {
      "entropy": 6.026930570602417,
      "epoch": 2.8663811937428996,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041822150031894397,
      "loss": 5.2243,
      "mean_token_accuracy": 0.20202300995588302,
      "num_tokens": 66469743.0,
      "step": 32800
    },
    {
      "entropy": 6.0277587413787845,
      "epoch": 2.866818142095604,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041819746614533824,
      "loss": 5.3118,
      "mean_token_accuracy": 0.2024797722697258,
      "num_tokens": 66479974.0,
      "step": 32805
    },
    {
      "entropy": 5.95627613067627,
      "epoch": 2.867255090448309,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041817342922513,
      "loss": 5.1775,
      "mean_token_accuracy": 0.2149068146944046,
      "num_tokens": 66489605.0,
      "step": 32810
    },
    {
      "entropy": 6.007072639465332,
      "epoch": 2.867692038801014,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041814938955878035,
      "loss": 5.2908,
      "mean_token_accuracy": 0.20900786817073821,
      "num_tokens": 66499454.0,
      "step": 32815
    },
    {
      "entropy": 6.034838581085205,
      "epoch": 2.8681289871537183,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041812534714675046,
      "loss": 5.3112,
      "mean_token_accuracy": 0.2060420110821724,
      "num_tokens": 66510076.0,
      "step": 32820
    },
    {
      "entropy": 6.0611337184906,
      "epoch": 2.8685659355064232,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041810130198950137,
      "loss": 5.2904,
      "mean_token_accuracy": 0.20395987778902053,
      "num_tokens": 66520659.0,
      "step": 32825
    },
    {
      "entropy": 6.035171175003052,
      "epoch": 2.8690028838591277,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004180772540874943,
      "loss": 5.2934,
      "mean_token_accuracy": 0.2059799998998642,
      "num_tokens": 66531387.0,
      "step": 32830
    },
    {
      "entropy": 6.030501747131348,
      "epoch": 2.8694398322118326,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041805320344119064,
      "loss": 5.2321,
      "mean_token_accuracy": 0.20933480709791183,
      "num_tokens": 66540654.0,
      "step": 32835
    },
    {
      "entropy": 6.03508152961731,
      "epoch": 2.869876780564537,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004180291500510515,
      "loss": 5.259,
      "mean_token_accuracy": 0.2054838463664055,
      "num_tokens": 66550966.0,
      "step": 32840
    },
    {
      "entropy": 6.062289476394653,
      "epoch": 2.870313728917242,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041800509391753834,
      "loss": 5.3609,
      "mean_token_accuracy": 0.19630265086889268,
      "num_tokens": 66561550.0,
      "step": 32845
    },
    {
      "entropy": 6.0305013179779055,
      "epoch": 2.870750677269947,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004179810350411125,
      "loss": 5.2355,
      "mean_token_accuracy": 0.21149528920650482,
      "num_tokens": 66571477.0,
      "step": 32850
    },
    {
      "entropy": 6.037065505981445,
      "epoch": 2.8711876256226514,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004179569734222355,
      "loss": 5.2954,
      "mean_token_accuracy": 0.2059851333498955,
      "num_tokens": 66581976.0,
      "step": 32855
    },
    {
      "entropy": 6.056600046157837,
      "epoch": 2.871624573975356,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041793290906136886,
      "loss": 5.2703,
      "mean_token_accuracy": 0.20036799758672713,
      "num_tokens": 66591747.0,
      "step": 32860
    },
    {
      "entropy": 6.089670848846436,
      "epoch": 2.872061522328061,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004179088419589741,
      "loss": 5.259,
      "mean_token_accuracy": 0.20014116466045379,
      "num_tokens": 66602681.0,
      "step": 32865
    },
    {
      "entropy": 6.0287481307983395,
      "epoch": 2.8724984706807657,
      "grad_norm": 1.25,
      "learning_rate": 0.00041788477211551295,
      "loss": 5.2765,
      "mean_token_accuracy": 0.2062819331884384,
      "num_tokens": 66612887.0,
      "step": 32870
    },
    {
      "entropy": 6.0087477684021,
      "epoch": 2.87293541903347,
      "grad_norm": 1.25,
      "learning_rate": 0.00041786069953144697,
      "loss": 5.2224,
      "mean_token_accuracy": 0.20712469667196273,
      "num_tokens": 66622418.0,
      "step": 32875
    },
    {
      "entropy": 6.016960620880127,
      "epoch": 2.873372367386175,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041783662420723786,
      "loss": 5.3175,
      "mean_token_accuracy": 0.2002807229757309,
      "num_tokens": 66633457.0,
      "step": 32880
    },
    {
      "entropy": 6.034476947784424,
      "epoch": 2.8738093157388795,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004178125461433475,
      "loss": 5.2224,
      "mean_token_accuracy": 0.20563973635435104,
      "num_tokens": 66643691.0,
      "step": 32885
    },
    {
      "entropy": 6.048425388336182,
      "epoch": 2.8742462640915845,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041778846534023763,
      "loss": 5.2514,
      "mean_token_accuracy": 0.20531596392393112,
      "num_tokens": 66653745.0,
      "step": 32890
    },
    {
      "entropy": 5.99768180847168,
      "epoch": 2.874683212444289,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041776438179837026,
      "loss": 5.2414,
      "mean_token_accuracy": 0.2139417201280594,
      "num_tokens": 66664093.0,
      "step": 32895
    },
    {
      "entropy": 6.008603715896607,
      "epoch": 2.875120160796994,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004177402955182071,
      "loss": 5.3119,
      "mean_token_accuracy": 0.20062779635190964,
      "num_tokens": 66675204.0,
      "step": 32900
    },
    {
      "entropy": 6.076387929916382,
      "epoch": 2.8755571091496988,
      "grad_norm": 1.25,
      "learning_rate": 0.00041771620650021043,
      "loss": 5.2981,
      "mean_token_accuracy": 0.20447054654359817,
      "num_tokens": 66684789.0,
      "step": 32905
    },
    {
      "entropy": 6.098245286941529,
      "epoch": 2.8759940575024032,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041769211474484204,
      "loss": 5.3021,
      "mean_token_accuracy": 0.20425219982862472,
      "num_tokens": 66694734.0,
      "step": 32910
    },
    {
      "entropy": 6.050508451461792,
      "epoch": 2.8764310058551077,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004176680202525641,
      "loss": 5.3128,
      "mean_token_accuracy": 0.20000965446233748,
      "num_tokens": 66705231.0,
      "step": 32915
    },
    {
      "entropy": 5.976842164993286,
      "epoch": 2.8768679542078126,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00041764392302383893,
      "loss": 5.2427,
      "mean_token_accuracy": 0.208383309841156,
      "num_tokens": 66715627.0,
      "step": 32920
    },
    {
      "entropy": 6.002675294876099,
      "epoch": 2.8773049025605175,
      "grad_norm": 1.25,
      "learning_rate": 0.0004176198230591284,
      "loss": 5.2376,
      "mean_token_accuracy": 0.2068549931049347,
      "num_tokens": 66725525.0,
      "step": 32925
    },
    {
      "entropy": 6.047699213027954,
      "epoch": 2.877741850913222,
      "grad_norm": 1.265625,
      "learning_rate": 0.000417595720358895,
      "loss": 5.2876,
      "mean_token_accuracy": 0.19729862362146378,
      "num_tokens": 66736352.0,
      "step": 32930
    },
    {
      "entropy": 6.04535551071167,
      "epoch": 2.878178799265927,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041757161492360096,
      "loss": 5.3449,
      "mean_token_accuracy": 0.19886990934610366,
      "num_tokens": 66745961.0,
      "step": 32935
    },
    {
      "entropy": 6.0463175773620605,
      "epoch": 2.8786157476186314,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041754750675370865,
      "loss": 5.2123,
      "mean_token_accuracy": 0.21161111295223237,
      "num_tokens": 66755295.0,
      "step": 32940
    },
    {
      "entropy": 6.004108238220215,
      "epoch": 2.8790526959713363,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004175233958496804,
      "loss": 5.2197,
      "mean_token_accuracy": 0.20530531257390977,
      "num_tokens": 66765089.0,
      "step": 32945
    },
    {
      "entropy": 6.053836154937744,
      "epoch": 2.8794896443240408,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004174992822119788,
      "loss": 5.2528,
      "mean_token_accuracy": 0.20455739498138428,
      "num_tokens": 66774995.0,
      "step": 32950
    },
    {
      "entropy": 6.018562126159668,
      "epoch": 2.8799265926767457,
      "grad_norm": 1.125,
      "learning_rate": 0.0004174751658410663,
      "loss": 5.2896,
      "mean_token_accuracy": 0.20586961954832078,
      "num_tokens": 66784584.0,
      "step": 32955
    },
    {
      "entropy": 5.995736217498779,
      "epoch": 2.8803635410294506,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041745104673740545,
      "loss": 5.2587,
      "mean_token_accuracy": 0.20573701858520507,
      "num_tokens": 66794747.0,
      "step": 32960
    },
    {
      "entropy": 6.072381067276001,
      "epoch": 2.880800489382155,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041742692490145895,
      "loss": 5.2666,
      "mean_token_accuracy": 0.20397120267152785,
      "num_tokens": 66804632.0,
      "step": 32965
    },
    {
      "entropy": 5.963957643508911,
      "epoch": 2.8812374377348595,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004174028003336893,
      "loss": 5.2123,
      "mean_token_accuracy": 0.20669952929019927,
      "num_tokens": 66814534.0,
      "step": 32970
    },
    {
      "entropy": 5.997437620162964,
      "epoch": 2.8816743860875644,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004173786730345594,
      "loss": 5.3374,
      "mean_token_accuracy": 0.2004629984498024,
      "num_tokens": 66824629.0,
      "step": 32975
    },
    {
      "entropy": 6.054879140853882,
      "epoch": 2.8821113344402693,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004173545430045319,
      "loss": 5.3023,
      "mean_token_accuracy": 0.20000162422657014,
      "num_tokens": 66834345.0,
      "step": 32980
    },
    {
      "entropy": 6.029374361038208,
      "epoch": 2.882548282792974,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041733041024406966,
      "loss": 5.2771,
      "mean_token_accuracy": 0.20429684072732926,
      "num_tokens": 66845272.0,
      "step": 32985
    },
    {
      "entropy": 6.068456649780273,
      "epoch": 2.8829852311456787,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004173062747536356,
      "loss": 5.2867,
      "mean_token_accuracy": 0.20449685156345368,
      "num_tokens": 66855022.0,
      "step": 32990
    },
    {
      "entropy": 6.024906301498413,
      "epoch": 2.883422179498383,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004172821365336926,
      "loss": 5.2217,
      "mean_token_accuracy": 0.20768261700868607,
      "num_tokens": 66864895.0,
      "step": 32995
    },
    {
      "entropy": 5.962719821929932,
      "epoch": 2.883859127851088,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004172579955847037,
      "loss": 5.2061,
      "mean_token_accuracy": 0.20856466442346572,
      "num_tokens": 66876134.0,
      "step": 33000
    },
    {
      "epoch": 2.883859127851088,
      "eval_entropy": 5.736183507514098,
      "eval_loss": 5.295504570007324,
      "eval_mean_token_accuracy": 0.21275744903112093,
      "eval_num_tokens": 66876134.0,
      "eval_runtime": 21.6035,
      "eval_samples_per_second": 1713.33,
      "eval_steps_per_second": 214.178,
      "step": 33000
    },
    {
      "entropy": 6.009463930130005,
      "epoch": 2.8842960762037926,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004172338519071318,
      "loss": 5.2361,
      "mean_token_accuracy": 0.20433570593595504,
      "num_tokens": 66886089.0,
      "step": 33005
    },
    {
      "entropy": 5.89949049949646,
      "epoch": 2.8847330245564975,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041720970550144024,
      "loss": 5.1531,
      "mean_token_accuracy": 0.21637639850378038,
      "num_tokens": 66896072.0,
      "step": 33010
    },
    {
      "entropy": 6.034364557266235,
      "epoch": 2.8851699729092024,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041718555636809194,
      "loss": 5.364,
      "mean_token_accuracy": 0.19315982460975648,
      "num_tokens": 66906030.0,
      "step": 33015
    },
    {
      "entropy": 6.016134262084961,
      "epoch": 2.885606921261907,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004171614045075501,
      "loss": 5.2297,
      "mean_token_accuracy": 0.2044316679239273,
      "num_tokens": 66916714.0,
      "step": 33020
    },
    {
      "entropy": 5.992003011703491,
      "epoch": 2.8860438696146113,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004171372499202781,
      "loss": 5.2941,
      "mean_token_accuracy": 0.20344114452600479,
      "num_tokens": 66926570.0,
      "step": 33025
    },
    {
      "entropy": 6.063661718368531,
      "epoch": 2.8864808179673163,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004171130926067391,
      "loss": 5.3695,
      "mean_token_accuracy": 0.1984993577003479,
      "num_tokens": 66937565.0,
      "step": 33030
    },
    {
      "entropy": 6.04550404548645,
      "epoch": 2.886917766320021,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004170889325673967,
      "loss": 5.3148,
      "mean_token_accuracy": 0.20163129568099974,
      "num_tokens": 66948363.0,
      "step": 33035
    },
    {
      "entropy": 5.970583438873291,
      "epoch": 2.8873547146727256,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041706476980271395,
      "loss": 5.2014,
      "mean_token_accuracy": 0.21069546341896056,
      "num_tokens": 66958312.0,
      "step": 33040
    },
    {
      "entropy": 5.997757196426392,
      "epoch": 2.8877916630254306,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041704060431315456,
      "loss": 5.1684,
      "mean_token_accuracy": 0.21596708595752717,
      "num_tokens": 66967880.0,
      "step": 33045
    },
    {
      "entropy": 5.952107572555542,
      "epoch": 2.888228611378135,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041701643609918187,
      "loss": 5.1949,
      "mean_token_accuracy": 0.21615796238183976,
      "num_tokens": 66977930.0,
      "step": 33050
    },
    {
      "entropy": 6.00415449142456,
      "epoch": 2.88866555973084,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004169922651612596,
      "loss": 5.3275,
      "mean_token_accuracy": 0.2016575664281845,
      "num_tokens": 66987780.0,
      "step": 33055
    },
    {
      "entropy": 6.055237293243408,
      "epoch": 2.8891025080835444,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041696809149985124,
      "loss": 5.2744,
      "mean_token_accuracy": 0.1983456790447235,
      "num_tokens": 66997910.0,
      "step": 33060
    },
    {
      "entropy": 6.0365385055542,
      "epoch": 2.8895394564362493,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041694391511542053,
      "loss": 5.2063,
      "mean_token_accuracy": 0.21041217148303987,
      "num_tokens": 67007365.0,
      "step": 33065
    },
    {
      "entropy": 5.960046339035034,
      "epoch": 2.889976404788954,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004169197360084311,
      "loss": 5.1735,
      "mean_token_accuracy": 0.21747081726789474,
      "num_tokens": 67016438.0,
      "step": 33070
    },
    {
      "entropy": 5.92432279586792,
      "epoch": 2.8904133531416587,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041689555417934675,
      "loss": 5.0794,
      "mean_token_accuracy": 0.2284829631447792,
      "num_tokens": 67026842.0,
      "step": 33075
    },
    {
      "entropy": 6.0024830341339115,
      "epoch": 2.890850301494363,
      "grad_norm": 1.25,
      "learning_rate": 0.00041687136962863144,
      "loss": 5.3259,
      "mean_token_accuracy": 0.20366783291101456,
      "num_tokens": 67037343.0,
      "step": 33080
    },
    {
      "entropy": 6.023594331741333,
      "epoch": 2.891287249847068,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041684718235674887,
      "loss": 5.2631,
      "mean_token_accuracy": 0.2005263090133667,
      "num_tokens": 67047352.0,
      "step": 33085
    },
    {
      "entropy": 6.0338866233825685,
      "epoch": 2.891724198199773,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004168229923641629,
      "loss": 5.3365,
      "mean_token_accuracy": 0.19312265664339065,
      "num_tokens": 67058717.0,
      "step": 33090
    },
    {
      "entropy": 6.044533586502075,
      "epoch": 2.8921611465524775,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004167987996513378,
      "loss": 5.2665,
      "mean_token_accuracy": 0.20684145092964173,
      "num_tokens": 67068124.0,
      "step": 33095
    },
    {
      "entropy": 6.016391801834106,
      "epoch": 2.892598094905182,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004167746042187373,
      "loss": 5.2917,
      "mean_token_accuracy": 0.20236933529376983,
      "num_tokens": 67078906.0,
      "step": 33100
    },
    {
      "entropy": 5.977835178375244,
      "epoch": 2.893035043257887,
      "grad_norm": 1.578125,
      "learning_rate": 0.00041675040606682564,
      "loss": 5.2325,
      "mean_token_accuracy": 0.21125175356864928,
      "num_tokens": 67090424.0,
      "step": 33105
    },
    {
      "entropy": 6.063391017913818,
      "epoch": 2.8934719916105918,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041672620519606687,
      "loss": 5.2824,
      "mean_token_accuracy": 0.2046969398856163,
      "num_tokens": 67100311.0,
      "step": 33110
    },
    {
      "entropy": 6.038122177124023,
      "epoch": 2.8939089399632962,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004167020016069253,
      "loss": 5.2822,
      "mean_token_accuracy": 0.20288049876689912,
      "num_tokens": 67109337.0,
      "step": 33115
    },
    {
      "entropy": 5.998903226852417,
      "epoch": 2.894345888316001,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041667779529986504,
      "loss": 5.2534,
      "mean_token_accuracy": 0.20209793746471405,
      "num_tokens": 67119795.0,
      "step": 33120
    },
    {
      "entropy": 6.002016019821167,
      "epoch": 2.8947828366687056,
      "grad_norm": 1.25,
      "learning_rate": 0.00041665358627535043,
      "loss": 5.2837,
      "mean_token_accuracy": 0.199890698492527,
      "num_tokens": 67129444.0,
      "step": 33125
    },
    {
      "entropy": 6.035126209259033,
      "epoch": 2.8952197850214105,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004166293745338458,
      "loss": 5.2882,
      "mean_token_accuracy": 0.20776375830173494,
      "num_tokens": 67138868.0,
      "step": 33130
    },
    {
      "entropy": 6.042664909362793,
      "epoch": 2.895656733374115,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004166051600758156,
      "loss": 5.3373,
      "mean_token_accuracy": 0.2021288737654686,
      "num_tokens": 67149615.0,
      "step": 33135
    },
    {
      "entropy": 6.001376724243164,
      "epoch": 2.89609368172682,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041658094290172415,
      "loss": 5.2683,
      "mean_token_accuracy": 0.20975396484136583,
      "num_tokens": 67159745.0,
      "step": 33140
    },
    {
      "entropy": 6.04534912109375,
      "epoch": 2.896530630079525,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004165567230120362,
      "loss": 5.2215,
      "mean_token_accuracy": 0.20742821395397187,
      "num_tokens": 67169900.0,
      "step": 33145
    },
    {
      "entropy": 5.961681890487671,
      "epoch": 2.8969675784322293,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004165325004072159,
      "loss": 5.2623,
      "mean_token_accuracy": 0.21174887865781783,
      "num_tokens": 67180752.0,
      "step": 33150
    },
    {
      "entropy": 6.012210702896118,
      "epoch": 2.8974045267849338,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041650827508772826,
      "loss": 5.3083,
      "mean_token_accuracy": 0.20231623500585555,
      "num_tokens": 67191466.0,
      "step": 33155
    },
    {
      "entropy": 5.9916229248046875,
      "epoch": 2.8978414751376387,
      "grad_norm": 1.125,
      "learning_rate": 0.00041648404705403765,
      "loss": 5.2335,
      "mean_token_accuracy": 0.20424074083566665,
      "num_tokens": 67202813.0,
      "step": 33160
    },
    {
      "entropy": 6.009753608703614,
      "epoch": 2.8982784234903436,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041645981630660894,
      "loss": 5.1715,
      "mean_token_accuracy": 0.21162269711494447,
      "num_tokens": 67213686.0,
      "step": 33165
    },
    {
      "entropy": 6.040389680862427,
      "epoch": 2.898715371843048,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004164355828459068,
      "loss": 5.2682,
      "mean_token_accuracy": 0.20552642196416854,
      "num_tokens": 67224955.0,
      "step": 33170
    },
    {
      "entropy": 6.003567504882812,
      "epoch": 2.899152320195753,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000416411346672396,
      "loss": 5.2456,
      "mean_token_accuracy": 0.2035496413707733,
      "num_tokens": 67234438.0,
      "step": 33175
    },
    {
      "entropy": 6.046557378768921,
      "epoch": 2.8995892685484574,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041638710778654157,
      "loss": 5.3085,
      "mean_token_accuracy": 0.20059916973114014,
      "num_tokens": 67245151.0,
      "step": 33180
    },
    {
      "entropy": 6.009650230407715,
      "epoch": 2.9000262169011624,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004163628661888083,
      "loss": 5.2646,
      "mean_token_accuracy": 0.20534302443265914,
      "num_tokens": 67254942.0,
      "step": 33185
    },
    {
      "entropy": 6.085052871704102,
      "epoch": 2.900463165253867,
      "grad_norm": 1.328125,
      "learning_rate": 0.00041633862187966125,
      "loss": 5.2841,
      "mean_token_accuracy": 0.20320933759212495,
      "num_tokens": 67264435.0,
      "step": 33190
    },
    {
      "entropy": 5.963392686843872,
      "epoch": 2.9009001136065717,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004163143748595653,
      "loss": 5.2369,
      "mean_token_accuracy": 0.20107194930315017,
      "num_tokens": 67273304.0,
      "step": 33195
    },
    {
      "entropy": 6.0007452964782715,
      "epoch": 2.9013370619592767,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041629012512898556,
      "loss": 5.3234,
      "mean_token_accuracy": 0.20249418616294862,
      "num_tokens": 67283713.0,
      "step": 33200
    },
    {
      "entropy": 6.016227960586548,
      "epoch": 2.901774010311981,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004162658726883873,
      "loss": 5.151,
      "mean_token_accuracy": 0.21649425923824311,
      "num_tokens": 67292420.0,
      "step": 33205
    },
    {
      "entropy": 6.002579402923584,
      "epoch": 2.9022109586646856,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004162416175382354,
      "loss": 5.2471,
      "mean_token_accuracy": 0.20596883893013002,
      "num_tokens": 67302138.0,
      "step": 33210
    },
    {
      "entropy": 5.999075794219971,
      "epoch": 2.9026479070173905,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004162173596789954,
      "loss": 5.2597,
      "mean_token_accuracy": 0.20615230798721312,
      "num_tokens": 67311776.0,
      "step": 33215
    },
    {
      "entropy": 6.010639429092407,
      "epoch": 2.9030848553700954,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004161930991111325,
      "loss": 5.2768,
      "mean_token_accuracy": 0.2008287400007248,
      "num_tokens": 67322296.0,
      "step": 33220
    },
    {
      "entropy": 6.017301845550537,
      "epoch": 2.9035218037228,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004161688358351118,
      "loss": 5.2325,
      "mean_token_accuracy": 0.19992891550064087,
      "num_tokens": 67332173.0,
      "step": 33225
    },
    {
      "entropy": 6.015643072128296,
      "epoch": 2.903958752075505,
      "grad_norm": 1.15625,
      "learning_rate": 0.000416144569851399,
      "loss": 5.2486,
      "mean_token_accuracy": 0.20667391568422316,
      "num_tokens": 67342156.0,
      "step": 33230
    },
    {
      "entropy": 5.99610013961792,
      "epoch": 2.9043957004282093,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041612030116045933,
      "loss": 5.2962,
      "mean_token_accuracy": 0.20230839401483536,
      "num_tokens": 67353208.0,
      "step": 33235
    },
    {
      "entropy": 6.021438980102539,
      "epoch": 2.904832648780914,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041609602976275836,
      "loss": 5.2051,
      "mean_token_accuracy": 0.2143350213766098,
      "num_tokens": 67362577.0,
      "step": 33240
    },
    {
      "entropy": 6.0322459697723385,
      "epoch": 2.9052695971336187,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004160717556587616,
      "loss": 5.1867,
      "mean_token_accuracy": 0.21318580061197281,
      "num_tokens": 67371996.0,
      "step": 33245
    },
    {
      "entropy": 5.957010412216187,
      "epoch": 2.9057065454863236,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004160474788489346,
      "loss": 5.2107,
      "mean_token_accuracy": 0.2105372041463852,
      "num_tokens": 67381714.0,
      "step": 33250
    },
    {
      "entropy": 5.952681398391723,
      "epoch": 2.9061434938390285,
      "grad_norm": 1.3125,
      "learning_rate": 0.00041602319933374305,
      "loss": 5.2296,
      "mean_token_accuracy": 0.21144041270017624,
      "num_tokens": 67391551.0,
      "step": 33255
    },
    {
      "entropy": 6.023816347122192,
      "epoch": 2.906580442191733,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004159989171136527,
      "loss": 5.2657,
      "mean_token_accuracy": 0.19586805403232574,
      "num_tokens": 67400816.0,
      "step": 33260
    },
    {
      "entropy": 6.014206171035767,
      "epoch": 2.9070173905444374,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004159746321891292,
      "loss": 5.2722,
      "mean_token_accuracy": 0.20385874062776566,
      "num_tokens": 67410633.0,
      "step": 33265
    },
    {
      "entropy": 6.014793729782104,
      "epoch": 2.9074543388971423,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004159503445606385,
      "loss": 5.2598,
      "mean_token_accuracy": 0.20967214703559875,
      "num_tokens": 67420588.0,
      "step": 33270
    },
    {
      "entropy": 6.035376024246216,
      "epoch": 2.9078912872498472,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041592605422864615,
      "loss": 5.2526,
      "mean_token_accuracy": 0.2083306282758713,
      "num_tokens": 67430494.0,
      "step": 33275
    },
    {
      "entropy": 5.935432720184326,
      "epoch": 2.9083282356025517,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004159017611936183,
      "loss": 5.1872,
      "mean_token_accuracy": 0.2190268248319626,
      "num_tokens": 67440692.0,
      "step": 33280
    },
    {
      "entropy": 6.005907392501831,
      "epoch": 2.9087651839552566,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041587746545602094,
      "loss": 5.2542,
      "mean_token_accuracy": 0.2050940990447998,
      "num_tokens": 67450936.0,
      "step": 33285
    },
    {
      "entropy": 5.991922664642334,
      "epoch": 2.909202132307961,
      "grad_norm": 1.25,
      "learning_rate": 0.0004158531670163199,
      "loss": 5.2416,
      "mean_token_accuracy": 0.2105686604976654,
      "num_tokens": 67460997.0,
      "step": 33290
    },
    {
      "entropy": 5.980035924911499,
      "epoch": 2.909639080660666,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004158288658749813,
      "loss": 5.2839,
      "mean_token_accuracy": 0.2089468777179718,
      "num_tokens": 67470685.0,
      "step": 33295
    },
    {
      "entropy": 6.010352277755738,
      "epoch": 2.9100760290133705,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041580456203247127,
      "loss": 5.2439,
      "mean_token_accuracy": 0.20725292414426805,
      "num_tokens": 67481734.0,
      "step": 33300
    },
    {
      "entropy": 5.89740686416626,
      "epoch": 2.9105129773660754,
      "grad_norm": 1.203125,
      "learning_rate": 0.000415780255489256,
      "loss": 5.0578,
      "mean_token_accuracy": 0.2310038149356842,
      "num_tokens": 67491290.0,
      "step": 33305
    },
    {
      "entropy": 5.991577434539795,
      "epoch": 2.9109499257187803,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004157559462458016,
      "loss": 5.3102,
      "mean_token_accuracy": 0.2042112961411476,
      "num_tokens": 67501758.0,
      "step": 33310
    },
    {
      "entropy": 6.022864818572998,
      "epoch": 2.9113868740714848,
      "grad_norm": 1.328125,
      "learning_rate": 0.00041573163430257435,
      "loss": 5.2859,
      "mean_token_accuracy": 0.20113715827465056,
      "num_tokens": 67511435.0,
      "step": 33315
    },
    {
      "entropy": 6.032228660583496,
      "epoch": 2.9118238224241892,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041570731966004065,
      "loss": 5.305,
      "mean_token_accuracy": 0.2028704583644867,
      "num_tokens": 67521408.0,
      "step": 33320
    },
    {
      "entropy": 6.030580663681031,
      "epoch": 2.912260770776894,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004156830023186669,
      "loss": 5.3316,
      "mean_token_accuracy": 0.1995191216468811,
      "num_tokens": 67531631.0,
      "step": 33325
    },
    {
      "entropy": 6.065176248550415,
      "epoch": 2.912697719129599,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004156586822789194,
      "loss": 5.3047,
      "mean_token_accuracy": 0.18796419799327851,
      "num_tokens": 67542265.0,
      "step": 33330
    },
    {
      "entropy": 6.032901954650879,
      "epoch": 2.9131346674823035,
      "grad_norm": 1.125,
      "learning_rate": 0.00041563435954126467,
      "loss": 5.2915,
      "mean_token_accuracy": 0.2057914763689041,
      "num_tokens": 67553664.0,
      "step": 33335
    },
    {
      "entropy": 6.019833421707153,
      "epoch": 2.9135716158350085,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004156100341061693,
      "loss": 5.2206,
      "mean_token_accuracy": 0.20197975933551787,
      "num_tokens": 67564500.0,
      "step": 33340
    },
    {
      "entropy": 5.99129204750061,
      "epoch": 2.914008564187713,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004155857059740997,
      "loss": 5.1969,
      "mean_token_accuracy": 0.21325988471508026,
      "num_tokens": 67573611.0,
      "step": 33345
    },
    {
      "entropy": 5.980174493789673,
      "epoch": 2.914445512540418,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004155613751455226,
      "loss": 5.2274,
      "mean_token_accuracy": 0.21080669462680818,
      "num_tokens": 67583721.0,
      "step": 33350
    },
    {
      "entropy": 6.02056622505188,
      "epoch": 2.9148824608931223,
      "grad_norm": 1.390625,
      "learning_rate": 0.00041553704162090475,
      "loss": 5.2695,
      "mean_token_accuracy": 0.2099211186170578,
      "num_tokens": 67593381.0,
      "step": 33355
    },
    {
      "entropy": 6.012459850311279,
      "epoch": 2.915319409245827,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004155127054007127,
      "loss": 5.2801,
      "mean_token_accuracy": 0.20474033802747726,
      "num_tokens": 67603799.0,
      "step": 33360
    },
    {
      "entropy": 6.06403431892395,
      "epoch": 2.9157563575985317,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041548836648541346,
      "loss": 5.2891,
      "mean_token_accuracy": 0.2021852269768715,
      "num_tokens": 67613543.0,
      "step": 33365
    },
    {
      "entropy": 6.030965185165405,
      "epoch": 2.9161933059512366,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041546402487547363,
      "loss": 5.2869,
      "mean_token_accuracy": 0.20267039835453032,
      "num_tokens": 67623302.0,
      "step": 33370
    },
    {
      "entropy": 5.992038536071777,
      "epoch": 2.916630254303941,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004154396805713602,
      "loss": 5.1788,
      "mean_token_accuracy": 0.21756059378385545,
      "num_tokens": 67633655.0,
      "step": 33375
    },
    {
      "entropy": 5.969665765762329,
      "epoch": 2.917067202656646,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041541533357354013,
      "loss": 5.2953,
      "mean_token_accuracy": 0.20458471179008483,
      "num_tokens": 67644832.0,
      "step": 33380
    },
    {
      "entropy": 6.078668594360352,
      "epoch": 2.917504151009351,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004153909838824804,
      "loss": 5.2601,
      "mean_token_accuracy": 0.20694745182991028,
      "num_tokens": 67655290.0,
      "step": 33385
    },
    {
      "entropy": 6.0007915019989015,
      "epoch": 2.9179410993620554,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000415366631498648,
      "loss": 5.1994,
      "mean_token_accuracy": 0.21110633611679078,
      "num_tokens": 67664252.0,
      "step": 33390
    },
    {
      "entropy": 6.009668636322021,
      "epoch": 2.9183780477147603,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004153422764225101,
      "loss": 5.2539,
      "mean_token_accuracy": 0.20599207729101182,
      "num_tokens": 67674655.0,
      "step": 33395
    },
    {
      "entropy": 5.963732004165649,
      "epoch": 2.9188149960674648,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004153179186545338,
      "loss": 5.1909,
      "mean_token_accuracy": 0.20825324803590775,
      "num_tokens": 67685126.0,
      "step": 33400
    },
    {
      "entropy": 5.95066819190979,
      "epoch": 2.9192519444201697,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041529355819518626,
      "loss": 5.2166,
      "mean_token_accuracy": 0.2079930678009987,
      "num_tokens": 67694111.0,
      "step": 33405
    },
    {
      "entropy": 6.03324670791626,
      "epoch": 2.919688892772874,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041526919504493475,
      "loss": 5.2626,
      "mean_token_accuracy": 0.20631029456853867,
      "num_tokens": 67704229.0,
      "step": 33410
    },
    {
      "entropy": 6.016604328155518,
      "epoch": 2.920125841125579,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041524482920424666,
      "loss": 5.2764,
      "mean_token_accuracy": 0.20705704689025878,
      "num_tokens": 67714625.0,
      "step": 33415
    },
    {
      "entropy": 6.010281610488891,
      "epoch": 2.9205627894782835,
      "grad_norm": 1.25,
      "learning_rate": 0.00041522046067358917,
      "loss": 5.2671,
      "mean_token_accuracy": 0.20161973237991332,
      "num_tokens": 67724094.0,
      "step": 33420
    },
    {
      "entropy": 6.018165445327758,
      "epoch": 2.9209997378309884,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041519608945342985,
      "loss": 5.2703,
      "mean_token_accuracy": 0.20774483382701875,
      "num_tokens": 67733233.0,
      "step": 33425
    },
    {
      "entropy": 6.035573434829712,
      "epoch": 2.921436686183693,
      "grad_norm": 1.265625,
      "learning_rate": 0.000415171715544236,
      "loss": 5.1952,
      "mean_token_accuracy": 0.2175191044807434,
      "num_tokens": 67742781.0,
      "step": 33430
    },
    {
      "entropy": 5.997231960296631,
      "epoch": 2.921873634536398,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041514733894647526,
      "loss": 5.2052,
      "mean_token_accuracy": 0.21491810381412507,
      "num_tokens": 67753166.0,
      "step": 33435
    },
    {
      "entropy": 6.033080625534057,
      "epoch": 2.9223105828891027,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004151229596606151,
      "loss": 5.3211,
      "mean_token_accuracy": 0.2046387553215027,
      "num_tokens": 67762599.0,
      "step": 33440
    },
    {
      "entropy": 6.007091903686524,
      "epoch": 2.922747531241807,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004150985776871231,
      "loss": 5.2417,
      "mean_token_accuracy": 0.2017530992627144,
      "num_tokens": 67772620.0,
      "step": 33445
    },
    {
      "entropy": 6.056059217453003,
      "epoch": 2.9231844795945117,
      "grad_norm": 1.28125,
      "learning_rate": 0.000415074193026467,
      "loss": 5.3018,
      "mean_token_accuracy": 0.20216749906539916,
      "num_tokens": 67782962.0,
      "step": 33450
    },
    {
      "entropy": 6.040071773529053,
      "epoch": 2.9236214279472166,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041504980567911453,
      "loss": 5.284,
      "mean_token_accuracy": 0.20583022236824036,
      "num_tokens": 67793168.0,
      "step": 33455
    },
    {
      "entropy": 6.017706441879272,
      "epoch": 2.9240583762999215,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041502541564553336,
      "loss": 5.2989,
      "mean_token_accuracy": 0.20088891983032225,
      "num_tokens": 67803317.0,
      "step": 33460
    },
    {
      "entropy": 5.985110855102539,
      "epoch": 2.924495324652626,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041500102292619135,
      "loss": 5.237,
      "mean_token_accuracy": 0.20245713889598846,
      "num_tokens": 67812594.0,
      "step": 33465
    },
    {
      "entropy": 6.009873342514038,
      "epoch": 2.924932273005331,
      "grad_norm": 1.34375,
      "learning_rate": 0.00041497662752155637,
      "loss": 5.204,
      "mean_token_accuracy": 0.2088009774684906,
      "num_tokens": 67822277.0,
      "step": 33470
    },
    {
      "entropy": 6.049144554138183,
      "epoch": 2.9253692213580353,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004149522294320963,
      "loss": 5.2617,
      "mean_token_accuracy": 0.20527509301900865,
      "num_tokens": 67832948.0,
      "step": 33475
    },
    {
      "entropy": 5.9757496356964115,
      "epoch": 2.9258061697107403,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041492782865827914,
      "loss": 5.1933,
      "mean_token_accuracy": 0.21052034050226212,
      "num_tokens": 67843965.0,
      "step": 33480
    },
    {
      "entropy": 5.956845426559449,
      "epoch": 2.9262431180634447,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041490342520057293,
      "loss": 5.2571,
      "mean_token_accuracy": 0.21082605570554733,
      "num_tokens": 67854079.0,
      "step": 33485
    },
    {
      "entropy": 6.017020797729492,
      "epoch": 2.9266800664161496,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041487901905944575,
      "loss": 5.2477,
      "mean_token_accuracy": 0.2122163325548172,
      "num_tokens": 67865218.0,
      "step": 33490
    },
    {
      "entropy": 6.01182370185852,
      "epoch": 2.9271170147688546,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041485461023536564,
      "loss": 5.1796,
      "mean_token_accuracy": 0.2215782016515732,
      "num_tokens": 67875229.0,
      "step": 33495
    },
    {
      "entropy": 6.037134504318237,
      "epoch": 2.927553963121559,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004148301987288008,
      "loss": 5.332,
      "mean_token_accuracy": 0.20255241692066192,
      "num_tokens": 67885795.0,
      "step": 33500
    },
    {
      "entropy": 5.992516088485718,
      "epoch": 2.9279909114742635,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004148057845402195,
      "loss": 5.234,
      "mean_token_accuracy": 0.21157575845718385,
      "num_tokens": 67896668.0,
      "step": 33505
    },
    {
      "entropy": 5.985220718383789,
      "epoch": 2.9284278598269684,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041478136767009004,
      "loss": 5.1993,
      "mean_token_accuracy": 0.21111261695623398,
      "num_tokens": 67907082.0,
      "step": 33510
    },
    {
      "entropy": 6.012921094894409,
      "epoch": 2.9288648081796733,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004147569481188806,
      "loss": 5.2726,
      "mean_token_accuracy": 0.20676218271255492,
      "num_tokens": 67917517.0,
      "step": 33515
    },
    {
      "entropy": 6.082888412475586,
      "epoch": 2.929301756532378,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041473252588705977,
      "loss": 5.3442,
      "mean_token_accuracy": 0.19882056266069412,
      "num_tokens": 67927332.0,
      "step": 33520
    },
    {
      "entropy": 6.0332036972045895,
      "epoch": 2.9297387048850827,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00041470810097509587,
      "loss": 5.2752,
      "mean_token_accuracy": 0.20374558120965958,
      "num_tokens": 67936939.0,
      "step": 33525
    },
    {
      "entropy": 5.956047534942627,
      "epoch": 2.930175653237787,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004146836733834573,
      "loss": 5.2552,
      "mean_token_accuracy": 0.21351383924484252,
      "num_tokens": 67947553.0,
      "step": 33530
    },
    {
      "entropy": 5.9642175197601315,
      "epoch": 2.930612601590492,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004146592431126128,
      "loss": 5.216,
      "mean_token_accuracy": 0.20653652548789977,
      "num_tokens": 67958489.0,
      "step": 33535
    },
    {
      "entropy": 6.042495822906494,
      "epoch": 2.9310495499431966,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004146348101630307,
      "loss": 5.1867,
      "mean_token_accuracy": 0.21158226877450942,
      "num_tokens": 67967305.0,
      "step": 33540
    },
    {
      "entropy": 6.002935886383057,
      "epoch": 2.9314864982959015,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004146103745351799,
      "loss": 5.2203,
      "mean_token_accuracy": 0.20415178388357164,
      "num_tokens": 67977400.0,
      "step": 33545
    },
    {
      "entropy": 5.971988725662231,
      "epoch": 2.9319234466486064,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004145859362295289,
      "loss": 5.2399,
      "mean_token_accuracy": 0.2089586988091469,
      "num_tokens": 67987275.0,
      "step": 33550
    },
    {
      "entropy": 6.116434955596924,
      "epoch": 2.932360395001311,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004145614952465465,
      "loss": 5.4028,
      "mean_token_accuracy": 0.19700236320495607,
      "num_tokens": 67999002.0,
      "step": 33555
    },
    {
      "entropy": 6.052365636825561,
      "epoch": 2.9327973433540153,
      "grad_norm": 1.328125,
      "learning_rate": 0.00041453705158670147,
      "loss": 5.2089,
      "mean_token_accuracy": 0.21424809694290162,
      "num_tokens": 68008504.0,
      "step": 33560
    },
    {
      "entropy": 5.994502210617066,
      "epoch": 2.9332342917067202,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041451260525046267,
      "loss": 5.2423,
      "mean_token_accuracy": 0.20627751797437668,
      "num_tokens": 68019522.0,
      "step": 33565
    },
    {
      "entropy": 6.006840562820434,
      "epoch": 2.933671240059425,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041448815623829907,
      "loss": 5.2376,
      "mean_token_accuracy": 0.2047195628285408,
      "num_tokens": 68029991.0,
      "step": 33570
    },
    {
      "entropy": 6.014827299118042,
      "epoch": 2.9341081884121296,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004144637045506795,
      "loss": 5.2957,
      "mean_token_accuracy": 0.20081343203783036,
      "num_tokens": 68040347.0,
      "step": 33575
    },
    {
      "entropy": 5.997933769226075,
      "epoch": 2.9345451367648345,
      "grad_norm": 1.140625,
      "learning_rate": 0.000414439250188073,
      "loss": 5.2336,
      "mean_token_accuracy": 0.21397892236709595,
      "num_tokens": 68050932.0,
      "step": 33580
    },
    {
      "entropy": 6.079803371429444,
      "epoch": 2.934982085117539,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004144147931509486,
      "loss": 5.2659,
      "mean_token_accuracy": 0.202189801633358,
      "num_tokens": 68061104.0,
      "step": 33585
    },
    {
      "entropy": 5.9991168022155765,
      "epoch": 2.935419033470244,
      "grad_norm": 1.46875,
      "learning_rate": 0.00041439033343977546,
      "loss": 5.2468,
      "mean_token_accuracy": 0.21102908551692962,
      "num_tokens": 68070374.0,
      "step": 33590
    },
    {
      "entropy": 6.012089443206787,
      "epoch": 2.9358559818229484,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041436587105502263,
      "loss": 5.2706,
      "mean_token_accuracy": 0.20141672044992448,
      "num_tokens": 68079996.0,
      "step": 33595
    },
    {
      "entropy": 6.013753604888916,
      "epoch": 2.9362929301756533,
      "grad_norm": 1.25,
      "learning_rate": 0.0004143414059971594,
      "loss": 5.2789,
      "mean_token_accuracy": 0.19848434925079345,
      "num_tokens": 68090322.0,
      "step": 33600
    },
    {
      "entropy": 5.973025798797607,
      "epoch": 2.936729878528358,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000414316938266655,
      "loss": 5.1436,
      "mean_token_accuracy": 0.20888953655958176,
      "num_tokens": 68099482.0,
      "step": 33605
    },
    {
      "entropy": 5.96286301612854,
      "epoch": 2.9371668268810627,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004142924678639787,
      "loss": 5.2373,
      "mean_token_accuracy": 0.20959723144769668,
      "num_tokens": 68109852.0,
      "step": 33610
    },
    {
      "entropy": 6.00406813621521,
      "epoch": 2.937603775233767,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041426799478959993,
      "loss": 5.2645,
      "mean_token_accuracy": 0.20527869313955308,
      "num_tokens": 68118721.0,
      "step": 33615
    },
    {
      "entropy": 6.038699007034301,
      "epoch": 2.938040723586472,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004142435190439879,
      "loss": 5.2179,
      "mean_token_accuracy": 0.2091053009033203,
      "num_tokens": 68129080.0,
      "step": 33620
    },
    {
      "entropy": 6.045088291168213,
      "epoch": 2.938477671939177,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041421904062761234,
      "loss": 5.2737,
      "mean_token_accuracy": 0.2045389711856842,
      "num_tokens": 68138154.0,
      "step": 33625
    },
    {
      "entropy": 6.019199228286743,
      "epoch": 2.9389146202918814,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004141945595409426,
      "loss": 5.3419,
      "mean_token_accuracy": 0.2015029862523079,
      "num_tokens": 68148736.0,
      "step": 33630
    },
    {
      "entropy": 5.982270526885986,
      "epoch": 2.9393515686445864,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004141700757844483,
      "loss": 5.2169,
      "mean_token_accuracy": 0.20551124066114426,
      "num_tokens": 68159041.0,
      "step": 33635
    },
    {
      "entropy": 6.045693588256836,
      "epoch": 2.939788516997291,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000414145589358599,
      "loss": 5.2622,
      "mean_token_accuracy": 0.2048383131623268,
      "num_tokens": 68168715.0,
      "step": 33640
    },
    {
      "entropy": 6.013275957107544,
      "epoch": 2.9402254653499957,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041412110026386436,
      "loss": 5.2666,
      "mean_token_accuracy": 0.20477730482816697,
      "num_tokens": 68178863.0,
      "step": 33645
    },
    {
      "entropy": 6.020035457611084,
      "epoch": 2.9406624137027,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004140966085007142,
      "loss": 5.2912,
      "mean_token_accuracy": 0.2009050577878952,
      "num_tokens": 68190131.0,
      "step": 33650
    },
    {
      "entropy": 6.001848268508911,
      "epoch": 2.941099362055405,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041407211406961815,
      "loss": 5.2366,
      "mean_token_accuracy": 0.21126120090484618,
      "num_tokens": 68200193.0,
      "step": 33655
    },
    {
      "entropy": 6.0120837688446045,
      "epoch": 2.94153631040811,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041404761697104605,
      "loss": 5.2292,
      "mean_token_accuracy": 0.20380585491657258,
      "num_tokens": 68210714.0,
      "step": 33660
    },
    {
      "entropy": 6.025084590911865,
      "epoch": 2.9419732587608145,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004140231172054678,
      "loss": 5.2758,
      "mean_token_accuracy": 0.20594558566808702,
      "num_tokens": 68220809.0,
      "step": 33665
    },
    {
      "entropy": 6.076620101928711,
      "epoch": 2.942410207113519,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004139986147733534,
      "loss": 5.2777,
      "mean_token_accuracy": 0.20476260781288147,
      "num_tokens": 68229887.0,
      "step": 33670
    },
    {
      "entropy": 6.01701865196228,
      "epoch": 2.942847155466224,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041397410967517265,
      "loss": 5.2264,
      "mean_token_accuracy": 0.2066827416419983,
      "num_tokens": 68240491.0,
      "step": 33675
    },
    {
      "entropy": 5.995053052902222,
      "epoch": 2.943284103818929,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041394960191139566,
      "loss": 5.3446,
      "mean_token_accuracy": 0.19916209429502488,
      "num_tokens": 68250302.0,
      "step": 33680
    },
    {
      "entropy": 6.028770017623901,
      "epoch": 2.9437210521716333,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004139250914824925,
      "loss": 5.1827,
      "mean_token_accuracy": 0.20761746019124985,
      "num_tokens": 68259245.0,
      "step": 33685
    },
    {
      "entropy": 6.0023092269897464,
      "epoch": 2.944158000524338,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041390057838893335,
      "loss": 5.1978,
      "mean_token_accuracy": 0.2121851161122322,
      "num_tokens": 68269096.0,
      "step": 33690
    },
    {
      "entropy": 6.026428937911987,
      "epoch": 2.9445949488770427,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041387606263118825,
      "loss": 5.2569,
      "mean_token_accuracy": 0.20656732469797134,
      "num_tokens": 68278949.0,
      "step": 33695
    },
    {
      "entropy": 5.949312162399292,
      "epoch": 2.9450318972297476,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004138515442097275,
      "loss": 5.1874,
      "mean_token_accuracy": 0.21111273616552353,
      "num_tokens": 68288526.0,
      "step": 33700
    },
    {
      "entropy": 6.018980979919434,
      "epoch": 2.945468845582452,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004138270231250214,
      "loss": 5.184,
      "mean_token_accuracy": 0.2095971077680588,
      "num_tokens": 68297997.0,
      "step": 33705
    },
    {
      "entropy": 5.983213996887207,
      "epoch": 2.945905793935157,
      "grad_norm": 1.25,
      "learning_rate": 0.0004138024993775403,
      "loss": 5.3197,
      "mean_token_accuracy": 0.20274700671434404,
      "num_tokens": 68307316.0,
      "step": 33710
    },
    {
      "entropy": 5.953754711151123,
      "epoch": 2.9463427422878614,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004137779729677544,
      "loss": 5.0846,
      "mean_token_accuracy": 0.2235663890838623,
      "num_tokens": 68317124.0,
      "step": 33715
    },
    {
      "entropy": 6.009392213821411,
      "epoch": 2.9467796906405663,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004137534438961344,
      "loss": 5.2469,
      "mean_token_accuracy": 0.20514061599969863,
      "num_tokens": 68326576.0,
      "step": 33720
    },
    {
      "entropy": 5.991183423995972,
      "epoch": 2.947216638993271,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041372891216315055,
      "loss": 5.2231,
      "mean_token_accuracy": 0.21076448559761046,
      "num_tokens": 68336999.0,
      "step": 33725
    },
    {
      "entropy": 6.041845989227295,
      "epoch": 2.9476535873459757,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041370437776927347,
      "loss": 5.2306,
      "mean_token_accuracy": 0.21018911451101302,
      "num_tokens": 68347084.0,
      "step": 33730
    },
    {
      "entropy": 6.0040806293487545,
      "epoch": 2.9480905356986806,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004136798407149738,
      "loss": 5.2328,
      "mean_token_accuracy": 0.2066108524799347,
      "num_tokens": 68357136.0,
      "step": 33735
    },
    {
      "entropy": 6.062993717193604,
      "epoch": 2.948527484051385,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041365530100072215,
      "loss": 5.3243,
      "mean_token_accuracy": 0.19896628260612487,
      "num_tokens": 68367020.0,
      "step": 33740
    },
    {
      "entropy": 6.024948835372925,
      "epoch": 2.9489644324040896,
      "grad_norm": 1.25,
      "learning_rate": 0.0004136307586269891,
      "loss": 5.2283,
      "mean_token_accuracy": 0.21261949390172957,
      "num_tokens": 68378436.0,
      "step": 33745
    },
    {
      "entropy": 5.995826625823975,
      "epoch": 2.9494013807567945,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041360621359424544,
      "loss": 5.1955,
      "mean_token_accuracy": 0.2071181133389473,
      "num_tokens": 68387777.0,
      "step": 33750
    },
    {
      "entropy": 5.960070991516114,
      "epoch": 2.9498383291094994,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041358166590296205,
      "loss": 5.2363,
      "mean_token_accuracy": 0.20598842203617096,
      "num_tokens": 68397648.0,
      "step": 33755
    },
    {
      "entropy": 5.9885331153869625,
      "epoch": 2.950275277462204,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041355711555360965,
      "loss": 5.3757,
      "mean_token_accuracy": 0.19784369468688964,
      "num_tokens": 68409539.0,
      "step": 33760
    },
    {
      "entropy": 6.048188257217407,
      "epoch": 2.9507122258149088,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041353256254665914,
      "loss": 5.2973,
      "mean_token_accuracy": 0.20291729569435119,
      "num_tokens": 68419626.0,
      "step": 33765
    },
    {
      "entropy": 6.04543285369873,
      "epoch": 2.9511491741676132,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041350800688258157,
      "loss": 5.2183,
      "mean_token_accuracy": 0.21193269938230513,
      "num_tokens": 68429289.0,
      "step": 33770
    },
    {
      "entropy": 6.035431337356568,
      "epoch": 2.951586122520318,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041348344856184785,
      "loss": 5.266,
      "mean_token_accuracy": 0.20717159956693648,
      "num_tokens": 68440186.0,
      "step": 33775
    },
    {
      "entropy": 6.010252189636231,
      "epoch": 2.9520230708730226,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041345888758492896,
      "loss": 5.2112,
      "mean_token_accuracy": 0.21144990921020507,
      "num_tokens": 68450292.0,
      "step": 33780
    },
    {
      "entropy": 6.031027555465698,
      "epoch": 2.9524600192257275,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041343432395229605,
      "loss": 5.2598,
      "mean_token_accuracy": 0.20555216819047928,
      "num_tokens": 68460419.0,
      "step": 33785
    },
    {
      "entropy": 5.972742748260498,
      "epoch": 2.9528969675784325,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004134097576644203,
      "loss": 5.2319,
      "mean_token_accuracy": 0.2092115804553032,
      "num_tokens": 68470488.0,
      "step": 33790
    },
    {
      "entropy": 6.008285188674927,
      "epoch": 2.953333915931137,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041338518872177286,
      "loss": 5.2664,
      "mean_token_accuracy": 0.20344794392585755,
      "num_tokens": 68479872.0,
      "step": 33795
    },
    {
      "entropy": 6.044109106063843,
      "epoch": 2.9537708642838414,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041336061712482497,
      "loss": 5.3067,
      "mean_token_accuracy": 0.19994090348482133,
      "num_tokens": 68491086.0,
      "step": 33800
    },
    {
      "entropy": 6.078249788284301,
      "epoch": 2.9542078126365463,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041333604287404805,
      "loss": 5.3258,
      "mean_token_accuracy": 0.1951776623725891,
      "num_tokens": 68501136.0,
      "step": 33805
    },
    {
      "entropy": 5.983725690841675,
      "epoch": 2.954644760989251,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004133114659699132,
      "loss": 5.2061,
      "mean_token_accuracy": 0.2094160720705986,
      "num_tokens": 68511579.0,
      "step": 33810
    },
    {
      "entropy": 5.993789339065552,
      "epoch": 2.9550817093419557,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041328688641289205,
      "loss": 5.2289,
      "mean_token_accuracy": 0.21123905181884767,
      "num_tokens": 68521716.0,
      "step": 33815
    },
    {
      "entropy": 6.017913055419922,
      "epoch": 2.9555186576946606,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041326230420345584,
      "loss": 5.2832,
      "mean_token_accuracy": 0.2080348923802376,
      "num_tokens": 68531814.0,
      "step": 33820
    },
    {
      "entropy": 5.951377105712891,
      "epoch": 2.955955606047365,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004132377193420763,
      "loss": 5.1869,
      "mean_token_accuracy": 0.21552499681711196,
      "num_tokens": 68541388.0,
      "step": 33825
    },
    {
      "entropy": 6.0375124454498295,
      "epoch": 2.95639255440007,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041321313182922484,
      "loss": 5.3629,
      "mean_token_accuracy": 0.19066479355096816,
      "num_tokens": 68552259.0,
      "step": 33830
    },
    {
      "entropy": 6.03316445350647,
      "epoch": 2.9568295027527745,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004131885416653731,
      "loss": 5.2198,
      "mean_token_accuracy": 0.21144597083330155,
      "num_tokens": 68562001.0,
      "step": 33835
    },
    {
      "entropy": 6.059235048294068,
      "epoch": 2.9572664511054794,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004131639488509927,
      "loss": 5.2271,
      "mean_token_accuracy": 0.20770546197891235,
      "num_tokens": 68571636.0,
      "step": 33840
    },
    {
      "entropy": 6.051633691787719,
      "epoch": 2.9577033994581843,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004131393533865554,
      "loss": 5.368,
      "mean_token_accuracy": 0.19412940740585327,
      "num_tokens": 68581448.0,
      "step": 33845
    },
    {
      "entropy": 6.087592744827271,
      "epoch": 2.9581403478108887,
      "grad_norm": 1.25,
      "learning_rate": 0.00041311475527253284,
      "loss": 5.3556,
      "mean_token_accuracy": 0.19898512810468674,
      "num_tokens": 68591880.0,
      "step": 33850
    },
    {
      "entropy": 6.105361270904541,
      "epoch": 2.958577296163593,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041309015450939694,
      "loss": 5.3255,
      "mean_token_accuracy": 0.2030830442905426,
      "num_tokens": 68603025.0,
      "step": 33855
    },
    {
      "entropy": 6.085260200500488,
      "epoch": 2.959014244516298,
      "grad_norm": 1.25,
      "learning_rate": 0.0004130655510976196,
      "loss": 5.3337,
      "mean_token_accuracy": 0.19896250665187837,
      "num_tokens": 68613580.0,
      "step": 33860
    },
    {
      "entropy": 6.042206335067749,
      "epoch": 2.959451192869003,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041304094503767256,
      "loss": 5.3257,
      "mean_token_accuracy": 0.20395378023386002,
      "num_tokens": 68624451.0,
      "step": 33865
    },
    {
      "entropy": 5.971810293197632,
      "epoch": 2.9598881412217075,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00041301633633002784,
      "loss": 5.2413,
      "mean_token_accuracy": 0.20841215848922728,
      "num_tokens": 68634408.0,
      "step": 33870
    },
    {
      "entropy": 5.958119440078735,
      "epoch": 2.9603250895744124,
      "grad_norm": 1.25,
      "learning_rate": 0.0004129917249751575,
      "loss": 5.2285,
      "mean_token_accuracy": 0.2078595742583275,
      "num_tokens": 68643795.0,
      "step": 33875
    },
    {
      "entropy": 6.008126258850098,
      "epoch": 2.960762037927117,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004129671109735336,
      "loss": 5.2085,
      "mean_token_accuracy": 0.20780719220638275,
      "num_tokens": 68653669.0,
      "step": 33880
    },
    {
      "entropy": 6.096936225891113,
      "epoch": 2.961198986279822,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041294249432562825,
      "loss": 5.3135,
      "mean_token_accuracy": 0.19756195247173308,
      "num_tokens": 68663094.0,
      "step": 33885
    },
    {
      "entropy": 6.001310205459594,
      "epoch": 2.9616359346325263,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00041291787503191347,
      "loss": 5.1911,
      "mean_token_accuracy": 0.21617559492588043,
      "num_tokens": 68673561.0,
      "step": 33890
    },
    {
      "entropy": 6.0517919063568115,
      "epoch": 2.962072882985231,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041289325309286167,
      "loss": 5.3346,
      "mean_token_accuracy": 0.1974238023161888,
      "num_tokens": 68683624.0,
      "step": 33895
    },
    {
      "entropy": 6.028034400939942,
      "epoch": 2.962509831337936,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041286862850894497,
      "loss": 5.2675,
      "mean_token_accuracy": 0.21045215874910356,
      "num_tokens": 68693805.0,
      "step": 33900
    },
    {
      "entropy": 6.054689025878906,
      "epoch": 2.9629467796906406,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004128440012806357,
      "loss": 5.3038,
      "mean_token_accuracy": 0.2059404045343399,
      "num_tokens": 68704020.0,
      "step": 33905
    },
    {
      "entropy": 6.029179191589355,
      "epoch": 2.963383728043345,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041281937140840634,
      "loss": 5.2589,
      "mean_token_accuracy": 0.20256270617246627,
      "num_tokens": 68714486.0,
      "step": 33910
    },
    {
      "entropy": 6.0339030742645265,
      "epoch": 2.96382067639605,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004127947388927292,
      "loss": 5.2088,
      "mean_token_accuracy": 0.211198627948761,
      "num_tokens": 68724759.0,
      "step": 33915
    },
    {
      "entropy": 5.961906242370605,
      "epoch": 2.964257624748755,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004127701037340767,
      "loss": 5.1912,
      "mean_token_accuracy": 0.21710798889398575,
      "num_tokens": 68734826.0,
      "step": 33920
    },
    {
      "entropy": 5.986735439300537,
      "epoch": 2.9646945731014593,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041274546593292143,
      "loss": 5.282,
      "mean_token_accuracy": 0.20269432812929153,
      "num_tokens": 68744786.0,
      "step": 33925
    },
    {
      "entropy": 6.065600061416626,
      "epoch": 2.9651315214541643,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004127208254897361,
      "loss": 5.3181,
      "mean_token_accuracy": 0.19696001410484315,
      "num_tokens": 68754751.0,
      "step": 33930
    },
    {
      "entropy": 6.040071964263916,
      "epoch": 2.9655684698068687,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041269618240499296,
      "loss": 5.2477,
      "mean_token_accuracy": 0.2046002611517906,
      "num_tokens": 68765652.0,
      "step": 33935
    },
    {
      "entropy": 6.021200656890869,
      "epoch": 2.9660054181595736,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041267153667916506,
      "loss": 5.2283,
      "mean_token_accuracy": 0.20971952378749847,
      "num_tokens": 68775587.0,
      "step": 33940
    },
    {
      "entropy": 6.040398073196411,
      "epoch": 2.966442366512278,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041264688831272483,
      "loss": 5.2468,
      "mean_token_accuracy": 0.20515124946832658,
      "num_tokens": 68786678.0,
      "step": 33945
    },
    {
      "entropy": 5.960202693939209,
      "epoch": 2.966879314864983,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004126222373061452,
      "loss": 5.1458,
      "mean_token_accuracy": 0.21740300804376603,
      "num_tokens": 68796108.0,
      "step": 33950
    },
    {
      "entropy": 5.971696567535401,
      "epoch": 2.967316263217688,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041259758365989896,
      "loss": 5.226,
      "mean_token_accuracy": 0.20649247914552687,
      "num_tokens": 68806364.0,
      "step": 33955
    },
    {
      "entropy": 6.045690822601318,
      "epoch": 2.9677532115703924,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004125729273744589,
      "loss": 5.2816,
      "mean_token_accuracy": 0.20205494463443757,
      "num_tokens": 68816099.0,
      "step": 33960
    },
    {
      "entropy": 6.01018214225769,
      "epoch": 2.968190159923097,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041254826845029806,
      "loss": 5.3166,
      "mean_token_accuracy": 0.19679210782051088,
      "num_tokens": 68826480.0,
      "step": 33965
    },
    {
      "entropy": 5.986596536636353,
      "epoch": 2.968627108275802,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041252360688788937,
      "loss": 5.2334,
      "mean_token_accuracy": 0.20977362096309662,
      "num_tokens": 68835657.0,
      "step": 33970
    },
    {
      "entropy": 6.05064058303833,
      "epoch": 2.9690640566285067,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004124989426877058,
      "loss": 5.288,
      "mean_token_accuracy": 0.2026844084262848,
      "num_tokens": 68845115.0,
      "step": 33975
    },
    {
      "entropy": 6.0535258769989015,
      "epoch": 2.969501004981211,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004124742758502205,
      "loss": 5.2649,
      "mean_token_accuracy": 0.20844239741563797,
      "num_tokens": 68854562.0,
      "step": 33980
    },
    {
      "entropy": 6.0686229228973385,
      "epoch": 2.969937953333916,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004124496063759066,
      "loss": 5.324,
      "mean_token_accuracy": 0.2014366164803505,
      "num_tokens": 68864787.0,
      "step": 33985
    },
    {
      "entropy": 5.960636234283447,
      "epoch": 2.9703749016866206,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004124249342652371,
      "loss": 5.2365,
      "mean_token_accuracy": 0.20812607407569886,
      "num_tokens": 68874518.0,
      "step": 33990
    },
    {
      "entropy": 6.061416149139404,
      "epoch": 2.9708118500393255,
      "grad_norm": 1.390625,
      "learning_rate": 0.00041240025951868545,
      "loss": 5.2143,
      "mean_token_accuracy": 0.20507186353206636,
      "num_tokens": 68883765.0,
      "step": 33995
    },
    {
      "entropy": 6.044732666015625,
      "epoch": 2.97124879839203,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041237558213672477,
      "loss": 5.2736,
      "mean_token_accuracy": 0.20273139625787734,
      "num_tokens": 68894191.0,
      "step": 34000
    },
    {
      "entropy": 6.0214629650115965,
      "epoch": 2.971685746744735,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041235090211982846,
      "loss": 5.2398,
      "mean_token_accuracy": 0.2048561915755272,
      "num_tokens": 68905136.0,
      "step": 34005
    },
    {
      "entropy": 6.003361082077026,
      "epoch": 2.9721226950974398,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004123262194684699,
      "loss": 5.2385,
      "mean_token_accuracy": 0.2099767342209816,
      "num_tokens": 68914620.0,
      "step": 34010
    },
    {
      "entropy": 6.030956220626831,
      "epoch": 2.9725596434501442,
      "grad_norm": 1.3125,
      "learning_rate": 0.00041230153418312244,
      "loss": 5.2386,
      "mean_token_accuracy": 0.2072659030556679,
      "num_tokens": 68923781.0,
      "step": 34015
    },
    {
      "entropy": 5.980956935882569,
      "epoch": 2.9729965918028487,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004122768462642596,
      "loss": 5.2977,
      "mean_token_accuracy": 0.20483383238315583,
      "num_tokens": 68934283.0,
      "step": 34020
    },
    {
      "entropy": 5.970917177200318,
      "epoch": 2.9734335401555536,
      "grad_norm": 1.1875,
      "learning_rate": 0.000412252155712355,
      "loss": 5.2041,
      "mean_token_accuracy": 0.21056757271289825,
      "num_tokens": 68943799.0,
      "step": 34025
    },
    {
      "entropy": 6.074224901199341,
      "epoch": 2.9738704885082585,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041222746252788213,
      "loss": 5.2834,
      "mean_token_accuracy": 0.20325959771871566,
      "num_tokens": 68953733.0,
      "step": 34030
    },
    {
      "entropy": 6.0292994499206545,
      "epoch": 2.974307436860963,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004122027667113146,
      "loss": 5.2941,
      "mean_token_accuracy": 0.1992412120103836,
      "num_tokens": 68963888.0,
      "step": 34035
    },
    {
      "entropy": 6.091340637207031,
      "epoch": 2.974744385213668,
      "grad_norm": 1.203125,
      "learning_rate": 0.000412178068263126,
      "loss": 5.3025,
      "mean_token_accuracy": 0.1987645909190178,
      "num_tokens": 68974652.0,
      "step": 34040
    },
    {
      "entropy": 6.07080340385437,
      "epoch": 2.9751813335663724,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004121533671837903,
      "loss": 5.2057,
      "mean_token_accuracy": 0.20880695879459382,
      "num_tokens": 68984583.0,
      "step": 34045
    },
    {
      "entropy": 5.972291135787964,
      "epoch": 2.9756182819190773,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004121286634737811,
      "loss": 5.3003,
      "mean_token_accuracy": 0.20376747399568557,
      "num_tokens": 68995790.0,
      "step": 34050
    },
    {
      "entropy": 6.027967643737793,
      "epoch": 2.9760552302717818,
      "grad_norm": 1.125,
      "learning_rate": 0.0004121039571335723,
      "loss": 5.32,
      "mean_token_accuracy": 0.19969379901885986,
      "num_tokens": 69006695.0,
      "step": 34055
    },
    {
      "entropy": 6.01713228225708,
      "epoch": 2.9764921786244867,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004120792481636378,
      "loss": 5.1921,
      "mean_token_accuracy": 0.220496267080307,
      "num_tokens": 69017504.0,
      "step": 34060
    },
    {
      "entropy": 6.0317127227783205,
      "epoch": 2.976929126977191,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004120545365644515,
      "loss": 5.354,
      "mean_token_accuracy": 0.19614437222480774,
      "num_tokens": 69026938.0,
      "step": 34065
    },
    {
      "entropy": 6.005792093276978,
      "epoch": 2.977366075329896,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004120298223364872,
      "loss": 5.2301,
      "mean_token_accuracy": 0.2099548652768135,
      "num_tokens": 69037404.0,
      "step": 34070
    },
    {
      "entropy": 6.038941383361816,
      "epoch": 2.9778030236826005,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041200510548021935,
      "loss": 5.2241,
      "mean_token_accuracy": 0.2087363675236702,
      "num_tokens": 69047507.0,
      "step": 34075
    },
    {
      "entropy": 5.992018842697144,
      "epoch": 2.9782399720353054,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041198038599612156,
      "loss": 5.2316,
      "mean_token_accuracy": 0.2107149913907051,
      "num_tokens": 69057626.0,
      "step": 34080
    },
    {
      "entropy": 6.0447601795196535,
      "epoch": 2.9786769203880104,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041195566388466836,
      "loss": 5.29,
      "mean_token_accuracy": 0.20510124117136003,
      "num_tokens": 69068142.0,
      "step": 34085
    },
    {
      "entropy": 6.07082724571228,
      "epoch": 2.979113868740715,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041193093914633364,
      "loss": 5.2977,
      "mean_token_accuracy": 0.20140518993139267,
      "num_tokens": 69077935.0,
      "step": 34090
    },
    {
      "entropy": 6.0274255752563475,
      "epoch": 2.9795508170934193,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004119062117815918,
      "loss": 5.3209,
      "mean_token_accuracy": 0.19755046516656877,
      "num_tokens": 69087397.0,
      "step": 34095
    },
    {
      "entropy": 6.053063201904297,
      "epoch": 2.979987765446124,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004118814817909171,
      "loss": 5.3378,
      "mean_token_accuracy": 0.20713470727205277,
      "num_tokens": 69097666.0,
      "step": 34100
    },
    {
      "entropy": 6.029000282287598,
      "epoch": 2.980424713798829,
      "grad_norm": 1.25,
      "learning_rate": 0.0004118567491747839,
      "loss": 5.2777,
      "mean_token_accuracy": 0.2008195623755455,
      "num_tokens": 69107516.0,
      "step": 34105
    },
    {
      "entropy": 6.030483961105347,
      "epoch": 2.9808616621515336,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004118320139336665,
      "loss": 5.2404,
      "mean_token_accuracy": 0.20853616446256637,
      "num_tokens": 69117462.0,
      "step": 34110
    },
    {
      "entropy": 6.044802904129028,
      "epoch": 2.9812986105042385,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004118072760680393,
      "loss": 5.2179,
      "mean_token_accuracy": 0.21511198580265045,
      "num_tokens": 69128171.0,
      "step": 34115
    },
    {
      "entropy": 5.99122052192688,
      "epoch": 2.981735558856943,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041178253557837697,
      "loss": 5.2615,
      "mean_token_accuracy": 0.20669524222612382,
      "num_tokens": 69138169.0,
      "step": 34120
    },
    {
      "entropy": 6.017171907424927,
      "epoch": 2.982172507209648,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041175779246515384,
      "loss": 5.2157,
      "mean_token_accuracy": 0.21101900637149812,
      "num_tokens": 69147518.0,
      "step": 34125
    },
    {
      "entropy": 6.050429105758667,
      "epoch": 2.9826094555623524,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004117330467288447,
      "loss": 5.2115,
      "mean_token_accuracy": 0.20436621755361556,
      "num_tokens": 69157049.0,
      "step": 34130
    },
    {
      "entropy": 5.984636974334717,
      "epoch": 2.9830464039150573,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041170829836992403,
      "loss": 5.2474,
      "mean_token_accuracy": 0.20895660519599915,
      "num_tokens": 69166502.0,
      "step": 34135
    },
    {
      "entropy": 6.019475650787354,
      "epoch": 2.983483352267762,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004116835473888665,
      "loss": 5.2127,
      "mean_token_accuracy": 0.21221507638692855,
      "num_tokens": 69176566.0,
      "step": 34140
    },
    {
      "entropy": 6.0594813346862795,
      "epoch": 2.9839203006204666,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004116587937861469,
      "loss": 5.2489,
      "mean_token_accuracy": 0.20049700736999512,
      "num_tokens": 69186677.0,
      "step": 34145
    },
    {
      "entropy": 6.019785165786743,
      "epoch": 2.984357248973171,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004116340375622402,
      "loss": 5.3348,
      "mean_token_accuracy": 0.19809505790472032,
      "num_tokens": 69197595.0,
      "step": 34150
    },
    {
      "entropy": 5.992626333236695,
      "epoch": 2.984794197325876,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004116092787176209,
      "loss": 5.2552,
      "mean_token_accuracy": 0.21301408261060714,
      "num_tokens": 69207384.0,
      "step": 34155
    },
    {
      "entropy": 5.999223136901856,
      "epoch": 2.985231145678581,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041158451725276405,
      "loss": 5.1612,
      "mean_token_accuracy": 0.22049519121646882,
      "num_tokens": 69217019.0,
      "step": 34160
    },
    {
      "entropy": 6.097411060333252,
      "epoch": 2.9856680940312854,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004115597531681447,
      "loss": 5.4002,
      "mean_token_accuracy": 0.1889390230178833,
      "num_tokens": 69229076.0,
      "step": 34165
    },
    {
      "entropy": 6.028207397460937,
      "epoch": 2.9861050423839903,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041153498646423763,
      "loss": 5.3301,
      "mean_token_accuracy": 0.19936158657073974,
      "num_tokens": 69238771.0,
      "step": 34170
    },
    {
      "entropy": 6.05636682510376,
      "epoch": 2.986541990736695,
      "grad_norm": 1.28125,
      "learning_rate": 0.000411510217141518,
      "loss": 5.2651,
      "mean_token_accuracy": 0.20046894401311874,
      "num_tokens": 69248864.0,
      "step": 34175
    },
    {
      "entropy": 6.052912425994873,
      "epoch": 2.9869789390893997,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041148544520046087,
      "loss": 5.3186,
      "mean_token_accuracy": 0.19935702979564668,
      "num_tokens": 69259372.0,
      "step": 34180
    },
    {
      "entropy": 6.006982898712158,
      "epoch": 2.987415887442104,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004114606706415415,
      "loss": 5.2035,
      "mean_token_accuracy": 0.20998591035604477,
      "num_tokens": 69270079.0,
      "step": 34185
    },
    {
      "entropy": 6.024829626083374,
      "epoch": 2.987852835794809,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004114358934652349,
      "loss": 5.2364,
      "mean_token_accuracy": 0.211351278424263,
      "num_tokens": 69279909.0,
      "step": 34190
    },
    {
      "entropy": 5.958320617675781,
      "epoch": 2.988289784147514,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004114111136720163,
      "loss": 5.1896,
      "mean_token_accuracy": 0.21199559569358825,
      "num_tokens": 69289598.0,
      "step": 34195
    },
    {
      "entropy": 6.064268350601196,
      "epoch": 2.9887267325002185,
      "grad_norm": 1.125,
      "learning_rate": 0.0004113863312623611,
      "loss": 5.296,
      "mean_token_accuracy": 0.19962669014930726,
      "num_tokens": 69301305.0,
      "step": 34200
    },
    {
      "entropy": 6.006138610839844,
      "epoch": 2.989163680852923,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004113615462367446,
      "loss": 5.2239,
      "mean_token_accuracy": 0.21035700738430024,
      "num_tokens": 69310560.0,
      "step": 34205
    },
    {
      "entropy": 6.000448894500733,
      "epoch": 2.989600629205628,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041133675859564213,
      "loss": 5.2854,
      "mean_token_accuracy": 0.20378048866987228,
      "num_tokens": 69320286.0,
      "step": 34210
    },
    {
      "entropy": 5.994059276580811,
      "epoch": 2.9900375775583328,
      "grad_norm": 1.25,
      "learning_rate": 0.00041131196833952925,
      "loss": 5.2257,
      "mean_token_accuracy": 0.20200383365154267,
      "num_tokens": 69329895.0,
      "step": 34215
    },
    {
      "entropy": 6.085097932815552,
      "epoch": 2.9904745259110372,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004112871754688814,
      "loss": 5.2295,
      "mean_token_accuracy": 0.20513238608837128,
      "num_tokens": 69340122.0,
      "step": 34220
    },
    {
      "entropy": 6.027937984466552,
      "epoch": 2.990911474263742,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041126237998417414,
      "loss": 5.263,
      "mean_token_accuracy": 0.2066078618168831,
      "num_tokens": 69349354.0,
      "step": 34225
    },
    {
      "entropy": 5.999767017364502,
      "epoch": 2.9913484226164466,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00041123758188588296,
      "loss": 5.321,
      "mean_token_accuracy": 0.19565083384513854,
      "num_tokens": 69360236.0,
      "step": 34230
    },
    {
      "entropy": 6.029934644699097,
      "epoch": 2.9917853709691515,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041121278117448365,
      "loss": 5.3108,
      "mean_token_accuracy": 0.20476776212453843,
      "num_tokens": 69369362.0,
      "step": 34235
    },
    {
      "entropy": 6.0578563690185545,
      "epoch": 2.992222319321856,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041118797785045174,
      "loss": 5.2781,
      "mean_token_accuracy": 0.20159876942634583,
      "num_tokens": 69379403.0,
      "step": 34240
    },
    {
      "entropy": 5.98863844871521,
      "epoch": 2.992659267674561,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004111631719142631,
      "loss": 5.3011,
      "mean_token_accuracy": 0.20822513550519944,
      "num_tokens": 69388667.0,
      "step": 34245
    },
    {
      "entropy": 6.022905158996582,
      "epoch": 2.993096216027266,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041113836336639347,
      "loss": 5.2725,
      "mean_token_accuracy": 0.19825261384248732,
      "num_tokens": 69399552.0,
      "step": 34250
    },
    {
      "entropy": 6.030688095092773,
      "epoch": 2.9935331643799703,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004111135522073187,
      "loss": 5.1566,
      "mean_token_accuracy": 0.21031396090984344,
      "num_tokens": 69409323.0,
      "step": 34255
    },
    {
      "entropy": 6.091666984558105,
      "epoch": 2.9939701127326748,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004110887384375147,
      "loss": 5.2886,
      "mean_token_accuracy": 0.1979842409491539,
      "num_tokens": 69418696.0,
      "step": 34260
    },
    {
      "entropy": 6.052858495712281,
      "epoch": 2.9944070610853797,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004110639220574573,
      "loss": 5.3221,
      "mean_token_accuracy": 0.19320282638072966,
      "num_tokens": 69429470.0,
      "step": 34265
    },
    {
      "entropy": 5.973674297332764,
      "epoch": 2.9948440094380846,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041103910306762267,
      "loss": 5.2038,
      "mean_token_accuracy": 0.2092508405447006,
      "num_tokens": 69440636.0,
      "step": 34270
    },
    {
      "entropy": 6.080009651184082,
      "epoch": 2.995280957790789,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041101428146848667,
      "loss": 5.317,
      "mean_token_accuracy": 0.2035638391971588,
      "num_tokens": 69451819.0,
      "step": 34275
    },
    {
      "entropy": 6.077949571609497,
      "epoch": 2.995717906143494,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041098945726052556,
      "loss": 5.3511,
      "mean_token_accuracy": 0.20005041658878325,
      "num_tokens": 69462690.0,
      "step": 34280
    },
    {
      "entropy": 5.988423633575439,
      "epoch": 2.9961548544961984,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004109646304442154,
      "loss": 5.2182,
      "mean_token_accuracy": 0.20732527375221252,
      "num_tokens": 69473338.0,
      "step": 34285
    },
    {
      "entropy": 6.025283908843994,
      "epoch": 2.9965918028489034,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004109398010200323,
      "loss": 5.2594,
      "mean_token_accuracy": 0.20615433305501937,
      "num_tokens": 69482189.0,
      "step": 34290
    },
    {
      "entropy": 5.931045007705689,
      "epoch": 2.997028751201608,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004109149689884527,
      "loss": 5.2212,
      "mean_token_accuracy": 0.2130722790956497,
      "num_tokens": 69492084.0,
      "step": 34295
    },
    {
      "entropy": 6.006751203536988,
      "epoch": 2.9974656995543127,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041089013434995273,
      "loss": 5.2436,
      "mean_token_accuracy": 0.2065196618437767,
      "num_tokens": 69502886.0,
      "step": 34300
    },
    {
      "entropy": 6.074687719345093,
      "epoch": 2.9979026479070177,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004108652971050088,
      "loss": 5.2125,
      "mean_token_accuracy": 0.21003158241510392,
      "num_tokens": 69512670.0,
      "step": 34305
    },
    {
      "entropy": 6.000804090499878,
      "epoch": 2.998339596259722,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041084045725409726,
      "loss": 5.306,
      "mean_token_accuracy": 0.20090585052967072,
      "num_tokens": 69522744.0,
      "step": 34310
    },
    {
      "entropy": 5.969648790359497,
      "epoch": 2.9987765446124266,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004108156147976945,
      "loss": 5.1599,
      "mean_token_accuracy": 0.2137812376022339,
      "num_tokens": 69533195.0,
      "step": 34315
    },
    {
      "entropy": 6.027792263031006,
      "epoch": 2.9992134929651315,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004107907697362772,
      "loss": 5.3332,
      "mean_token_accuracy": 0.20077544450759888,
      "num_tokens": 69543058.0,
      "step": 34320
    },
    {
      "entropy": 6.068245697021484,
      "epoch": 2.9996504413178364,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004107659220703217,
      "loss": 5.2846,
      "mean_token_accuracy": 0.20309805870056152,
      "num_tokens": 69552428.0,
      "step": 34325
    },
    {
      "entropy": 6.059314107894897,
      "epoch": 3.000087389670541,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004107410718003047,
      "loss": 5.2374,
      "mean_token_accuracy": 0.20899544805288314,
      "num_tokens": 69561735.0,
      "step": 34330
    },
    {
      "entropy": 5.986356782913208,
      "epoch": 3.000524338023246,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004107162189267028,
      "loss": 5.2256,
      "mean_token_accuracy": 0.2058022752404213,
      "num_tokens": 69573165.0,
      "step": 34335
    },
    {
      "entropy": 5.988177156448364,
      "epoch": 3.0009612863759503,
      "grad_norm": 1.328125,
      "learning_rate": 0.00041069136344999273,
      "loss": 5.1832,
      "mean_token_accuracy": 0.21526970565319062,
      "num_tokens": 69583484.0,
      "step": 34340
    },
    {
      "entropy": 5.971571969985962,
      "epoch": 3.001398234728655,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004106665053706512,
      "loss": 5.2238,
      "mean_token_accuracy": 0.20604751110076905,
      "num_tokens": 69593486.0,
      "step": 34345
    },
    {
      "entropy": 6.061177206039429,
      "epoch": 3.0018351830813597,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000410641644689155,
      "loss": 5.3085,
      "mean_token_accuracy": 0.20720209926366806,
      "num_tokens": 69604245.0,
      "step": 34350
    },
    {
      "entropy": 6.039223146438599,
      "epoch": 3.0022721314340646,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000410616781405981,
      "loss": 5.2172,
      "mean_token_accuracy": 0.2065788760781288,
      "num_tokens": 69614237.0,
      "step": 34355
    },
    {
      "entropy": 6.073290109634399,
      "epoch": 3.002709079786769,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004105919155216061,
      "loss": 5.249,
      "mean_token_accuracy": 0.21293893158435823,
      "num_tokens": 69623695.0,
      "step": 34360
    },
    {
      "entropy": 6.038322257995605,
      "epoch": 3.003146028139474,
      "grad_norm": 1.34375,
      "learning_rate": 0.00041056704703650715,
      "loss": 5.1209,
      "mean_token_accuracy": 0.21962381452322005,
      "num_tokens": 69633299.0,
      "step": 34365
    },
    {
      "entropy": 5.9754057884216305,
      "epoch": 3.0035829764921784,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004105421759511612,
      "loss": 5.2127,
      "mean_token_accuracy": 0.20403510183095933,
      "num_tokens": 69642853.0,
      "step": 34370
    },
    {
      "entropy": 6.041961002349853,
      "epoch": 3.0040199248448833,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004105173022660453,
      "loss": 5.2547,
      "mean_token_accuracy": 0.2018100365996361,
      "num_tokens": 69652697.0,
      "step": 34375
    },
    {
      "entropy": 6.00643744468689,
      "epoch": 3.0044568731975883,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004104924259816366,
      "loss": 5.2552,
      "mean_token_accuracy": 0.2130083531141281,
      "num_tokens": 69664261.0,
      "step": 34380
    },
    {
      "entropy": 6.0444403171539305,
      "epoch": 3.0048938215502927,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004104675470984121,
      "loss": 5.2612,
      "mean_token_accuracy": 0.2144293174147606,
      "num_tokens": 69674819.0,
      "step": 34385
    },
    {
      "entropy": 6.033697938919067,
      "epoch": 3.0053307699029976,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041044266561684914,
      "loss": 5.2114,
      "mean_token_accuracy": 0.20518747419118882,
      "num_tokens": 69685546.0,
      "step": 34390
    },
    {
      "entropy": 5.907575178146362,
      "epoch": 3.005767718255702,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004104177815374248,
      "loss": 5.1252,
      "mean_token_accuracy": 0.22015803158283234,
      "num_tokens": 69695192.0,
      "step": 34395
    },
    {
      "entropy": 6.037266397476197,
      "epoch": 3.006204666608407,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004103928948606164,
      "loss": 5.1808,
      "mean_token_accuracy": 0.20717159658670425,
      "num_tokens": 69705669.0,
      "step": 34400
    },
    {
      "entropy": 5.977775239944458,
      "epoch": 3.0066416149611115,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004103680055869015,
      "loss": 5.168,
      "mean_token_accuracy": 0.22085942924022675,
      "num_tokens": 69716274.0,
      "step": 34405
    },
    {
      "entropy": 6.053186893463135,
      "epoch": 3.0070785633138164,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004103431137167572,
      "loss": 5.2452,
      "mean_token_accuracy": 0.2090905174612999,
      "num_tokens": 69726438.0,
      "step": 34410
    },
    {
      "entropy": 6.011644077301026,
      "epoch": 3.007515511666521,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00041031821925066115,
      "loss": 5.2187,
      "mean_token_accuracy": 0.21340183168649673,
      "num_tokens": 69735467.0,
      "step": 34415
    },
    {
      "entropy": 6.009003686904907,
      "epoch": 3.007952460019226,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041029332218909067,
      "loss": 5.2163,
      "mean_token_accuracy": 0.20692601352930068,
      "num_tokens": 69744886.0,
      "step": 34420
    },
    {
      "entropy": 5.997377729415893,
      "epoch": 3.0083894083719303,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041026842253252333,
      "loss": 5.2383,
      "mean_token_accuracy": 0.2047009140253067,
      "num_tokens": 69754337.0,
      "step": 34425
    },
    {
      "entropy": 5.975416898727417,
      "epoch": 3.008826356724635,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004102435202814369,
      "loss": 5.2085,
      "mean_token_accuracy": 0.21449752002954484,
      "num_tokens": 69764660.0,
      "step": 34430
    },
    {
      "entropy": 5.957406663894654,
      "epoch": 3.00926330507734,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004102186154363088,
      "loss": 5.196,
      "mean_token_accuracy": 0.210871322453022,
      "num_tokens": 69774339.0,
      "step": 34435
    },
    {
      "entropy": 6.026563215255737,
      "epoch": 3.0097002534300445,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041019370799761685,
      "loss": 5.2729,
      "mean_token_accuracy": 0.2051939845085144,
      "num_tokens": 69785272.0,
      "step": 34440
    },
    {
      "entropy": 6.017344236373901,
      "epoch": 3.0101372017827495,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004101687979658386,
      "loss": 5.3126,
      "mean_token_accuracy": 0.2052341014146805,
      "num_tokens": 69796646.0,
      "step": 34445
    },
    {
      "entropy": 5.964749193191528,
      "epoch": 3.010574150135454,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041014388534145213,
      "loss": 5.1624,
      "mean_token_accuracy": 0.21618392020463945,
      "num_tokens": 69807233.0,
      "step": 34450
    },
    {
      "entropy": 6.025935411453247,
      "epoch": 3.011011098488159,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004101189701249351,
      "loss": 5.2613,
      "mean_token_accuracy": 0.20286018699407576,
      "num_tokens": 69818131.0,
      "step": 34455
    },
    {
      "entropy": 6.076148462295532,
      "epoch": 3.0114480468408633,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004100940523167654,
      "loss": 5.3108,
      "mean_token_accuracy": 0.20156297236680984,
      "num_tokens": 69829537.0,
      "step": 34460
    },
    {
      "entropy": 6.027623844146729,
      "epoch": 3.0118849951935682,
      "grad_norm": 1.203125,
      "learning_rate": 0.000410069131917421,
      "loss": 5.2573,
      "mean_token_accuracy": 0.20280795842409133,
      "num_tokens": 69840964.0,
      "step": 34465
    },
    {
      "entropy": 6.00153546333313,
      "epoch": 3.0123219435462727,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041004420892737993,
      "loss": 5.171,
      "mean_token_accuracy": 0.20813340544700623,
      "num_tokens": 69849919.0,
      "step": 34470
    },
    {
      "entropy": 5.963567876815796,
      "epoch": 3.0127588918989776,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004100192833471201,
      "loss": 5.2539,
      "mean_token_accuracy": 0.21299227327108383,
      "num_tokens": 69859832.0,
      "step": 34475
    },
    {
      "entropy": 6.011879348754883,
      "epoch": 3.013195840251682,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004099943551771196,
      "loss": 5.1488,
      "mean_token_accuracy": 0.21071719974279404,
      "num_tokens": 69869382.0,
      "step": 34480
    },
    {
      "entropy": 5.9668961524963375,
      "epoch": 3.013632788604387,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004099694244178568,
      "loss": 5.1832,
      "mean_token_accuracy": 0.2124067425727844,
      "num_tokens": 69879282.0,
      "step": 34485
    },
    {
      "entropy": 5.940767812728882,
      "epoch": 3.0140697369570915,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004099444910698096,
      "loss": 5.1407,
      "mean_token_accuracy": 0.2236734986305237,
      "num_tokens": 69889876.0,
      "step": 34490
    },
    {
      "entropy": 6.057441377639771,
      "epoch": 3.0145066853097964,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004099195551334564,
      "loss": 5.2567,
      "mean_token_accuracy": 0.19934893995523453,
      "num_tokens": 69899642.0,
      "step": 34495
    },
    {
      "entropy": 5.973689365386963,
      "epoch": 3.0149436336625013,
      "grad_norm": 1.375,
      "learning_rate": 0.00040989461660927536,
      "loss": 5.1351,
      "mean_token_accuracy": 0.21481484025716782,
      "num_tokens": 69909877.0,
      "step": 34500
    },
    {
      "entropy": 5.966961193084717,
      "epoch": 3.0153805820152058,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040986967549774497,
      "loss": 5.1592,
      "mean_token_accuracy": 0.21692205369472503,
      "num_tokens": 69920061.0,
      "step": 34505
    },
    {
      "entropy": 5.966281986236572,
      "epoch": 3.0158175303679107,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004098447317993435,
      "loss": 5.2037,
      "mean_token_accuracy": 0.21201127171516418,
      "num_tokens": 69930739.0,
      "step": 34510
    },
    {
      "entropy": 5.960740470886231,
      "epoch": 3.016254478720615,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040981978551454947,
      "loss": 5.1223,
      "mean_token_accuracy": 0.21325641572475434,
      "num_tokens": 69939825.0,
      "step": 34515
    },
    {
      "entropy": 6.101427936553955,
      "epoch": 3.01669142707332,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040979483664384123,
      "loss": 5.3615,
      "mean_token_accuracy": 0.19593294709920883,
      "num_tokens": 69949662.0,
      "step": 34520
    },
    {
      "entropy": 5.97711820602417,
      "epoch": 3.0171283754260245,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040976988518769743,
      "loss": 5.231,
      "mean_token_accuracy": 0.20933706909418107,
      "num_tokens": 69959950.0,
      "step": 34525
    },
    {
      "entropy": 6.043479156494141,
      "epoch": 3.0175653237787294,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040974493114659654,
      "loss": 5.281,
      "mean_token_accuracy": 0.19929179102182387,
      "num_tokens": 69970451.0,
      "step": 34530
    },
    {
      "entropy": 6.0046624660491945,
      "epoch": 3.018002272131434,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004097199745210174,
      "loss": 5.199,
      "mean_token_accuracy": 0.2125801146030426,
      "num_tokens": 69979940.0,
      "step": 34535
    },
    {
      "entropy": 6.014459943771362,
      "epoch": 3.018439220484139,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040969501531143844,
      "loss": 5.2001,
      "mean_token_accuracy": 0.2175571456551552,
      "num_tokens": 69989507.0,
      "step": 34540
    },
    {
      "entropy": 6.046332120895386,
      "epoch": 3.0188761688368433,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040967005351833863,
      "loss": 5.1763,
      "mean_token_accuracy": 0.21482913047075272,
      "num_tokens": 69999184.0,
      "step": 34545
    },
    {
      "entropy": 6.046880197525025,
      "epoch": 3.019313117189548,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004096450891421965,
      "loss": 5.2627,
      "mean_token_accuracy": 0.20536292642354964,
      "num_tokens": 70009569.0,
      "step": 34550
    },
    {
      "entropy": 6.04979100227356,
      "epoch": 3.019750065542253,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040962012218349105,
      "loss": 5.2378,
      "mean_token_accuracy": 0.20715751647949218,
      "num_tokens": 70019824.0,
      "step": 34555
    },
    {
      "entropy": 5.95647611618042,
      "epoch": 3.0201870138949576,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004095951526427011,
      "loss": 5.1082,
      "mean_token_accuracy": 0.21716589033603667,
      "num_tokens": 70029043.0,
      "step": 34560
    },
    {
      "entropy": 6.012052822113037,
      "epoch": 3.0206239622476625,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004095701805203056,
      "loss": 5.2464,
      "mean_token_accuracy": 0.2084464520215988,
      "num_tokens": 70039794.0,
      "step": 34565
    },
    {
      "entropy": 6.061886548995972,
      "epoch": 3.021060910600367,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004095452058167837,
      "loss": 5.2462,
      "mean_token_accuracy": 0.20718639940023423,
      "num_tokens": 70049866.0,
      "step": 34570
    },
    {
      "entropy": 5.991275930404663,
      "epoch": 3.021497858953072,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000409520228532614,
      "loss": 5.1556,
      "mean_token_accuracy": 0.21354180276393891,
      "num_tokens": 70059471.0,
      "step": 34575
    },
    {
      "entropy": 5.980598878860474,
      "epoch": 3.0219348073057763,
      "grad_norm": 1.359375,
      "learning_rate": 0.000409495248668276,
      "loss": 5.1763,
      "mean_token_accuracy": 0.2158389613032341,
      "num_tokens": 70069003.0,
      "step": 34580
    },
    {
      "entropy": 6.011398124694824,
      "epoch": 3.0223717556584813,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040947026622424853,
      "loss": 5.2733,
      "mean_token_accuracy": 0.2067425385117531,
      "num_tokens": 70078650.0,
      "step": 34585
    },
    {
      "entropy": 6.030223989486695,
      "epoch": 3.0228087040111857,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040944528120101097,
      "loss": 5.1757,
      "mean_token_accuracy": 0.213797989487648,
      "num_tokens": 70087361.0,
      "step": 34590
    },
    {
      "entropy": 6.013439130783081,
      "epoch": 3.0232456523638906,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040942029359904245,
      "loss": 5.2424,
      "mean_token_accuracy": 0.2128133073449135,
      "num_tokens": 70097289.0,
      "step": 34595
    },
    {
      "entropy": 5.992819547653198,
      "epoch": 3.023682600716595,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040939530341882225,
      "loss": 5.1937,
      "mean_token_accuracy": 0.2195806697010994,
      "num_tokens": 70108099.0,
      "step": 34600
    },
    {
      "entropy": 6.035924005508423,
      "epoch": 3.0241195490693,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040937031066082974,
      "loss": 5.2891,
      "mean_token_accuracy": 0.20180162489414216,
      "num_tokens": 70119243.0,
      "step": 34605
    },
    {
      "entropy": 6.0490686893463135,
      "epoch": 3.0245564974220045,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004093453153255442,
      "loss": 5.1909,
      "mean_token_accuracy": 0.2141481250524521,
      "num_tokens": 70129073.0,
      "step": 34610
    },
    {
      "entropy": 5.944206285476684,
      "epoch": 3.0249934457747094,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004093203174134452,
      "loss": 5.1466,
      "mean_token_accuracy": 0.217157681286335,
      "num_tokens": 70139778.0,
      "step": 34615
    },
    {
      "entropy": 6.046713924407959,
      "epoch": 3.0254303941274143,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040929531692501206,
      "loss": 5.3272,
      "mean_token_accuracy": 0.19973155856132507,
      "num_tokens": 70151088.0,
      "step": 34620
    },
    {
      "entropy": 5.980036497116089,
      "epoch": 3.025867342480119,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040927031386072435,
      "loss": 5.1729,
      "mean_token_accuracy": 0.21793546080589293,
      "num_tokens": 70160133.0,
      "step": 34625
    },
    {
      "entropy": 6.034684181213379,
      "epoch": 3.0263042908328237,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040924530822106177,
      "loss": 5.1927,
      "mean_token_accuracy": 0.2073066160082817,
      "num_tokens": 70171176.0,
      "step": 34630
    },
    {
      "entropy": 6.032678842544556,
      "epoch": 3.026741239185528,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040922030000650376,
      "loss": 5.2245,
      "mean_token_accuracy": 0.200888229906559,
      "num_tokens": 70181413.0,
      "step": 34635
    },
    {
      "entropy": 6.0451654434204105,
      "epoch": 3.027178187538233,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004091952892175301,
      "loss": 5.2405,
      "mean_token_accuracy": 0.2054998829960823,
      "num_tokens": 70190535.0,
      "step": 34640
    },
    {
      "entropy": 5.981273126602173,
      "epoch": 3.0276151358909376,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00040917027585462044,
      "loss": 5.2101,
      "mean_token_accuracy": 0.2085585966706276,
      "num_tokens": 70199440.0,
      "step": 34645
    },
    {
      "entropy": 6.01178674697876,
      "epoch": 3.0280520842436425,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004091452599182546,
      "loss": 5.1823,
      "mean_token_accuracy": 0.21370837539434434,
      "num_tokens": 70209846.0,
      "step": 34650
    },
    {
      "entropy": 6.031817197799683,
      "epoch": 3.028489032596347,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040912024140891235,
      "loss": 5.2675,
      "mean_token_accuracy": 0.20391411185264588,
      "num_tokens": 70219910.0,
      "step": 34655
    },
    {
      "entropy": 6.027023363113403,
      "epoch": 3.028925980949052,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004090952203270737,
      "loss": 5.2165,
      "mean_token_accuracy": 0.20560523271560668,
      "num_tokens": 70230937.0,
      "step": 34660
    },
    {
      "entropy": 6.00078821182251,
      "epoch": 3.0293629293017563,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040907019667321833,
      "loss": 5.2125,
      "mean_token_accuracy": 0.21440518647432327,
      "num_tokens": 70240791.0,
      "step": 34665
    },
    {
      "entropy": 6.031328105926514,
      "epoch": 3.0297998776544612,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004090451704478264,
      "loss": 5.216,
      "mean_token_accuracy": 0.21656785011291504,
      "num_tokens": 70251231.0,
      "step": 34670
    },
    {
      "entropy": 6.024787759780883,
      "epoch": 3.030236826007166,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040902014165137787,
      "loss": 5.1393,
      "mean_token_accuracy": 0.21764903962612153,
      "num_tokens": 70260993.0,
      "step": 34675
    },
    {
      "entropy": 6.081805038452148,
      "epoch": 3.0306737743598706,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004089951102843528,
      "loss": 5.3107,
      "mean_token_accuracy": 0.2045293778181076,
      "num_tokens": 70271380.0,
      "step": 34680
    },
    {
      "entropy": 6.055787086486816,
      "epoch": 3.0311107227125755,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040897007634723125,
      "loss": 5.2586,
      "mean_token_accuracy": 0.20813646763563157,
      "num_tokens": 70282614.0,
      "step": 34685
    },
    {
      "entropy": 6.039513444900512,
      "epoch": 3.03154767106528,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004089450398404935,
      "loss": 5.2107,
      "mean_token_accuracy": 0.21309684514999389,
      "num_tokens": 70292084.0,
      "step": 34690
    },
    {
      "entropy": 5.968467903137207,
      "epoch": 3.031984619417985,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040892000076461966,
      "loss": 5.1275,
      "mean_token_accuracy": 0.21681503504514693,
      "num_tokens": 70302205.0,
      "step": 34695
    },
    {
      "entropy": 6.010453462600708,
      "epoch": 3.0324215677706894,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040889495912009,
      "loss": 5.2823,
      "mean_token_accuracy": 0.20425643026828766,
      "num_tokens": 70312264.0,
      "step": 34700
    },
    {
      "entropy": 6.00281252861023,
      "epoch": 3.0328585161233943,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040886991490738495,
      "loss": 5.1604,
      "mean_token_accuracy": 0.21152516901493074,
      "num_tokens": 70321733.0,
      "step": 34705
    },
    {
      "entropy": 6.006563758850097,
      "epoch": 3.0332954644760988,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040884486812698475,
      "loss": 5.2975,
      "mean_token_accuracy": 0.20453153550624847,
      "num_tokens": 70331621.0,
      "step": 34710
    },
    {
      "entropy": 5.999759149551392,
      "epoch": 3.0337324128288037,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040881981877936987,
      "loss": 5.21,
      "mean_token_accuracy": 0.2052023246884346,
      "num_tokens": 70342247.0,
      "step": 34715
    },
    {
      "entropy": 6.0163867473602295,
      "epoch": 3.034169361181508,
      "grad_norm": 1.25,
      "learning_rate": 0.0004087947668650207,
      "loss": 5.1852,
      "mean_token_accuracy": 0.21541795283555984,
      "num_tokens": 70352423.0,
      "step": 34720
    },
    {
      "entropy": 6.014359188079834,
      "epoch": 3.034606309534213,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004087697123844178,
      "loss": 5.2661,
      "mean_token_accuracy": 0.2060771808028221,
      "num_tokens": 70362740.0,
      "step": 34725
    },
    {
      "entropy": 5.991729784011841,
      "epoch": 3.035043257886918,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004087446553380417,
      "loss": 5.1545,
      "mean_token_accuracy": 0.21577748358249665,
      "num_tokens": 70372456.0,
      "step": 34730
    },
    {
      "entropy": 6.041048574447632,
      "epoch": 3.0354802062396224,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004087195957263731,
      "loss": 5.1795,
      "mean_token_accuracy": 0.2191232070326805,
      "num_tokens": 70382741.0,
      "step": 34735
    },
    {
      "entropy": 5.952575206756592,
      "epoch": 3.0359171545923274,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040869453354989255,
      "loss": 5.2016,
      "mean_token_accuracy": 0.20889885127544403,
      "num_tokens": 70392712.0,
      "step": 34740
    },
    {
      "entropy": 6.0112708568572994,
      "epoch": 3.036354102945032,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040866946880908076,
      "loss": 5.2694,
      "mean_token_accuracy": 0.20501707792282103,
      "num_tokens": 70403655.0,
      "step": 34745
    },
    {
      "entropy": 6.111040878295898,
      "epoch": 3.0367910512977367,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004086444015044186,
      "loss": 5.3084,
      "mean_token_accuracy": 0.19941156655550002,
      "num_tokens": 70413720.0,
      "step": 34750
    },
    {
      "entropy": 6.007580757141113,
      "epoch": 3.037227999650441,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004086193316363867,
      "loss": 5.2202,
      "mean_token_accuracy": 0.2059991776943207,
      "num_tokens": 70423910.0,
      "step": 34755
    },
    {
      "entropy": 6.005898141860962,
      "epoch": 3.037664948003146,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040859425920546604,
      "loss": 5.2411,
      "mean_token_accuracy": 0.20772541612386702,
      "num_tokens": 70434593.0,
      "step": 34760
    },
    {
      "entropy": 5.999711561203003,
      "epoch": 3.0381018963558506,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004085691842121375,
      "loss": 5.207,
      "mean_token_accuracy": 0.20979429334402083,
      "num_tokens": 70443967.0,
      "step": 34765
    },
    {
      "entropy": 6.065043878555298,
      "epoch": 3.0385388447085555,
      "grad_norm": 1.28125,
      "learning_rate": 0.000408544106656882,
      "loss": 5.2364,
      "mean_token_accuracy": 0.20480695962905884,
      "num_tokens": 70454098.0,
      "step": 34770
    },
    {
      "entropy": 6.021069002151489,
      "epoch": 3.03897579306126,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040851902654018057,
      "loss": 5.2154,
      "mean_token_accuracy": 0.20753195285797119,
      "num_tokens": 70463946.0,
      "step": 34775
    },
    {
      "entropy": 6.042910099029541,
      "epoch": 3.039412741413965,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040849394386251425,
      "loss": 5.2424,
      "mean_token_accuracy": 0.20946975350379943,
      "num_tokens": 70474805.0,
      "step": 34780
    },
    {
      "entropy": 5.972177267074585,
      "epoch": 3.03984968976667,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004084688586243641,
      "loss": 5.1549,
      "mean_token_accuracy": 0.21932613402605056,
      "num_tokens": 70484585.0,
      "step": 34785
    },
    {
      "entropy": 6.044469261169434,
      "epoch": 3.0402866381193743,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004084437708262113,
      "loss": 5.2392,
      "mean_token_accuracy": 0.20639195442199706,
      "num_tokens": 70494811.0,
      "step": 34790
    },
    {
      "entropy": 6.05676498413086,
      "epoch": 3.040723586472079,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004084186804685371,
      "loss": 5.2455,
      "mean_token_accuracy": 0.20747948437929153,
      "num_tokens": 70505392.0,
      "step": 34795
    },
    {
      "entropy": 6.007638597488404,
      "epoch": 3.0411605348247837,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004083935875518227,
      "loss": 5.1998,
      "mean_token_accuracy": 0.21727822870016097,
      "num_tokens": 70515066.0,
      "step": 34800
    },
    {
      "entropy": 6.040521478652954,
      "epoch": 3.0415974831774886,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040836849207654937,
      "loss": 5.1946,
      "mean_token_accuracy": 0.20868303030729293,
      "num_tokens": 70525100.0,
      "step": 34805
    },
    {
      "entropy": 6.0022131443023685,
      "epoch": 3.042034431530193,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004083433940431985,
      "loss": 5.2049,
      "mean_token_accuracy": 0.20967471450567246,
      "num_tokens": 70535286.0,
      "step": 34810
    },
    {
      "entropy": 6.031146812438965,
      "epoch": 3.042471379882898,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004083182934522514,
      "loss": 5.2463,
      "mean_token_accuracy": 0.21288681924343109,
      "num_tokens": 70545724.0,
      "step": 34815
    },
    {
      "entropy": 6.012356376647949,
      "epoch": 3.0429083282356024,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040829319030418967,
      "loss": 5.1967,
      "mean_token_accuracy": 0.20941566079854965,
      "num_tokens": 70555179.0,
      "step": 34820
    },
    {
      "entropy": 6.068049573898316,
      "epoch": 3.0433452765883073,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004082680845994946,
      "loss": 5.2654,
      "mean_token_accuracy": 0.20767101794481277,
      "num_tokens": 70564355.0,
      "step": 34825
    },
    {
      "entropy": 5.972356557846069,
      "epoch": 3.043782224941012,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004082429763386479,
      "loss": 5.1671,
      "mean_token_accuracy": 0.2178034409880638,
      "num_tokens": 70575367.0,
      "step": 34830
    },
    {
      "entropy": 5.965447664260864,
      "epoch": 3.0442191732937167,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040821786552213107,
      "loss": 5.2013,
      "mean_token_accuracy": 0.21113389134407043,
      "num_tokens": 70585047.0,
      "step": 34835
    },
    {
      "entropy": 6.019962739944458,
      "epoch": 3.044656121646421,
      "grad_norm": 1.359375,
      "learning_rate": 0.00040819275215042586,
      "loss": 5.2286,
      "mean_token_accuracy": 0.20631040036678314,
      "num_tokens": 70594736.0,
      "step": 34840
    },
    {
      "entropy": 6.039164257049561,
      "epoch": 3.045093069999126,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040816763622401376,
      "loss": 5.2185,
      "mean_token_accuracy": 0.2104601800441742,
      "num_tokens": 70604762.0,
      "step": 34845
    },
    {
      "entropy": 6.011320114135742,
      "epoch": 3.045530018351831,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040814251774337666,
      "loss": 5.2347,
      "mean_token_accuracy": 0.20616308152675628,
      "num_tokens": 70614438.0,
      "step": 34850
    },
    {
      "entropy": 6.090040969848633,
      "epoch": 3.0459669667045355,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004081173967089963,
      "loss": 5.3459,
      "mean_token_accuracy": 0.20093904286623002,
      "num_tokens": 70624669.0,
      "step": 34855
    },
    {
      "entropy": 6.012873649597168,
      "epoch": 3.0464039150572404,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040809227312135456,
      "loss": 5.1696,
      "mean_token_accuracy": 0.2129221573472023,
      "num_tokens": 70635082.0,
      "step": 34860
    },
    {
      "entropy": 6.030430459976197,
      "epoch": 3.046840863409945,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004080671469809332,
      "loss": 5.2231,
      "mean_token_accuracy": 0.21189550310373306,
      "num_tokens": 70644117.0,
      "step": 34865
    },
    {
      "entropy": 6.031175708770752,
      "epoch": 3.04727781176265,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004080420182882144,
      "loss": 5.2592,
      "mean_token_accuracy": 0.20560390800237655,
      "num_tokens": 70653299.0,
      "step": 34870
    },
    {
      "entropy": 5.994077730178833,
      "epoch": 3.0477147601153542,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040801688704367975,
      "loss": 5.2219,
      "mean_token_accuracy": 0.2080558642745018,
      "num_tokens": 70664477.0,
      "step": 34875
    },
    {
      "entropy": 6.014699077606201,
      "epoch": 3.048151708468059,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00040799175324781164,
      "loss": 5.239,
      "mean_token_accuracy": 0.20281548500061036,
      "num_tokens": 70674635.0,
      "step": 34880
    },
    {
      "entropy": 5.967408895492554,
      "epoch": 3.0485886568207636,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000407966616901092,
      "loss": 5.1864,
      "mean_token_accuracy": 0.21450987309217454,
      "num_tokens": 70684633.0,
      "step": 34885
    },
    {
      "entropy": 6.027512454986573,
      "epoch": 3.0490256051734685,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004079414780040029,
      "loss": 5.2646,
      "mean_token_accuracy": 0.2115102380514145,
      "num_tokens": 70694416.0,
      "step": 34890
    },
    {
      "entropy": 6.05652928352356,
      "epoch": 3.049462553526173,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004079163365570267,
      "loss": 5.2096,
      "mean_token_accuracy": 0.20696344077587128,
      "num_tokens": 70703373.0,
      "step": 34895
    },
    {
      "entropy": 5.997592782974243,
      "epoch": 3.049899501878878,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004078911925606455,
      "loss": 5.1859,
      "mean_token_accuracy": 0.20805690288543702,
      "num_tokens": 70713868.0,
      "step": 34900
    },
    {
      "entropy": 6.058240699768066,
      "epoch": 3.050336450231583,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004078660460153415,
      "loss": 5.3438,
      "mean_token_accuracy": 0.196154448390007,
      "num_tokens": 70724306.0,
      "step": 34905
    },
    {
      "entropy": 6.049051523208618,
      "epoch": 3.0507733985842873,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00040784089692159724,
      "loss": 5.2681,
      "mean_token_accuracy": 0.2077073022723198,
      "num_tokens": 70733923.0,
      "step": 34910
    },
    {
      "entropy": 6.064835166931152,
      "epoch": 3.0512103469369922,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004078157452798949,
      "loss": 5.3048,
      "mean_token_accuracy": 0.20158809870481492,
      "num_tokens": 70744964.0,
      "step": 34915
    },
    {
      "entropy": 6.008375072479248,
      "epoch": 3.0516472952896967,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040779059109071706,
      "loss": 5.1635,
      "mean_token_accuracy": 0.2124011442065239,
      "num_tokens": 70754639.0,
      "step": 34920
    },
    {
      "entropy": 6.080964183807373,
      "epoch": 3.0520842436424016,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040776543435454597,
      "loss": 5.2619,
      "mean_token_accuracy": 0.2117330923676491,
      "num_tokens": 70764810.0,
      "step": 34925
    },
    {
      "entropy": 6.061050033569336,
      "epoch": 3.052521191995106,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004077402750718643,
      "loss": 5.2465,
      "mean_token_accuracy": 0.20695436149835586,
      "num_tokens": 70774698.0,
      "step": 34930
    },
    {
      "entropy": 6.029809761047363,
      "epoch": 3.052958140347811,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004077151132431546,
      "loss": 5.1934,
      "mean_token_accuracy": 0.2169630855321884,
      "num_tokens": 70785522.0,
      "step": 34935
    },
    {
      "entropy": 6.007316827774048,
      "epoch": 3.0533950887005155,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004076899488688996,
      "loss": 5.2676,
      "mean_token_accuracy": 0.20527447611093522,
      "num_tokens": 70797001.0,
      "step": 34940
    },
    {
      "entropy": 5.980831003189087,
      "epoch": 3.0538320370532204,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040766478194958176,
      "loss": 5.2748,
      "mean_token_accuracy": 0.20794493407011033,
      "num_tokens": 70807247.0,
      "step": 34945
    },
    {
      "entropy": 6.056809806823731,
      "epoch": 3.054268985405925,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004076396124856839,
      "loss": 5.3112,
      "mean_token_accuracy": 0.19599460661411286,
      "num_tokens": 70818269.0,
      "step": 34950
    },
    {
      "entropy": 6.105883169174194,
      "epoch": 3.0547059337586298,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004076144404776888,
      "loss": 5.3182,
      "mean_token_accuracy": 0.2014859139919281,
      "num_tokens": 70827759.0,
      "step": 34955
    },
    {
      "entropy": 6.018387365341186,
      "epoch": 3.0551428821113342,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004075892659260792,
      "loss": 5.2347,
      "mean_token_accuracy": 0.2056772977113724,
      "num_tokens": 70838296.0,
      "step": 34960
    },
    {
      "entropy": 6.003496694564819,
      "epoch": 3.055579830464039,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040756408883133797,
      "loss": 5.2253,
      "mean_token_accuracy": 0.20396629869937896,
      "num_tokens": 70848364.0,
      "step": 34965
    },
    {
      "entropy": 6.04559326171875,
      "epoch": 3.056016778816744,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004075389091939481,
      "loss": 5.2178,
      "mean_token_accuracy": 0.2170301139354706,
      "num_tokens": 70858845.0,
      "step": 34970
    },
    {
      "entropy": 6.0101603984832765,
      "epoch": 3.0564537271694485,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004075137270143925,
      "loss": 5.2751,
      "mean_token_accuracy": 0.2146001562476158,
      "num_tokens": 70868976.0,
      "step": 34975
    },
    {
      "entropy": 5.990287828445434,
      "epoch": 3.0568906755221534,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004074885422931542,
      "loss": 5.1904,
      "mean_token_accuracy": 0.22054076045751572,
      "num_tokens": 70878287.0,
      "step": 34980
    },
    {
      "entropy": 6.0224062442779545,
      "epoch": 3.057327623874858,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040746335503071625,
      "loss": 5.2438,
      "mean_token_accuracy": 0.21478271484375,
      "num_tokens": 70889374.0,
      "step": 34985
    },
    {
      "entropy": 5.97563738822937,
      "epoch": 3.057764572227563,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040743816522756165,
      "loss": 5.2198,
      "mean_token_accuracy": 0.21523999720811843,
      "num_tokens": 70899741.0,
      "step": 34990
    },
    {
      "entropy": 6.011527824401855,
      "epoch": 3.0582015205802673,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040741297288417373,
      "loss": 5.285,
      "mean_token_accuracy": 0.20689035803079606,
      "num_tokens": 70909962.0,
      "step": 34995
    },
    {
      "entropy": 6.039534091949463,
      "epoch": 3.058638468932972,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004073877780010356,
      "loss": 5.2185,
      "mean_token_accuracy": 0.20905487537384032,
      "num_tokens": 70921129.0,
      "step": 35000
    },
    {
      "entropy": 6.0252564430236815,
      "epoch": 3.0590754172856767,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004073625805786305,
      "loss": 5.2194,
      "mean_token_accuracy": 0.2144826680421829,
      "num_tokens": 70930492.0,
      "step": 35005
    },
    {
      "entropy": 6.073181200027466,
      "epoch": 3.0595123656383816,
      "grad_norm": 1.25,
      "learning_rate": 0.0004073373806174417,
      "loss": 5.2338,
      "mean_token_accuracy": 0.20402433723211288,
      "num_tokens": 70940667.0,
      "step": 35010
    },
    {
      "entropy": 6.018658304214478,
      "epoch": 3.059949313991086,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004073121781179527,
      "loss": 5.1913,
      "mean_token_accuracy": 0.21208285838365554,
      "num_tokens": 70949815.0,
      "step": 35015
    },
    {
      "entropy": 6.074837732315063,
      "epoch": 3.060386262343791,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040728697308064674,
      "loss": 5.3317,
      "mean_token_accuracy": 0.19238968044519425,
      "num_tokens": 70959931.0,
      "step": 35020
    },
    {
      "entropy": 6.025677013397217,
      "epoch": 3.060823210696496,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004072617655060073,
      "loss": 5.1971,
      "mean_token_accuracy": 0.20690015703439713,
      "num_tokens": 70970204.0,
      "step": 35025
    },
    {
      "entropy": 6.054167795181274,
      "epoch": 3.0612601590492003,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040723655539451787,
      "loss": 5.2954,
      "mean_token_accuracy": 0.2055946946144104,
      "num_tokens": 70981841.0,
      "step": 35030
    },
    {
      "entropy": 5.9966148853302,
      "epoch": 3.0616971074019053,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004072113427466621,
      "loss": 5.2334,
      "mean_token_accuracy": 0.21001926064491272,
      "num_tokens": 70992281.0,
      "step": 35035
    },
    {
      "entropy": 5.952709293365478,
      "epoch": 3.0621340557546097,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040718612756292347,
      "loss": 5.1813,
      "mean_token_accuracy": 0.21410521417856215,
      "num_tokens": 71002737.0,
      "step": 35040
    },
    {
      "entropy": 6.012992477416992,
      "epoch": 3.0625710041073146,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004071609098437856,
      "loss": 5.2357,
      "mean_token_accuracy": 0.2084740549325943,
      "num_tokens": 71013604.0,
      "step": 35045
    },
    {
      "entropy": 5.9639317989349365,
      "epoch": 3.063007952460019,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040713568958973225,
      "loss": 5.2154,
      "mean_token_accuracy": 0.20805667638778685,
      "num_tokens": 71023278.0,
      "step": 35050
    },
    {
      "entropy": 6.019426965713501,
      "epoch": 3.063444900812724,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040711046680124715,
      "loss": 5.2548,
      "mean_token_accuracy": 0.20929258912801743,
      "num_tokens": 71033823.0,
      "step": 35055
    },
    {
      "entropy": 5.95811972618103,
      "epoch": 3.0638818491654285,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004070852414788141,
      "loss": 5.2705,
      "mean_token_accuracy": 0.2031094714999199,
      "num_tokens": 71043802.0,
      "step": 35060
    },
    {
      "entropy": 6.063987684249878,
      "epoch": 3.0643187975181334,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004070600136229168,
      "loss": 5.2873,
      "mean_token_accuracy": 0.20351822674274445,
      "num_tokens": 71054152.0,
      "step": 35065
    },
    {
      "entropy": 6.01295804977417,
      "epoch": 3.064755745870838,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040703478323403944,
      "loss": 5.1836,
      "mean_token_accuracy": 0.21090617626905442,
      "num_tokens": 71065346.0,
      "step": 35070
    },
    {
      "entropy": 6.000601196289063,
      "epoch": 3.065192694223543,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040700955031266556,
      "loss": 5.2328,
      "mean_token_accuracy": 0.20841710418462753,
      "num_tokens": 71074855.0,
      "step": 35075
    },
    {
      "entropy": 5.99900541305542,
      "epoch": 3.0656296425762477,
      "grad_norm": 1.25,
      "learning_rate": 0.0004069843148592794,
      "loss": 5.1821,
      "mean_token_accuracy": 0.21553916186094285,
      "num_tokens": 71084372.0,
      "step": 35080
    },
    {
      "entropy": 6.020778322219849,
      "epoch": 3.066066590928952,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000406959076874365,
      "loss": 5.2768,
      "mean_token_accuracy": 0.2092842623591423,
      "num_tokens": 71093365.0,
      "step": 35085
    },
    {
      "entropy": 5.941695070266723,
      "epoch": 3.066503539281657,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040693383635840625,
      "loss": 5.1786,
      "mean_token_accuracy": 0.21484682559967042,
      "num_tokens": 71102748.0,
      "step": 35090
    },
    {
      "entropy": 6.0295110702514645,
      "epoch": 3.0669404876343616,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004069085933118874,
      "loss": 5.2706,
      "mean_token_accuracy": 0.2097337007522583,
      "num_tokens": 71113878.0,
      "step": 35095
    },
    {
      "entropy": 6.120940732955932,
      "epoch": 3.0673774359870665,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004068833477352927,
      "loss": 5.31,
      "mean_token_accuracy": 0.20274885296821593,
      "num_tokens": 71124849.0,
      "step": 35100
    },
    {
      "entropy": 6.017226982116699,
      "epoch": 3.067814384339771,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040685809962910614,
      "loss": 5.2118,
      "mean_token_accuracy": 0.20979052186012268,
      "num_tokens": 71134064.0,
      "step": 35105
    },
    {
      "entropy": 5.948977470397949,
      "epoch": 3.068251332692476,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040683284899381225,
      "loss": 5.1227,
      "mean_token_accuracy": 0.2102445349097252,
      "num_tokens": 71144270.0,
      "step": 35110
    },
    {
      "entropy": 6.001692008972168,
      "epoch": 3.0686882810451803,
      "grad_norm": 1.359375,
      "learning_rate": 0.00040680759582989516,
      "loss": 5.1478,
      "mean_token_accuracy": 0.21597563624382018,
      "num_tokens": 71153949.0,
      "step": 35115
    },
    {
      "entropy": 6.03813853263855,
      "epoch": 3.0691252293978852,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040678234013783945,
      "loss": 5.2278,
      "mean_token_accuracy": 0.20983532816171646,
      "num_tokens": 71164858.0,
      "step": 35120
    },
    {
      "entropy": 6.0059154510498045,
      "epoch": 3.0695621777505897,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004067570819181293,
      "loss": 5.1365,
      "mean_token_accuracy": 0.2176334410905838,
      "num_tokens": 71174832.0,
      "step": 35125
    },
    {
      "entropy": 5.997531700134277,
      "epoch": 3.0699991261032946,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004067318211712492,
      "loss": 5.247,
      "mean_token_accuracy": 0.20526065230369567,
      "num_tokens": 71185089.0,
      "step": 35130
    },
    {
      "entropy": 6.032149314880371,
      "epoch": 3.070436074455999,
      "grad_norm": 1.25,
      "learning_rate": 0.0004067065578976839,
      "loss": 5.2049,
      "mean_token_accuracy": 0.20928356498479844,
      "num_tokens": 71195889.0,
      "step": 35135
    },
    {
      "entropy": 6.1108794689178465,
      "epoch": 3.070873022808704,
      "grad_norm": 1.359375,
      "learning_rate": 0.00040668129209791773,
      "loss": 5.2746,
      "mean_token_accuracy": 0.20242840051651,
      "num_tokens": 71205825.0,
      "step": 35140
    },
    {
      "entropy": 6.010470724105835,
      "epoch": 3.071309971161409,
      "grad_norm": 1.25,
      "learning_rate": 0.0004066560237724354,
      "loss": 5.1617,
      "mean_token_accuracy": 0.21582695841789246,
      "num_tokens": 71215201.0,
      "step": 35145
    },
    {
      "entropy": 5.904430294036866,
      "epoch": 3.0717469195141134,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004066307529217215,
      "loss": 5.117,
      "mean_token_accuracy": 0.22237349301576614,
      "num_tokens": 71225329.0,
      "step": 35150
    },
    {
      "entropy": 5.974742460250854,
      "epoch": 3.0721838678668183,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040660547954626076,
      "loss": 5.138,
      "mean_token_accuracy": 0.21650433987379075,
      "num_tokens": 71235301.0,
      "step": 35155
    },
    {
      "entropy": 6.0100654602050785,
      "epoch": 3.0726208162195228,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040658020364653803,
      "loss": 5.2432,
      "mean_token_accuracy": 0.20512788444757463,
      "num_tokens": 71244991.0,
      "step": 35160
    },
    {
      "entropy": 5.997658538818359,
      "epoch": 3.0730577645722277,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000406554925223038,
      "loss": 5.2226,
      "mean_token_accuracy": 0.2121350273489952,
      "num_tokens": 71255480.0,
      "step": 35165
    },
    {
      "entropy": 6.058477067947388,
      "epoch": 3.073494712924932,
      "grad_norm": 1.25,
      "learning_rate": 0.0004065296442762456,
      "loss": 5.3243,
      "mean_token_accuracy": 0.20239043235778809,
      "num_tokens": 71265996.0,
      "step": 35170
    },
    {
      "entropy": 6.073267602920533,
      "epoch": 3.073931661277637,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040650436080664576,
      "loss": 5.224,
      "mean_token_accuracy": 0.20498018860816955,
      "num_tokens": 71277293.0,
      "step": 35175
    },
    {
      "entropy": 6.039472579956055,
      "epoch": 3.0743686096303415,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040647907481472333,
      "loss": 5.1843,
      "mean_token_accuracy": 0.2042487308382988,
      "num_tokens": 71285903.0,
      "step": 35180
    },
    {
      "entropy": 5.969787359237671,
      "epoch": 3.0748055579830464,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004064537863009633,
      "loss": 5.171,
      "mean_token_accuracy": 0.21520075649023057,
      "num_tokens": 71294979.0,
      "step": 35185
    },
    {
      "entropy": 5.97256293296814,
      "epoch": 3.075242506335751,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040642849526585087,
      "loss": 5.1921,
      "mean_token_accuracy": 0.2144932195544243,
      "num_tokens": 71306223.0,
      "step": 35190
    },
    {
      "entropy": 6.095684432983399,
      "epoch": 3.075679454688456,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004064032017098709,
      "loss": 5.3254,
      "mean_token_accuracy": 0.20019950419664384,
      "num_tokens": 71316984.0,
      "step": 35195
    },
    {
      "entropy": 6.092527627944946,
      "epoch": 3.0761164030411607,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040637790563350886,
      "loss": 5.298,
      "mean_token_accuracy": 0.20986131578683853,
      "num_tokens": 71327855.0,
      "step": 35200
    },
    {
      "entropy": 6.016704559326172,
      "epoch": 3.076553351393865,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004063526070372496,
      "loss": 5.202,
      "mean_token_accuracy": 0.2095435932278633,
      "num_tokens": 71338115.0,
      "step": 35205
    },
    {
      "entropy": 5.955842113494873,
      "epoch": 3.07699029974657,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004063273059215785,
      "loss": 5.1504,
      "mean_token_accuracy": 0.21693524569272996,
      "num_tokens": 71348533.0,
      "step": 35210
    },
    {
      "entropy": 5.907676553726196,
      "epoch": 3.0774272480992746,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004063020022869809,
      "loss": 5.0263,
      "mean_token_accuracy": 0.2314365267753601,
      "num_tokens": 71358368.0,
      "step": 35215
    },
    {
      "entropy": 6.020469284057617,
      "epoch": 3.0778641964519795,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040627669613394206,
      "loss": 5.2364,
      "mean_token_accuracy": 0.20737760812044143,
      "num_tokens": 71368443.0,
      "step": 35220
    },
    {
      "entropy": 6.0260374546051025,
      "epoch": 3.078301144804684,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040625138746294747,
      "loss": 5.2328,
      "mean_token_accuracy": 0.20047150999307634,
      "num_tokens": 71378508.0,
      "step": 35225
    },
    {
      "entropy": 6.024166297912598,
      "epoch": 3.078738093157389,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004062260762744825,
      "loss": 5.2077,
      "mean_token_accuracy": 0.20810189694166184,
      "num_tokens": 71388512.0,
      "step": 35230
    },
    {
      "entropy": 6.030933570861817,
      "epoch": 3.0791750415100934,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040620076256903256,
      "loss": 5.2336,
      "mean_token_accuracy": 0.2056839257478714,
      "num_tokens": 71397908.0,
      "step": 35235
    },
    {
      "entropy": 6.094190359115601,
      "epoch": 3.0796119898627983,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004061754463470833,
      "loss": 5.306,
      "mean_token_accuracy": 0.20198949724435805,
      "num_tokens": 71408287.0,
      "step": 35240
    },
    {
      "entropy": 6.054951906204224,
      "epoch": 3.0800489382155027,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004061501276091202,
      "loss": 5.2914,
      "mean_token_accuracy": 0.2047638013958931,
      "num_tokens": 71419227.0,
      "step": 35245
    },
    {
      "entropy": 6.068400287628174,
      "epoch": 3.0804858865682077,
      "grad_norm": 1.28125,
      "learning_rate": 0.000406124806355629,
      "loss": 5.2342,
      "mean_token_accuracy": 0.2101852148771286,
      "num_tokens": 71429398.0,
      "step": 35250
    },
    {
      "entropy": 6.0089493751525875,
      "epoch": 3.080922834920912,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040609948258709517,
      "loss": 5.2069,
      "mean_token_accuracy": 0.21311398297548295,
      "num_tokens": 71440185.0,
      "step": 35255
    },
    {
      "entropy": 5.916995620727539,
      "epoch": 3.081359783273617,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004060741563040047,
      "loss": 5.1863,
      "mean_token_accuracy": 0.2196153774857521,
      "num_tokens": 71451160.0,
      "step": 35260
    },
    {
      "entropy": 6.044189167022705,
      "epoch": 3.081796731626322,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004060488275068432,
      "loss": 5.2761,
      "mean_token_accuracy": 0.20288509875535965,
      "num_tokens": 71461459.0,
      "step": 35265
    },
    {
      "entropy": 6.007672262191773,
      "epoch": 3.0822336799790264,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040602349619609646,
      "loss": 5.2035,
      "mean_token_accuracy": 0.20832560658454896,
      "num_tokens": 71470806.0,
      "step": 35270
    },
    {
      "entropy": 5.971236705780029,
      "epoch": 3.0826706283317313,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040599816237225044,
      "loss": 5.1927,
      "mean_token_accuracy": 0.2122786298394203,
      "num_tokens": 71480929.0,
      "step": 35275
    },
    {
      "entropy": 6.066132402420044,
      "epoch": 3.083107576684436,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040597282603579103,
      "loss": 5.2314,
      "mean_token_accuracy": 0.21009001582860948,
      "num_tokens": 71490360.0,
      "step": 35280
    },
    {
      "entropy": 6.020212078094483,
      "epoch": 3.0835445250371407,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040594748718720424,
      "loss": 5.2765,
      "mean_token_accuracy": 0.20450647324323654,
      "num_tokens": 71500355.0,
      "step": 35285
    },
    {
      "entropy": 6.000193786621094,
      "epoch": 3.083981473389845,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040592214582697595,
      "loss": 5.2054,
      "mean_token_accuracy": 0.21377289593219756,
      "num_tokens": 71509581.0,
      "step": 35290
    },
    {
      "entropy": 6.019990396499634,
      "epoch": 3.08441842174255,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004058968019555923,
      "loss": 5.2562,
      "mean_token_accuracy": 0.20292546451091767,
      "num_tokens": 71519967.0,
      "step": 35295
    },
    {
      "entropy": 5.961777591705323,
      "epoch": 3.0848553700952546,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040587145557353943,
      "loss": 5.1684,
      "mean_token_accuracy": 0.21345409452915193,
      "num_tokens": 71530619.0,
      "step": 35300
    },
    {
      "entropy": 6.0125480651855465,
      "epoch": 3.0852923184479595,
      "grad_norm": 1.25,
      "learning_rate": 0.0004058461066813035,
      "loss": 5.2614,
      "mean_token_accuracy": 0.20366816222667694,
      "num_tokens": 71541682.0,
      "step": 35305
    },
    {
      "entropy": 6.014077997207641,
      "epoch": 3.085729266800664,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040582075527937066,
      "loss": 5.2024,
      "mean_token_accuracy": 0.2110506147146225,
      "num_tokens": 71552971.0,
      "step": 35310
    },
    {
      "entropy": 5.985378646850586,
      "epoch": 3.086166215153369,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004057954013682272,
      "loss": 5.2165,
      "mean_token_accuracy": 0.21821094900369645,
      "num_tokens": 71562932.0,
      "step": 35315
    },
    {
      "entropy": 6.023525762557983,
      "epoch": 3.0866031635060738,
      "grad_norm": 1.375,
      "learning_rate": 0.00040577004494835934,
      "loss": 5.1769,
      "mean_token_accuracy": 0.21321842074394226,
      "num_tokens": 71572662.0,
      "step": 35320
    },
    {
      "entropy": 6.006003284454346,
      "epoch": 3.0870401118587782,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040574468602025355,
      "loss": 5.1979,
      "mean_token_accuracy": 0.21367767453193665,
      "num_tokens": 71582503.0,
      "step": 35325
    },
    {
      "entropy": 6.075115442276001,
      "epoch": 3.087477060211483,
      "grad_norm": 1.34375,
      "learning_rate": 0.00040571932458439617,
      "loss": 5.2522,
      "mean_token_accuracy": 0.2061188906431198,
      "num_tokens": 71592261.0,
      "step": 35330
    },
    {
      "entropy": 6.016394376754761,
      "epoch": 3.0879140085641876,
      "grad_norm": 1.25,
      "learning_rate": 0.00040569396064127355,
      "loss": 5.2029,
      "mean_token_accuracy": 0.21774541288614274,
      "num_tokens": 71602514.0,
      "step": 35335
    },
    {
      "entropy": 5.964048862457275,
      "epoch": 3.0883509569168925,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004056685941913724,
      "loss": 5.2326,
      "mean_token_accuracy": 0.21718014925718307,
      "num_tokens": 71612912.0,
      "step": 35340
    },
    {
      "entropy": 5.97691388130188,
      "epoch": 3.088787905269597,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040564322523517905,
      "loss": 5.1456,
      "mean_token_accuracy": 0.215986068546772,
      "num_tokens": 71622673.0,
      "step": 35345
    },
    {
      "entropy": 6.075341653823853,
      "epoch": 3.089224853622302,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004056178537731803,
      "loss": 5.2042,
      "mean_token_accuracy": 0.2084778994321823,
      "num_tokens": 71633315.0,
      "step": 35350
    },
    {
      "entropy": 6.039299631118775,
      "epoch": 3.0896618019750064,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004055924798058626,
      "loss": 5.2754,
      "mean_token_accuracy": 0.20532296746969222,
      "num_tokens": 71644159.0,
      "step": 35355
    },
    {
      "entropy": 6.012940073013306,
      "epoch": 3.0900987503277113,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040556710333371275,
      "loss": 5.2092,
      "mean_token_accuracy": 0.2093348503112793,
      "num_tokens": 71654227.0,
      "step": 35360
    },
    {
      "entropy": 5.992197418212891,
      "epoch": 3.0905356986804158,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004055417243572174,
      "loss": 5.1605,
      "mean_token_accuracy": 0.2140216663479805,
      "num_tokens": 71664711.0,
      "step": 35365
    },
    {
      "entropy": 6.062667179107666,
      "epoch": 3.0909726470331207,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040551634287686343,
      "loss": 5.223,
      "mean_token_accuracy": 0.2138304129242897,
      "num_tokens": 71674880.0,
      "step": 35370
    },
    {
      "entropy": 6.072262477874756,
      "epoch": 3.0914095953858256,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004054909588931375,
      "loss": 5.2667,
      "mean_token_accuracy": 0.2013191819190979,
      "num_tokens": 71685386.0,
      "step": 35375
    },
    {
      "entropy": 6.052088117599487,
      "epoch": 3.09184654373853,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004054655724065268,
      "loss": 5.3496,
      "mean_token_accuracy": 0.19947656095027924,
      "num_tokens": 71696008.0,
      "step": 35380
    },
    {
      "entropy": 6.009516859054566,
      "epoch": 3.092283492091235,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004054401834175179,
      "loss": 5.1801,
      "mean_token_accuracy": 0.21114340871572496,
      "num_tokens": 71705548.0,
      "step": 35385
    },
    {
      "entropy": 5.97347583770752,
      "epoch": 3.0927204404439395,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040541479192659797,
      "loss": 5.2298,
      "mean_token_accuracy": 0.2080102816224098,
      "num_tokens": 71716891.0,
      "step": 35390
    },
    {
      "entropy": 6.068667984008789,
      "epoch": 3.0931573887966444,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040538939793425394,
      "loss": 5.2709,
      "mean_token_accuracy": 0.20601681917905806,
      "num_tokens": 71727954.0,
      "step": 35395
    },
    {
      "entropy": 6.015737867355346,
      "epoch": 3.093594337149349,
      "grad_norm": 1.203125,
      "learning_rate": 0.000405364001440973,
      "loss": 5.2737,
      "mean_token_accuracy": 0.2034907341003418,
      "num_tokens": 71738282.0,
      "step": 35400
    },
    {
      "entropy": 6.0878458499908445,
      "epoch": 3.0940312855020538,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004053386024472422,
      "loss": 5.2918,
      "mean_token_accuracy": 0.202457395195961,
      "num_tokens": 71747895.0,
      "step": 35405
    },
    {
      "entropy": 6.004248094558716,
      "epoch": 3.094468233854758,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004053132009535487,
      "loss": 5.2388,
      "mean_token_accuracy": 0.20858814418315888,
      "num_tokens": 71758107.0,
      "step": 35410
    },
    {
      "entropy": 6.062949514389038,
      "epoch": 3.094905182207463,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004052877969603797,
      "loss": 5.3084,
      "mean_token_accuracy": 0.19861139953136445,
      "num_tokens": 71767661.0,
      "step": 35415
    },
    {
      "entropy": 6.060808372497559,
      "epoch": 3.0953421305601676,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004052623904682225,
      "loss": 5.2584,
      "mean_token_accuracy": 0.2059008374810219,
      "num_tokens": 71777908.0,
      "step": 35420
    },
    {
      "entropy": 6.0312704086303714,
      "epoch": 3.0957790789128725,
      "grad_norm": 1.375,
      "learning_rate": 0.00040523698147756443,
      "loss": 5.2009,
      "mean_token_accuracy": 0.21320036947727203,
      "num_tokens": 71786809.0,
      "step": 35425
    },
    {
      "entropy": 5.9630097389221195,
      "epoch": 3.0962160272655774,
      "grad_norm": 1.25,
      "learning_rate": 0.00040521156998889283,
      "loss": 5.2195,
      "mean_token_accuracy": 0.20651869028806685,
      "num_tokens": 71797299.0,
      "step": 35430
    },
    {
      "entropy": 6.0179784297943115,
      "epoch": 3.096652975618282,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040518615600269506,
      "loss": 5.2102,
      "mean_token_accuracy": 0.2068721607327461,
      "num_tokens": 71807565.0,
      "step": 35435
    },
    {
      "entropy": 5.997705698013306,
      "epoch": 3.097089923970987,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004051607395194586,
      "loss": 5.2096,
      "mean_token_accuracy": 0.21258843094110488,
      "num_tokens": 71817288.0,
      "step": 35440
    },
    {
      "entropy": 6.029693555831909,
      "epoch": 3.0975268723236913,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00040513532053967086,
      "loss": 5.1959,
      "mean_token_accuracy": 0.21290574967861176,
      "num_tokens": 71827371.0,
      "step": 35445
    },
    {
      "entropy": 6.098662185668945,
      "epoch": 3.097963820676396,
      "grad_norm": 1.25,
      "learning_rate": 0.00040510989906381955,
      "loss": 5.3551,
      "mean_token_accuracy": 0.19999365508556366,
      "num_tokens": 71837872.0,
      "step": 35450
    },
    {
      "entropy": 5.965522766113281,
      "epoch": 3.0984007690291007,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040508447509239224,
      "loss": 5.1574,
      "mean_token_accuracy": 0.21903432309627532,
      "num_tokens": 71848507.0,
      "step": 35455
    },
    {
      "entropy": 6.029659366607666,
      "epoch": 3.0988377173818056,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004050590486258764,
      "loss": 5.2675,
      "mean_token_accuracy": 0.19831760823726655,
      "num_tokens": 71860495.0,
      "step": 35460
    },
    {
      "entropy": 6.057018280029297,
      "epoch": 3.09927466573451,
      "grad_norm": 1.78125,
      "learning_rate": 0.00040503361966476005,
      "loss": 5.1102,
      "mean_token_accuracy": 0.22711047381162644,
      "num_tokens": 71870081.0,
      "step": 35465
    },
    {
      "entropy": 6.018096017837524,
      "epoch": 3.099711614087215,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004050081882095307,
      "loss": 5.2204,
      "mean_token_accuracy": 0.20631353706121444,
      "num_tokens": 71880072.0,
      "step": 35470
    },
    {
      "entropy": 5.98187108039856,
      "epoch": 3.1001485624399194,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004049827542606761,
      "loss": 5.209,
      "mean_token_accuracy": 0.20812605917453766,
      "num_tokens": 71889225.0,
      "step": 35475
    },
    {
      "entropy": 6.052856683731079,
      "epoch": 3.1005855107926243,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004049573178186842,
      "loss": 5.2313,
      "mean_token_accuracy": 0.2155118241906166,
      "num_tokens": 71898601.0,
      "step": 35480
    },
    {
      "entropy": 6.063455963134766,
      "epoch": 3.101022459145329,
      "grad_norm": 1.125,
      "learning_rate": 0.00040493187888404283,
      "loss": 5.2359,
      "mean_token_accuracy": 0.20553145706653594,
      "num_tokens": 71909267.0,
      "step": 35485
    },
    {
      "entropy": 6.072769594192505,
      "epoch": 3.1014594074980337,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040490643745724,
      "loss": 5.2495,
      "mean_token_accuracy": 0.2047132208943367,
      "num_tokens": 71919066.0,
      "step": 35490
    },
    {
      "entropy": 6.024191045761109,
      "epoch": 3.1018963558507386,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004048809935387635,
      "loss": 5.2018,
      "mean_token_accuracy": 0.209603451192379,
      "num_tokens": 71929067.0,
      "step": 35495
    },
    {
      "entropy": 6.06879358291626,
      "epoch": 3.102333304203443,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040485554712910155,
      "loss": 5.1502,
      "mean_token_accuracy": 0.20849586874246598,
      "num_tokens": 71938044.0,
      "step": 35500
    },
    {
      "entropy": 5.996237707138062,
      "epoch": 3.102770252556148,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040483009822874216,
      "loss": 5.2567,
      "mean_token_accuracy": 0.20862601548433304,
      "num_tokens": 71947714.0,
      "step": 35505
    },
    {
      "entropy": 5.993281364440918,
      "epoch": 3.1032072009088525,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004048046468381735,
      "loss": 5.2007,
      "mean_token_accuracy": 0.21347277611494064,
      "num_tokens": 71957660.0,
      "step": 35510
    },
    {
      "entropy": 5.983157444000244,
      "epoch": 3.1036441492615574,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004047791929578837,
      "loss": 5.1566,
      "mean_token_accuracy": 0.2184414491057396,
      "num_tokens": 71967063.0,
      "step": 35515
    },
    {
      "entropy": 5.954027462005615,
      "epoch": 3.104081097614262,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004047537365883609,
      "loss": 5.238,
      "mean_token_accuracy": 0.20868772119283677,
      "num_tokens": 71977331.0,
      "step": 35520
    },
    {
      "entropy": 5.957418727874756,
      "epoch": 3.104518045966967,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040472827773009346,
      "loss": 5.1559,
      "mean_token_accuracy": 0.22018727660179138,
      "num_tokens": 71987047.0,
      "step": 35525
    },
    {
      "entropy": 6.030376291275024,
      "epoch": 3.1049549943196713,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040470281638356974,
      "loss": 5.1764,
      "mean_token_accuracy": 0.20421637892723082,
      "num_tokens": 71997057.0,
      "step": 35530
    },
    {
      "entropy": 6.039985656738281,
      "epoch": 3.105391942672376,
      "grad_norm": 1.28125,
      "learning_rate": 0.000404677352549278,
      "loss": 5.2759,
      "mean_token_accuracy": 0.20468209385871888,
      "num_tokens": 72007092.0,
      "step": 35535
    },
    {
      "entropy": 6.033021020889282,
      "epoch": 3.1058288910250806,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040465188622770663,
      "loss": 5.3104,
      "mean_token_accuracy": 0.20367663651704787,
      "num_tokens": 72016847.0,
      "step": 35540
    },
    {
      "entropy": 5.99982705116272,
      "epoch": 3.1062658393777856,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004046264174193443,
      "loss": 5.2132,
      "mean_token_accuracy": 0.20842845290899276,
      "num_tokens": 72026706.0,
      "step": 35545
    },
    {
      "entropy": 6.1082823276519775,
      "epoch": 3.1067027877304905,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004046009461246792,
      "loss": 5.3324,
      "mean_token_accuracy": 0.19781127125024794,
      "num_tokens": 72036358.0,
      "step": 35550
    },
    {
      "entropy": 6.0485859394073485,
      "epoch": 3.107139736083195,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004045754723442001,
      "loss": 5.3059,
      "mean_token_accuracy": 0.19911132007837296,
      "num_tokens": 72049156.0,
      "step": 35555
    },
    {
      "entropy": 5.995420169830322,
      "epoch": 3.1075766844359,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004045499960783955,
      "loss": 5.1765,
      "mean_token_accuracy": 0.21164429634809495,
      "num_tokens": 72058986.0,
      "step": 35560
    },
    {
      "entropy": 6.01118311882019,
      "epoch": 3.1080136327886043,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004045245173277543,
      "loss": 5.2064,
      "mean_token_accuracy": 0.2135972261428833,
      "num_tokens": 72069141.0,
      "step": 35565
    },
    {
      "entropy": 5.962675094604492,
      "epoch": 3.1084505811413092,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040449903609276474,
      "loss": 5.1615,
      "mean_token_accuracy": 0.22182875424623488,
      "num_tokens": 72079009.0,
      "step": 35570
    },
    {
      "entropy": 6.013104200363159,
      "epoch": 3.1088875294940137,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004044735523739159,
      "loss": 5.2431,
      "mean_token_accuracy": 0.20345504730939865,
      "num_tokens": 72089551.0,
      "step": 35575
    },
    {
      "entropy": 6.062737894058228,
      "epoch": 3.1093244778467186,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040444806617169655,
      "loss": 5.2154,
      "mean_token_accuracy": 0.2108215108513832,
      "num_tokens": 72098965.0,
      "step": 35580
    },
    {
      "entropy": 6.035129070281982,
      "epoch": 3.109761426199423,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040442257748659537,
      "loss": 5.2487,
      "mean_token_accuracy": 0.21122245341539383,
      "num_tokens": 72109568.0,
      "step": 35585
    },
    {
      "entropy": 5.985316610336303,
      "epoch": 3.110198374552128,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040439708631910136,
      "loss": 5.2317,
      "mean_token_accuracy": 0.2109475702047348,
      "num_tokens": 72119988.0,
      "step": 35590
    },
    {
      "entropy": 5.9615209102630615,
      "epoch": 3.1106353229048325,
      "grad_norm": 1.125,
      "learning_rate": 0.0004043715926697034,
      "loss": 5.3018,
      "mean_token_accuracy": 0.209038408100605,
      "num_tokens": 72130515.0,
      "step": 35595
    },
    {
      "entropy": 6.095059108734131,
      "epoch": 3.1110722712575374,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004043460965388906,
      "loss": 5.2539,
      "mean_token_accuracy": 0.20233892649412155,
      "num_tokens": 72140362.0,
      "step": 35600
    },
    {
      "entropy": 6.022561740875244,
      "epoch": 3.111509219610242,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004043205979271518,
      "loss": 5.2313,
      "mean_token_accuracy": 0.2102019041776657,
      "num_tokens": 72150726.0,
      "step": 35605
    },
    {
      "entropy": 6.008059024810791,
      "epoch": 3.1119461679629468,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004042950968349762,
      "loss": 5.19,
      "mean_token_accuracy": 0.21488413214683533,
      "num_tokens": 72160264.0,
      "step": 35610
    },
    {
      "entropy": 5.996623897552491,
      "epoch": 3.1123831163156517,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004042695932628529,
      "loss": 5.2514,
      "mean_token_accuracy": 0.2069787234067917,
      "num_tokens": 72171407.0,
      "step": 35615
    },
    {
      "entropy": 6.047266721725464,
      "epoch": 3.112820064668356,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004042440872112711,
      "loss": 5.2833,
      "mean_token_accuracy": 0.20628097504377366,
      "num_tokens": 72182132.0,
      "step": 35620
    },
    {
      "entropy": 6.034156847000122,
      "epoch": 3.113257013021061,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040421857868071996,
      "loss": 5.2429,
      "mean_token_accuracy": 0.2010910615324974,
      "num_tokens": 72192129.0,
      "step": 35625
    },
    {
      "entropy": 5.9754609107971195,
      "epoch": 3.1136939613737655,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040419306767168874,
      "loss": 5.1981,
      "mean_token_accuracy": 0.21330669522285461,
      "num_tokens": 72202053.0,
      "step": 35630
    },
    {
      "entropy": 6.032208871841431,
      "epoch": 3.1141309097264704,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004041675541846668,
      "loss": 5.1963,
      "mean_token_accuracy": 0.21216461658477784,
      "num_tokens": 72211352.0,
      "step": 35635
    },
    {
      "entropy": 6.007864141464234,
      "epoch": 3.114567858079175,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00040414203822014343,
      "loss": 5.1859,
      "mean_token_accuracy": 0.20958243161439896,
      "num_tokens": 72220816.0,
      "step": 35640
    },
    {
      "entropy": 6.049524879455566,
      "epoch": 3.11500480643188,
      "grad_norm": 1.25,
      "learning_rate": 0.00040411651977860817,
      "loss": 5.2843,
      "mean_token_accuracy": 0.19933838099241258,
      "num_tokens": 72231318.0,
      "step": 35645
    },
    {
      "entropy": 6.014575815200805,
      "epoch": 3.1154417547845843,
      "grad_norm": 1.34375,
      "learning_rate": 0.00040409099886055044,
      "loss": 5.2174,
      "mean_token_accuracy": 0.20998117625713347,
      "num_tokens": 72241035.0,
      "step": 35650
    },
    {
      "entropy": 6.079341220855713,
      "epoch": 3.115878703137289,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004040654754664596,
      "loss": 5.1664,
      "mean_token_accuracy": 0.20289439857006072,
      "num_tokens": 72249967.0,
      "step": 35655
    },
    {
      "entropy": 6.0470929622650145,
      "epoch": 3.1163156514899937,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040403994959682546,
      "loss": 5.2682,
      "mean_token_accuracy": 0.2079853281378746,
      "num_tokens": 72260114.0,
      "step": 35660
    },
    {
      "entropy": 6.037948322296143,
      "epoch": 3.1167525998426986,
      "grad_norm": 1.375,
      "learning_rate": 0.00040401442125213735,
      "loss": 5.2016,
      "mean_token_accuracy": 0.2055423989892006,
      "num_tokens": 72270257.0,
      "step": 35665
    },
    {
      "entropy": 6.039092063903809,
      "epoch": 3.1171895481954035,
      "grad_norm": 1.359375,
      "learning_rate": 0.00040398889043288507,
      "loss": 5.1704,
      "mean_token_accuracy": 0.2076101765036583,
      "num_tokens": 72279628.0,
      "step": 35670
    },
    {
      "entropy": 6.006291913986206,
      "epoch": 3.117626496548108,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004039633571395583,
      "loss": 5.2129,
      "mean_token_accuracy": 0.20878246128559114,
      "num_tokens": 72289493.0,
      "step": 35675
    },
    {
      "entropy": 5.967185163497925,
      "epoch": 3.118063444900813,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040393782137264676,
      "loss": 5.1177,
      "mean_token_accuracy": 0.22080607563257218,
      "num_tokens": 72300007.0,
      "step": 35680
    },
    {
      "entropy": 6.005072546005249,
      "epoch": 3.1185003932535174,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004039122831326402,
      "loss": 5.186,
      "mean_token_accuracy": 0.21216347962617874,
      "num_tokens": 72310596.0,
      "step": 35685
    },
    {
      "entropy": 6.040719366073608,
      "epoch": 3.1189373416062223,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004038867424200285,
      "loss": 5.271,
      "mean_token_accuracy": 0.20900741517543792,
      "num_tokens": 72320156.0,
      "step": 35690
    },
    {
      "entropy": 6.038656425476074,
      "epoch": 3.1193742899589267,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004038611992353016,
      "loss": 5.207,
      "mean_token_accuracy": 0.20911463648080825,
      "num_tokens": 72329608.0,
      "step": 35695
    },
    {
      "entropy": 6.005929231643677,
      "epoch": 3.1198112383116317,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004038356535789494,
      "loss": 5.1722,
      "mean_token_accuracy": 0.21747828423976898,
      "num_tokens": 72339565.0,
      "step": 35700
    },
    {
      "entropy": 6.016811037063599,
      "epoch": 3.120248186664336,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040381010545146174,
      "loss": 5.2795,
      "mean_token_accuracy": 0.20711213052272798,
      "num_tokens": 72349566.0,
      "step": 35705
    },
    {
      "entropy": 6.015627384185791,
      "epoch": 3.120685135017041,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004037845548533289,
      "loss": 5.1776,
      "mean_token_accuracy": 0.21411222666502,
      "num_tokens": 72360366.0,
      "step": 35710
    },
    {
      "entropy": 6.084981966018677,
      "epoch": 3.1211220833697455,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040375900178504066,
      "loss": 5.3387,
      "mean_token_accuracy": 0.20667025446891785,
      "num_tokens": 72369893.0,
      "step": 35715
    },
    {
      "entropy": 6.017996072769165,
      "epoch": 3.1215590317224504,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004037334462470874,
      "loss": 5.2573,
      "mean_token_accuracy": 0.20346357226371764,
      "num_tokens": 72380484.0,
      "step": 35720
    },
    {
      "entropy": 6.025841760635376,
      "epoch": 3.1219959800751553,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004037078882399591,
      "loss": 5.172,
      "mean_token_accuracy": 0.21428311467170716,
      "num_tokens": 72391088.0,
      "step": 35725
    },
    {
      "entropy": 6.065987014770508,
      "epoch": 3.12243292842786,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004036823277641462,
      "loss": 5.2861,
      "mean_token_accuracy": 0.20000345706939698,
      "num_tokens": 72401253.0,
      "step": 35730
    },
    {
      "entropy": 6.044170141220093,
      "epoch": 3.1228698767805647,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040365676482013877,
      "loss": 5.1785,
      "mean_token_accuracy": 0.20945405066013337,
      "num_tokens": 72411235.0,
      "step": 35735
    },
    {
      "entropy": 6.052325439453125,
      "epoch": 3.123306825133269,
      "grad_norm": 1.25,
      "learning_rate": 0.0004036311994084271,
      "loss": 5.2853,
      "mean_token_accuracy": 0.20147834867238998,
      "num_tokens": 72421387.0,
      "step": 35740
    },
    {
      "entropy": 6.030856800079346,
      "epoch": 3.123743773485974,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040360563152950173,
      "loss": 5.2614,
      "mean_token_accuracy": 0.20649877786636353,
      "num_tokens": 72431431.0,
      "step": 35745
    },
    {
      "entropy": 6.042762374877929,
      "epoch": 3.1241807218386786,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004035800611838529,
      "loss": 5.2599,
      "mean_token_accuracy": 0.20651472359895706,
      "num_tokens": 72441286.0,
      "step": 35750
    },
    {
      "entropy": 6.03114504814148,
      "epoch": 3.1246176701913835,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040355448837197115,
      "loss": 5.2437,
      "mean_token_accuracy": 0.2050869345664978,
      "num_tokens": 72451661.0,
      "step": 35755
    },
    {
      "entropy": 6.004827642440796,
      "epoch": 3.125054618544088,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040352891309434693,
      "loss": 5.241,
      "mean_token_accuracy": 0.20289792269468307,
      "num_tokens": 72462013.0,
      "step": 35760
    },
    {
      "entropy": 5.986980724334717,
      "epoch": 3.125491566896793,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004035033353514708,
      "loss": 5.1611,
      "mean_token_accuracy": 0.2093708485364914,
      "num_tokens": 72472804.0,
      "step": 35765
    },
    {
      "entropy": 6.057681560516357,
      "epoch": 3.1259285152494973,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040347775514383344,
      "loss": 5.2219,
      "mean_token_accuracy": 0.20812643617391585,
      "num_tokens": 72482893.0,
      "step": 35770
    },
    {
      "entropy": 6.0173095703125,
      "epoch": 3.1263654636022022,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040345217247192536,
      "loss": 5.2129,
      "mean_token_accuracy": 0.20861918479204178,
      "num_tokens": 72492882.0,
      "step": 35775
    },
    {
      "entropy": 5.936790609359742,
      "epoch": 3.126802411954907,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004034265873362373,
      "loss": 5.1576,
      "mean_token_accuracy": 0.21425104588270188,
      "num_tokens": 72502322.0,
      "step": 35780
    },
    {
      "entropy": 6.025757265090943,
      "epoch": 3.1272393603076116,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040340099973726,
      "loss": 5.2426,
      "mean_token_accuracy": 0.20877794176340103,
      "num_tokens": 72511401.0,
      "step": 35785
    },
    {
      "entropy": 6.030284738540649,
      "epoch": 3.1276763086603165,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040337540967548427,
      "loss": 5.2413,
      "mean_token_accuracy": 0.2116603970527649,
      "num_tokens": 72522071.0,
      "step": 35790
    },
    {
      "entropy": 6.056809043884277,
      "epoch": 3.128113257013021,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040334981715140086,
      "loss": 5.3034,
      "mean_token_accuracy": 0.20534511506557465,
      "num_tokens": 72532899.0,
      "step": 35795
    },
    {
      "entropy": 6.052888441085815,
      "epoch": 3.128550205365726,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040332422216550067,
      "loss": 5.267,
      "mean_token_accuracy": 0.2066643938422203,
      "num_tokens": 72543372.0,
      "step": 35800
    },
    {
      "entropy": 6.032076835632324,
      "epoch": 3.1289871537184304,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004032986247182748,
      "loss": 5.2442,
      "mean_token_accuracy": 0.2078570008277893,
      "num_tokens": 72553014.0,
      "step": 35805
    },
    {
      "entropy": 5.982944345474243,
      "epoch": 3.1294241020711353,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040327302481021394,
      "loss": 5.1645,
      "mean_token_accuracy": 0.21645728200674058,
      "num_tokens": 72562281.0,
      "step": 35810
    },
    {
      "entropy": 5.991694784164428,
      "epoch": 3.1298610504238398,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00040324742244180937,
      "loss": 5.1548,
      "mean_token_accuracy": 0.21279390752315522,
      "num_tokens": 72571817.0,
      "step": 35815
    },
    {
      "entropy": 5.98730320930481,
      "epoch": 3.1302979987765447,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040322181761355194,
      "loss": 5.2043,
      "mean_token_accuracy": 0.21460814476013185,
      "num_tokens": 72582935.0,
      "step": 35820
    },
    {
      "entropy": 6.020169496536255,
      "epoch": 3.130734947129249,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004031962103259329,
      "loss": 5.2202,
      "mean_token_accuracy": 0.21587630212306977,
      "num_tokens": 72593280.0,
      "step": 35825
    },
    {
      "entropy": 5.980169820785522,
      "epoch": 3.131171895481954,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004031706005794434,
      "loss": 5.179,
      "mean_token_accuracy": 0.20740045458078385,
      "num_tokens": 72603193.0,
      "step": 35830
    },
    {
      "entropy": 5.977092361450195,
      "epoch": 3.1316088438346585,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004031449883745746,
      "loss": 5.2565,
      "mean_token_accuracy": 0.21288931965827942,
      "num_tokens": 72613363.0,
      "step": 35835
    },
    {
      "entropy": 6.056025791168213,
      "epoch": 3.1320457921873635,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040311937371181774,
      "loss": 5.3899,
      "mean_token_accuracy": 0.19665200710296632,
      "num_tokens": 72624151.0,
      "step": 35840
    },
    {
      "entropy": 6.046934986114502,
      "epoch": 3.1324827405400684,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040309375659166414,
      "loss": 5.1555,
      "mean_token_accuracy": 0.22092944532632827,
      "num_tokens": 72633692.0,
      "step": 35845
    },
    {
      "entropy": 6.002097749710083,
      "epoch": 3.132919688892773,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040306813701460525,
      "loss": 5.2512,
      "mean_token_accuracy": 0.20525816082954407,
      "num_tokens": 72644016.0,
      "step": 35850
    },
    {
      "entropy": 5.977407646179199,
      "epoch": 3.1333566372454777,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004030425149811324,
      "loss": 5.2424,
      "mean_token_accuracy": 0.21118716448545455,
      "num_tokens": 72654033.0,
      "step": 35855
    },
    {
      "entropy": 6.055842065811158,
      "epoch": 3.133793585598182,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004030168904917369,
      "loss": 5.2135,
      "mean_token_accuracy": 0.19989307820796967,
      "num_tokens": 72663452.0,
      "step": 35860
    },
    {
      "entropy": 6.0575953960418705,
      "epoch": 3.134230533950887,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040299126354691043,
      "loss": 5.2764,
      "mean_token_accuracy": 0.20706392228603362,
      "num_tokens": 72674766.0,
      "step": 35865
    },
    {
      "entropy": 6.059757471084595,
      "epoch": 3.1346674823035916,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040296563414714444,
      "loss": 5.2455,
      "mean_token_accuracy": 0.2102657899260521,
      "num_tokens": 72683816.0,
      "step": 35870
    },
    {
      "entropy": 5.996105194091797,
      "epoch": 3.1351044306562965,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004029400022929306,
      "loss": 5.1681,
      "mean_token_accuracy": 0.21678656339645386,
      "num_tokens": 72693604.0,
      "step": 35875
    },
    {
      "entropy": 5.9949122905731205,
      "epoch": 3.135541379009001,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004029143679847604,
      "loss": 5.2334,
      "mean_token_accuracy": 0.21108418256044387,
      "num_tokens": 72703936.0,
      "step": 35880
    },
    {
      "entropy": 6.0471498489379885,
      "epoch": 3.135978327361706,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004028887312231256,
      "loss": 5.2001,
      "mean_token_accuracy": 0.21284110546112062,
      "num_tokens": 72713646.0,
      "step": 35885
    },
    {
      "entropy": 6.03625078201294,
      "epoch": 3.1364152757144104,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040286309200851793,
      "loss": 5.1583,
      "mean_token_accuracy": 0.21291491836309434,
      "num_tokens": 72723573.0,
      "step": 35890
    },
    {
      "entropy": 6.0308146476745605,
      "epoch": 3.1368522240671153,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004028374503414292,
      "loss": 5.2826,
      "mean_token_accuracy": 0.20338870733976364,
      "num_tokens": 72733805.0,
      "step": 35895
    },
    {
      "entropy": 5.988727569580078,
      "epoch": 3.1372891724198197,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004028118062223512,
      "loss": 5.1771,
      "mean_token_accuracy": 0.21397020071744918,
      "num_tokens": 72744254.0,
      "step": 35900
    },
    {
      "entropy": 6.065953207015991,
      "epoch": 3.1377261207725247,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004027861596517757,
      "loss": 5.2693,
      "mean_token_accuracy": 0.19957565069198607,
      "num_tokens": 72754267.0,
      "step": 35905
    },
    {
      "entropy": 5.854194402694702,
      "epoch": 3.1381630691252296,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004027605106301948,
      "loss": 5.1388,
      "mean_token_accuracy": 0.22477860003709793,
      "num_tokens": 72764598.0,
      "step": 35910
    },
    {
      "entropy": 6.018028306961059,
      "epoch": 3.138600017477934,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040273485915810026,
      "loss": 5.2673,
      "mean_token_accuracy": 0.20320285856723785,
      "num_tokens": 72774145.0,
      "step": 35915
    },
    {
      "entropy": 6.128881406784058,
      "epoch": 3.139036965830639,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040270920523598425,
      "loss": 5.3473,
      "mean_token_accuracy": 0.19538600593805314,
      "num_tokens": 72783766.0,
      "step": 35920
    },
    {
      "entropy": 6.071776103973389,
      "epoch": 3.1394739141833434,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004026835488643387,
      "loss": 5.2137,
      "mean_token_accuracy": 0.21833940893411635,
      "num_tokens": 72793907.0,
      "step": 35925
    },
    {
      "entropy": 6.069908952713012,
      "epoch": 3.1399108625360483,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040265789004365586,
      "loss": 5.3309,
      "mean_token_accuracy": 0.19862293004989623,
      "num_tokens": 72802927.0,
      "step": 35930
    },
    {
      "entropy": 5.9121935844421385,
      "epoch": 3.140347810888753,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004026322287744277,
      "loss": 5.0947,
      "mean_token_accuracy": 0.2228113368153572,
      "num_tokens": 72813387.0,
      "step": 35935
    },
    {
      "entropy": 5.991809988021851,
      "epoch": 3.1407847592414577,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004026065650571466,
      "loss": 5.1683,
      "mean_token_accuracy": 0.21300536543130874,
      "num_tokens": 72824109.0,
      "step": 35940
    },
    {
      "entropy": 6.0380774974823,
      "epoch": 3.141221707594162,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004025808988923047,
      "loss": 5.1753,
      "mean_token_accuracy": 0.210610431432724,
      "num_tokens": 72834107.0,
      "step": 35945
    },
    {
      "entropy": 6.043208456039428,
      "epoch": 3.141658655946867,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040255523028039433,
      "loss": 5.1994,
      "mean_token_accuracy": 0.21187988072633743,
      "num_tokens": 72843726.0,
      "step": 35950
    },
    {
      "entropy": 6.030082702636719,
      "epoch": 3.1420956042995716,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040252955922190776,
      "loss": 5.286,
      "mean_token_accuracy": 0.20802717804908752,
      "num_tokens": 72854203.0,
      "step": 35955
    },
    {
      "entropy": 6.008528852462769,
      "epoch": 3.1425325526522765,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040250388571733744,
      "loss": 5.2207,
      "mean_token_accuracy": 0.205269019305706,
      "num_tokens": 72864689.0,
      "step": 35960
    },
    {
      "entropy": 5.980512571334839,
      "epoch": 3.1429695010049814,
      "grad_norm": 1.25,
      "learning_rate": 0.0004024782097671758,
      "loss": 5.2397,
      "mean_token_accuracy": 0.21287189722061156,
      "num_tokens": 72875040.0,
      "step": 35965
    },
    {
      "entropy": 6.003823375701904,
      "epoch": 3.143406449357686,
      "grad_norm": 1.25,
      "learning_rate": 0.0004024525313719153,
      "loss": 5.2223,
      "mean_token_accuracy": 0.2093981996178627,
      "num_tokens": 72885174.0,
      "step": 35970
    },
    {
      "entropy": 5.990322685241699,
      "epoch": 3.143843397710391,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040242685053204844,
      "loss": 5.2095,
      "mean_token_accuracy": 0.21005143374204635,
      "num_tokens": 72895344.0,
      "step": 35975
    },
    {
      "entropy": 6.100254917144776,
      "epoch": 3.1442803460630953,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004024011672480678,
      "loss": 5.2645,
      "mean_token_accuracy": 0.21050978899002076,
      "num_tokens": 72905950.0,
      "step": 35980
    },
    {
      "entropy": 6.069291591644287,
      "epoch": 3.1447172944158,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040237548152046603,
      "loss": 5.2332,
      "mean_token_accuracy": 0.21040431261062623,
      "num_tokens": 72916828.0,
      "step": 35985
    },
    {
      "entropy": 5.995930337905884,
      "epoch": 3.1451542427685046,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040234979334973577,
      "loss": 5.2744,
      "mean_token_accuracy": 0.21201652586460112,
      "num_tokens": 72926951.0,
      "step": 35990
    },
    {
      "entropy": 5.9421618461608885,
      "epoch": 3.1455911911212096,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004023241027363698,
      "loss": 5.1429,
      "mean_token_accuracy": 0.21444669514894485,
      "num_tokens": 72937909.0,
      "step": 35995
    },
    {
      "entropy": 6.025350141525268,
      "epoch": 3.146028139473914,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004022984096808608,
      "loss": 5.2211,
      "mean_token_accuracy": 0.20832187086343765,
      "num_tokens": 72949898.0,
      "step": 36000
    },
    {
      "epoch": 3.146028139473914,
      "eval_entropy": 5.7144571270931355,
      "eval_loss": 5.288760662078857,
      "eval_mean_token_accuracy": 0.21441947408737053,
      "eval_num_tokens": 72949898.0,
      "eval_runtime": 21.6021,
      "eval_samples_per_second": 1713.444,
      "eval_steps_per_second": 214.192,
      "step": 36000
    },
    {
      "entropy": 6.018662595748902,
      "epoch": 3.146465087826619,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040227271418370156,
      "loss": 5.2445,
      "mean_token_accuracy": 0.2091535970568657,
      "num_tokens": 72959643.0,
      "step": 36005
    },
    {
      "entropy": 5.987507390975952,
      "epoch": 3.1469020361793234,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000402247016245385,
      "loss": 5.227,
      "mean_token_accuracy": 0.20913294702768326,
      "num_tokens": 72969745.0,
      "step": 36010
    },
    {
      "entropy": 6.048810625076294,
      "epoch": 3.1473389845320283,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040222131586640397,
      "loss": 5.2185,
      "mean_token_accuracy": 0.208919158577919,
      "num_tokens": 72979892.0,
      "step": 36015
    },
    {
      "entropy": 6.022184801101685,
      "epoch": 3.1477759328847332,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040219561304725147,
      "loss": 5.2159,
      "mean_token_accuracy": 0.21614740937948226,
      "num_tokens": 72989526.0,
      "step": 36020
    },
    {
      "entropy": 5.98011999130249,
      "epoch": 3.1482128812374377,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040216990778842046,
      "loss": 5.2715,
      "mean_token_accuracy": 0.204580220580101,
      "num_tokens": 72999771.0,
      "step": 36025
    },
    {
      "entropy": 5.984656953811646,
      "epoch": 3.1486498295901426,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040214420009040394,
      "loss": 5.2011,
      "mean_token_accuracy": 0.21009377390146255,
      "num_tokens": 73011602.0,
      "step": 36030
    },
    {
      "entropy": 6.035901880264282,
      "epoch": 3.149086777942847,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000402118489953695,
      "loss": 5.2408,
      "mean_token_accuracy": 0.21202756613492965,
      "num_tokens": 73021838.0,
      "step": 36035
    },
    {
      "entropy": 6.046678686141968,
      "epoch": 3.149523726295552,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040209277737878683,
      "loss": 5.2301,
      "mean_token_accuracy": 0.2117896020412445,
      "num_tokens": 73032242.0,
      "step": 36040
    },
    {
      "entropy": 6.0710451126098635,
      "epoch": 3.1499606746482565,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004020670623661727,
      "loss": 5.3117,
      "mean_token_accuracy": 0.19735803753137587,
      "num_tokens": 73042903.0,
      "step": 36045
    },
    {
      "entropy": 6.014759349822998,
      "epoch": 3.1503976230009614,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004020413449163457,
      "loss": 5.157,
      "mean_token_accuracy": 0.2179866537451744,
      "num_tokens": 73054083.0,
      "step": 36050
    },
    {
      "entropy": 6.097563171386719,
      "epoch": 3.150834571353666,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004020156250297991,
      "loss": 5.2513,
      "mean_token_accuracy": 0.20054984390735625,
      "num_tokens": 73064215.0,
      "step": 36055
    },
    {
      "entropy": 5.920040225982666,
      "epoch": 3.1512715197063708,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004019899027070262,
      "loss": 5.1627,
      "mean_token_accuracy": 0.22033274173736572,
      "num_tokens": 73074319.0,
      "step": 36060
    },
    {
      "entropy": 6.000773906707764,
      "epoch": 3.1517084680590752,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004019641779485206,
      "loss": 5.2851,
      "mean_token_accuracy": 0.2044035777449608,
      "num_tokens": 73083972.0,
      "step": 36065
    },
    {
      "entropy": 6.0478147029876705,
      "epoch": 3.15214541641178,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040193845075477537,
      "loss": 5.2623,
      "mean_token_accuracy": 0.2032086506485939,
      "num_tokens": 73094495.0,
      "step": 36070
    },
    {
      "entropy": 5.974339723587036,
      "epoch": 3.152582364764485,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004019127211262842,
      "loss": 5.1029,
      "mean_token_accuracy": 0.22249964028596877,
      "num_tokens": 73104355.0,
      "step": 36075
    },
    {
      "entropy": 5.9823261260986325,
      "epoch": 3.1530193131171895,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040188698906354055,
      "loss": 5.1899,
      "mean_token_accuracy": 0.21401175409555434,
      "num_tokens": 73115795.0,
      "step": 36080
    },
    {
      "entropy": 6.024014568328857,
      "epoch": 3.1534562614698944,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004018612545670379,
      "loss": 5.1372,
      "mean_token_accuracy": 0.21943406164646148,
      "num_tokens": 73125351.0,
      "step": 36085
    },
    {
      "entropy": 5.972230291366577,
      "epoch": 3.153893209822599,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040183551763727,
      "loss": 5.2122,
      "mean_token_accuracy": 0.21284074932336808,
      "num_tokens": 73135087.0,
      "step": 36090
    },
    {
      "entropy": 5.932225370407105,
      "epoch": 3.154330158175304,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040180977827473035,
      "loss": 5.1918,
      "mean_token_accuracy": 0.2087901398539543,
      "num_tokens": 73145382.0,
      "step": 36095
    },
    {
      "entropy": 6.072624778747558,
      "epoch": 3.1547671065280083,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040178403647991276,
      "loss": 5.2909,
      "mean_token_accuracy": 0.1992565631866455,
      "num_tokens": 73155689.0,
      "step": 36100
    },
    {
      "entropy": 6.146011304855347,
      "epoch": 3.155204054880713,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004017582922533109,
      "loss": 5.2749,
      "mean_token_accuracy": 0.19953782856464386,
      "num_tokens": 73165803.0,
      "step": 36105
    },
    {
      "entropy": 6.056471872329712,
      "epoch": 3.1556410032334177,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040173254559541853,
      "loss": 5.2714,
      "mean_token_accuracy": 0.2106264889240265,
      "num_tokens": 73176029.0,
      "step": 36110
    },
    {
      "entropy": 6.017730474472046,
      "epoch": 3.1560779515861226,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004017067965067296,
      "loss": 5.2464,
      "mean_token_accuracy": 0.21143615245819092,
      "num_tokens": 73187708.0,
      "step": 36115
    },
    {
      "entropy": 6.060139226913452,
      "epoch": 3.156514899938827,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004016810449877379,
      "loss": 5.24,
      "mean_token_accuracy": 0.20628948658704757,
      "num_tokens": 73197458.0,
      "step": 36120
    },
    {
      "entropy": 6.025058889389038,
      "epoch": 3.156951848291532,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004016552910389373,
      "loss": 5.2205,
      "mean_token_accuracy": 0.20549193024635315,
      "num_tokens": 73206931.0,
      "step": 36125
    },
    {
      "entropy": 5.994244766235352,
      "epoch": 3.157388796644237,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040162953466082194,
      "loss": 5.2282,
      "mean_token_accuracy": 0.20890741795301437,
      "num_tokens": 73217048.0,
      "step": 36130
    },
    {
      "entropy": 6.010069179534912,
      "epoch": 3.1578257449969414,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004016037758538857,
      "loss": 5.2013,
      "mean_token_accuracy": 0.20899028182029725,
      "num_tokens": 73226649.0,
      "step": 36135
    },
    {
      "entropy": 6.003392124176026,
      "epoch": 3.1582626933496463,
      "grad_norm": 1.25,
      "learning_rate": 0.0004015780146186227,
      "loss": 5.1521,
      "mean_token_accuracy": 0.21440722197294235,
      "num_tokens": 73236822.0,
      "step": 36140
    },
    {
      "entropy": 6.079858922958374,
      "epoch": 3.1586996417023507,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004015522509555271,
      "loss": 5.3264,
      "mean_token_accuracy": 0.20484449714422226,
      "num_tokens": 73247078.0,
      "step": 36145
    },
    {
      "entropy": 5.945835018157959,
      "epoch": 3.1591365900550556,
      "grad_norm": 1.1875,
      "learning_rate": 0.000401526484865093,
      "loss": 5.0723,
      "mean_token_accuracy": 0.22305436581373214,
      "num_tokens": 73257001.0,
      "step": 36150
    },
    {
      "entropy": 5.9633337497711185,
      "epoch": 3.15957353840776,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004015007163478146,
      "loss": 5.1526,
      "mean_token_accuracy": 0.21569494754076005,
      "num_tokens": 73266922.0,
      "step": 36155
    },
    {
      "entropy": 5.982828664779663,
      "epoch": 3.160010486760465,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040147494540418617,
      "loss": 5.2097,
      "mean_token_accuracy": 0.20693878084421158,
      "num_tokens": 73277858.0,
      "step": 36160
    },
    {
      "entropy": 5.989380264282227,
      "epoch": 3.1604474351131695,
      "grad_norm": 1.25,
      "learning_rate": 0.00040144917203470203,
      "loss": 5.2331,
      "mean_token_accuracy": 0.2086735561490059,
      "num_tokens": 73287361.0,
      "step": 36165
    },
    {
      "entropy": 6.008219051361084,
      "epoch": 3.1608843834658744,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004014233962398565,
      "loss": 5.1695,
      "mean_token_accuracy": 0.21115922182798386,
      "num_tokens": 73296521.0,
      "step": 36170
    },
    {
      "entropy": 5.978036880493164,
      "epoch": 3.161321331818579,
      "grad_norm": 1.25,
      "learning_rate": 0.000401397618020144,
      "loss": 5.1848,
      "mean_token_accuracy": 0.20822983384132385,
      "num_tokens": 73306257.0,
      "step": 36175
    },
    {
      "entropy": 5.951703214645386,
      "epoch": 3.161758280171284,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000401371837376059,
      "loss": 5.1395,
      "mean_token_accuracy": 0.21668287813663484,
      "num_tokens": 73315939.0,
      "step": 36180
    },
    {
      "entropy": 5.99874062538147,
      "epoch": 3.1621952285239883,
      "grad_norm": 1.125,
      "learning_rate": 0.0004013460543080959,
      "loss": 5.2012,
      "mean_token_accuracy": 0.21362730264663696,
      "num_tokens": 73326407.0,
      "step": 36185
    },
    {
      "entropy": 6.038590621948242,
      "epoch": 3.162632176876693,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004013202688167493,
      "loss": 5.2322,
      "mean_token_accuracy": 0.2085847109556198,
      "num_tokens": 73336188.0,
      "step": 36190
    },
    {
      "entropy": 6.041966962814331,
      "epoch": 3.163069125229398,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004012944809025136,
      "loss": 5.235,
      "mean_token_accuracy": 0.21073623299598693,
      "num_tokens": 73346824.0,
      "step": 36195
    },
    {
      "entropy": 6.013591003417969,
      "epoch": 3.1635060735821026,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040126869056588375,
      "loss": 5.2101,
      "mean_token_accuracy": 0.20540666580200195,
      "num_tokens": 73356688.0,
      "step": 36200
    },
    {
      "entropy": 6.054212331771851,
      "epoch": 3.1639430219348075,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004012428978073543,
      "loss": 5.3145,
      "mean_token_accuracy": 0.20424061566591262,
      "num_tokens": 73366905.0,
      "step": 36205
    },
    {
      "entropy": 6.082866239547729,
      "epoch": 3.164379970287512,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040121710262741974,
      "loss": 5.3047,
      "mean_token_accuracy": 0.19962154030799867,
      "num_tokens": 73376870.0,
      "step": 36210
    },
    {
      "entropy": 5.998090839385986,
      "epoch": 3.164816918640217,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004011913050265751,
      "loss": 5.1655,
      "mean_token_accuracy": 0.21200965195894242,
      "num_tokens": 73386153.0,
      "step": 36215
    },
    {
      "entropy": 5.996933031082153,
      "epoch": 3.1652538669929213,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040116550500531506,
      "loss": 5.2676,
      "mean_token_accuracy": 0.20871747881174088,
      "num_tokens": 73395598.0,
      "step": 36220
    },
    {
      "entropy": 5.9988627433776855,
      "epoch": 3.1656908153456262,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040113970256413465,
      "loss": 5.1886,
      "mean_token_accuracy": 0.21102489531040192,
      "num_tokens": 73406883.0,
      "step": 36225
    },
    {
      "entropy": 6.010448980331421,
      "epoch": 3.1661277636983307,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040111389770352846,
      "loss": 5.2277,
      "mean_token_accuracy": 0.2145373448729515,
      "num_tokens": 73418504.0,
      "step": 36230
    },
    {
      "entropy": 6.017597293853759,
      "epoch": 3.1665647120510356,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040108809042399183,
      "loss": 5.2231,
      "mean_token_accuracy": 0.21049027442932128,
      "num_tokens": 73429228.0,
      "step": 36235
    },
    {
      "entropy": 5.98540849685669,
      "epoch": 3.16700166040374,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004010622807260194,
      "loss": 5.1925,
      "mean_token_accuracy": 0.21120257377624513,
      "num_tokens": 73439269.0,
      "step": 36240
    },
    {
      "entropy": 6.016873216629028,
      "epoch": 3.167438608756445,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004010364686101064,
      "loss": 5.2325,
      "mean_token_accuracy": 0.2072332724928856,
      "num_tokens": 73449554.0,
      "step": 36245
    },
    {
      "entropy": 5.954187393188477,
      "epoch": 3.1678755571091495,
      "grad_norm": 1.265625,
      "learning_rate": 0.000401010654076748,
      "loss": 5.1901,
      "mean_token_accuracy": 0.22083909958600997,
      "num_tokens": 73460453.0,
      "step": 36250
    },
    {
      "entropy": 5.989920902252197,
      "epoch": 3.1683125054618544,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004009848371264391,
      "loss": 5.132,
      "mean_token_accuracy": 0.21628358215093613,
      "num_tokens": 73469905.0,
      "step": 36255
    },
    {
      "entropy": 5.985828685760498,
      "epoch": 3.1687494538145593,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040095901775967507,
      "loss": 5.3024,
      "mean_token_accuracy": 0.20673198699951173,
      "num_tokens": 73480363.0,
      "step": 36260
    },
    {
      "entropy": 5.983976602554321,
      "epoch": 3.1691864021672638,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040093319597695105,
      "loss": 5.2407,
      "mean_token_accuracy": 0.21114132851362227,
      "num_tokens": 73490080.0,
      "step": 36265
    },
    {
      "entropy": 6.125428628921509,
      "epoch": 3.1696233505199687,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004009073717787624,
      "loss": 5.2775,
      "mean_token_accuracy": 0.20503613650798796,
      "num_tokens": 73500464.0,
      "step": 36270
    },
    {
      "entropy": 6.026804876327515,
      "epoch": 3.170060298872673,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004008815451656043,
      "loss": 5.2236,
      "mean_token_accuracy": 0.2125499963760376,
      "num_tokens": 73509834.0,
      "step": 36275
    },
    {
      "entropy": 5.996546936035156,
      "epoch": 3.170497247225378,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040085571613797226,
      "loss": 5.3227,
      "mean_token_accuracy": 0.20069922506809235,
      "num_tokens": 73520345.0,
      "step": 36280
    },
    {
      "entropy": 6.021788692474365,
      "epoch": 3.1709341955780825,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004008298846963617,
      "loss": 5.2939,
      "mean_token_accuracy": 0.20736991614103317,
      "num_tokens": 73530769.0,
      "step": 36285
    },
    {
      "entropy": 6.086397266387939,
      "epoch": 3.1713711439307875,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004008040508412678,
      "loss": 5.2067,
      "mean_token_accuracy": 0.20459921956062316,
      "num_tokens": 73539989.0,
      "step": 36290
    },
    {
      "entropy": 6.05887804031372,
      "epoch": 3.171808092283492,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004007782145731865,
      "loss": 5.2504,
      "mean_token_accuracy": 0.20105236619710923,
      "num_tokens": 73551344.0,
      "step": 36295
    },
    {
      "entropy": 6.00713963508606,
      "epoch": 3.172245040636197,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040075237589261303,
      "loss": 5.1832,
      "mean_token_accuracy": 0.2129138857126236,
      "num_tokens": 73560580.0,
      "step": 36300
    },
    {
      "entropy": 6.012425518035888,
      "epoch": 3.1726819889889013,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004007265348000432,
      "loss": 5.2061,
      "mean_token_accuracy": 0.2090657353401184,
      "num_tokens": 73569500.0,
      "step": 36305
    },
    {
      "entropy": 6.103154850006104,
      "epoch": 3.173118937341606,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040070069129597243,
      "loss": 5.2938,
      "mean_token_accuracy": 0.20265535712242128,
      "num_tokens": 73578658.0,
      "step": 36310
    },
    {
      "entropy": 6.0304208278656,
      "epoch": 3.173555885694311,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040067484538089653,
      "loss": 5.1943,
      "mean_token_accuracy": 0.21130180060863496,
      "num_tokens": 73588842.0,
      "step": 36315
    },
    {
      "entropy": 6.050020933151245,
      "epoch": 3.1739928340470156,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004006489970553114,
      "loss": 5.263,
      "mean_token_accuracy": 0.20930064022541045,
      "num_tokens": 73598962.0,
      "step": 36320
    },
    {
      "entropy": 6.029398441314697,
      "epoch": 3.1744297823997205,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040062314631971255,
      "loss": 5.2923,
      "mean_token_accuracy": 0.20233575850725175,
      "num_tokens": 73609890.0,
      "step": 36325
    },
    {
      "entropy": 6.003971576690674,
      "epoch": 3.174866730752425,
      "grad_norm": 1.25,
      "learning_rate": 0.0004005972931745959,
      "loss": 5.1914,
      "mean_token_accuracy": 0.21138090789318084,
      "num_tokens": 73620453.0,
      "step": 36330
    },
    {
      "entropy": 6.045704698562622,
      "epoch": 3.17530367910513,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040057143762045746,
      "loss": 5.229,
      "mean_token_accuracy": 0.20869613140821458,
      "num_tokens": 73630385.0,
      "step": 36335
    },
    {
      "entropy": 6.048480415344239,
      "epoch": 3.1757406274578344,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040054557965779294,
      "loss": 5.2471,
      "mean_token_accuracy": 0.2025328606367111,
      "num_tokens": 73640081.0,
      "step": 36340
    },
    {
      "entropy": 6.014483118057251,
      "epoch": 3.1761775758105393,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040051971928709853,
      "loss": 5.2302,
      "mean_token_accuracy": 0.20629858374595642,
      "num_tokens": 73650934.0,
      "step": 36345
    },
    {
      "entropy": 5.975817728042602,
      "epoch": 3.1766145241632437,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040049385650887004,
      "loss": 5.1889,
      "mean_token_accuracy": 0.21139660030603408,
      "num_tokens": 73661835.0,
      "step": 36350
    },
    {
      "entropy": 6.082211542129516,
      "epoch": 3.1770514725159487,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004004679913236036,
      "loss": 5.2724,
      "mean_token_accuracy": 0.20645512044429778,
      "num_tokens": 73671738.0,
      "step": 36355
    },
    {
      "entropy": 6.055068302154541,
      "epoch": 3.177488420868653,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040044212373179547,
      "loss": 5.2879,
      "mean_token_accuracy": 0.1978387102484703,
      "num_tokens": 73681834.0,
      "step": 36360
    },
    {
      "entropy": 5.9891037940979,
      "epoch": 3.177925369221358,
      "grad_norm": 1.25,
      "learning_rate": 0.00040041625373394153,
      "loss": 5.2224,
      "mean_token_accuracy": 0.20601088255643846,
      "num_tokens": 73692137.0,
      "step": 36365
    },
    {
      "entropy": 6.011704158782959,
      "epoch": 3.178362317574063,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004003903813305383,
      "loss": 5.2738,
      "mean_token_accuracy": 0.20722942501306535,
      "num_tokens": 73702263.0,
      "step": 36370
    },
    {
      "entropy": 6.020955562591553,
      "epoch": 3.1787992659267674,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004003645065220817,
      "loss": 5.2885,
      "mean_token_accuracy": 0.20056906789541246,
      "num_tokens": 73711945.0,
      "step": 36375
    },
    {
      "entropy": 6.136260652542115,
      "epoch": 3.1792362142794723,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004003386293090682,
      "loss": 5.3022,
      "mean_token_accuracy": 0.203248493373394,
      "num_tokens": 73722205.0,
      "step": 36380
    },
    {
      "entropy": 6.084401178359985,
      "epoch": 3.179673162632177,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004003127496919942,
      "loss": 5.2784,
      "mean_token_accuracy": 0.2041214868426323,
      "num_tokens": 73732049.0,
      "step": 36385
    },
    {
      "entropy": 5.987223768234253,
      "epoch": 3.1801101109848817,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004002868676713559,
      "loss": 5.2748,
      "mean_token_accuracy": 0.20778426229953767,
      "num_tokens": 73741559.0,
      "step": 36390
    },
    {
      "entropy": 6.064154148101807,
      "epoch": 3.180547059337586,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040026098324764985,
      "loss": 5.3538,
      "mean_token_accuracy": 0.201216422021389,
      "num_tokens": 73751576.0,
      "step": 36395
    },
    {
      "entropy": 6.066591691970825,
      "epoch": 3.180984007690291,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004002350964213725,
      "loss": 5.3187,
      "mean_token_accuracy": 0.2011834517121315,
      "num_tokens": 73762155.0,
      "step": 36400
    },
    {
      "entropy": 6.070414209365845,
      "epoch": 3.1814209560429956,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004002092071930204,
      "loss": 5.2234,
      "mean_token_accuracy": 0.2075077101588249,
      "num_tokens": 73772427.0,
      "step": 36405
    },
    {
      "entropy": 6.012972593307495,
      "epoch": 3.1818579043957005,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004001833155630901,
      "loss": 5.1911,
      "mean_token_accuracy": 0.21417371332645416,
      "num_tokens": 73782868.0,
      "step": 36410
    },
    {
      "entropy": 5.978199434280396,
      "epoch": 3.182294852748405,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040015742153207823,
      "loss": 5.2426,
      "mean_token_accuracy": 0.21116983145475388,
      "num_tokens": 73792480.0,
      "step": 36415
    },
    {
      "entropy": 6.013310098648072,
      "epoch": 3.18273180110111,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004001315251004814,
      "loss": 5.2431,
      "mean_token_accuracy": 0.2066675066947937,
      "num_tokens": 73803264.0,
      "step": 36420
    },
    {
      "entropy": 6.023541498184204,
      "epoch": 3.183168749453815,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040010562626879636,
      "loss": 5.1276,
      "mean_token_accuracy": 0.21885699927806854,
      "num_tokens": 73812491.0,
      "step": 36425
    },
    {
      "entropy": 6.029627895355224,
      "epoch": 3.1836056978065193,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040007972503751984,
      "loss": 5.2197,
      "mean_token_accuracy": 0.20767299830913544,
      "num_tokens": 73821672.0,
      "step": 36430
    },
    {
      "entropy": 6.017544555664062,
      "epoch": 3.184042646159224,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040005382140714867,
      "loss": 5.1924,
      "mean_token_accuracy": 0.21226154267787933,
      "num_tokens": 73831709.0,
      "step": 36435
    },
    {
      "entropy": 5.9809972763061525,
      "epoch": 3.1844795945119286,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004000279153781797,
      "loss": 5.2362,
      "mean_token_accuracy": 0.2022416189312935,
      "num_tokens": 73841121.0,
      "step": 36440
    },
    {
      "entropy": 6.0462888240814205,
      "epoch": 3.1849165428646335,
      "grad_norm": 1.375,
      "learning_rate": 0.00040000200695110977,
      "loss": 5.2685,
      "mean_token_accuracy": 0.20554581731557847,
      "num_tokens": 73851299.0,
      "step": 36445
    },
    {
      "entropy": 6.1236591815948485,
      "epoch": 3.185353491217338,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003999760961264359,
      "loss": 5.2939,
      "mean_token_accuracy": 0.20533522069454194,
      "num_tokens": 73862020.0,
      "step": 36450
    },
    {
      "entropy": 6.0688619136810305,
      "epoch": 3.185790439570043,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039995018290465495,
      "loss": 5.2558,
      "mean_token_accuracy": 0.2063014402985573,
      "num_tokens": 73872786.0,
      "step": 36455
    },
    {
      "entropy": 5.9863025665283205,
      "epoch": 3.1862273879227474,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003999242672862641,
      "loss": 5.1736,
      "mean_token_accuracy": 0.2113913968205452,
      "num_tokens": 73882303.0,
      "step": 36460
    },
    {
      "entropy": 5.995978116989136,
      "epoch": 3.1866643362754523,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003998983492717602,
      "loss": 5.2147,
      "mean_token_accuracy": 0.21492076218128203,
      "num_tokens": 73892124.0,
      "step": 36465
    },
    {
      "entropy": 6.009947395324707,
      "epoch": 3.187101284628157,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003998724288616406,
      "loss": 5.1748,
      "mean_token_accuracy": 0.21938615143299103,
      "num_tokens": 73901803.0,
      "step": 36470
    },
    {
      "entropy": 6.014356899261474,
      "epoch": 3.1875382329808617,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039984650605640244,
      "loss": 5.1377,
      "mean_token_accuracy": 0.22109656184911727,
      "num_tokens": 73911767.0,
      "step": 36475
    },
    {
      "entropy": 6.073317813873291,
      "epoch": 3.187975181333566,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003998205808565428,
      "loss": 5.3045,
      "mean_token_accuracy": 0.20174396634101868,
      "num_tokens": 73921746.0,
      "step": 36480
    },
    {
      "entropy": 5.963896131515503,
      "epoch": 3.188412129686271,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039979465326255904,
      "loss": 5.1975,
      "mean_token_accuracy": 0.21256877928972245,
      "num_tokens": 73932282.0,
      "step": 36485
    },
    {
      "entropy": 6.04523983001709,
      "epoch": 3.188849078038976,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003997687232749484,
      "loss": 5.2805,
      "mean_token_accuracy": 0.2068416267633438,
      "num_tokens": 73942739.0,
      "step": 36490
    },
    {
      "entropy": 6.021734523773193,
      "epoch": 3.1892860263916805,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039974279089420833,
      "loss": 5.242,
      "mean_token_accuracy": 0.21120132207870485,
      "num_tokens": 73953739.0,
      "step": 36495
    },
    {
      "entropy": 6.050042247772216,
      "epoch": 3.1897229747443854,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00039971685612083616,
      "loss": 5.243,
      "mean_token_accuracy": 0.20837754160165786,
      "num_tokens": 73963808.0,
      "step": 36500
    },
    {
      "entropy": 6.135912322998047,
      "epoch": 3.19015992309709,
      "grad_norm": 1.34375,
      "learning_rate": 0.00039969091895532933,
      "loss": 5.3313,
      "mean_token_accuracy": 0.2000255063176155,
      "num_tokens": 73973083.0,
      "step": 36505
    },
    {
      "entropy": 5.953465175628662,
      "epoch": 3.1905968714497948,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003996649793981853,
      "loss": 5.1259,
      "mean_token_accuracy": 0.22405130416154861,
      "num_tokens": 73983966.0,
      "step": 36510
    },
    {
      "entropy": 6.057453441619873,
      "epoch": 3.1910338198024992,
      "grad_norm": 1.25,
      "learning_rate": 0.00039963903744990166,
      "loss": 5.2854,
      "mean_token_accuracy": 0.2066226378083229,
      "num_tokens": 73993874.0,
      "step": 36515
    },
    {
      "entropy": 6.007268381118775,
      "epoch": 3.191470768155204,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000399613093110976,
      "loss": 5.2699,
      "mean_token_accuracy": 0.2003816083073616,
      "num_tokens": 74004179.0,
      "step": 36520
    },
    {
      "entropy": 5.998674201965332,
      "epoch": 3.1919077165079086,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003995871463819059,
      "loss": 5.2426,
      "mean_token_accuracy": 0.2086450219154358,
      "num_tokens": 74014941.0,
      "step": 36525
    },
    {
      "entropy": 6.052500820159912,
      "epoch": 3.1923446648606135,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039956119726318907,
      "loss": 5.3228,
      "mean_token_accuracy": 0.19496934562921525,
      "num_tokens": 74026620.0,
      "step": 36530
    },
    {
      "entropy": 6.041390228271484,
      "epoch": 3.192781613213318,
      "grad_norm": 1.515625,
      "learning_rate": 0.00039953524575532317,
      "loss": 5.269,
      "mean_token_accuracy": 0.20255627632141113,
      "num_tokens": 74036341.0,
      "step": 36535
    },
    {
      "entropy": 6.0642174243927,
      "epoch": 3.193218561566023,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039950929185880604,
      "loss": 5.2216,
      "mean_token_accuracy": 0.20995814651250838,
      "num_tokens": 74046884.0,
      "step": 36540
    },
    {
      "entropy": 5.9744078636169435,
      "epoch": 3.1936555099187274,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003994833355741354,
      "loss": 5.1821,
      "mean_token_accuracy": 0.21019198298454284,
      "num_tokens": 74056706.0,
      "step": 36545
    },
    {
      "entropy": 5.8437378883361815,
      "epoch": 3.1940924582714323,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039945737690180917,
      "loss": 5.1023,
      "mean_token_accuracy": 0.2309470370411873,
      "num_tokens": 74067046.0,
      "step": 36550
    },
    {
      "entropy": 6.047464609146118,
      "epoch": 3.194529406624137,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003994314158423252,
      "loss": 5.1947,
      "mean_token_accuracy": 0.20993087142705918,
      "num_tokens": 74076884.0,
      "step": 36555
    },
    {
      "entropy": 6.032649135589599,
      "epoch": 3.1949663549768417,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003994054523961816,
      "loss": 5.2692,
      "mean_token_accuracy": 0.20278876572847365,
      "num_tokens": 74086482.0,
      "step": 36560
    },
    {
      "entropy": 5.929864406585693,
      "epoch": 3.1954033033295466,
      "grad_norm": 1.25,
      "learning_rate": 0.00039937948656387616,
      "loss": 5.1673,
      "mean_token_accuracy": 0.21546151787042617,
      "num_tokens": 74095916.0,
      "step": 36565
    },
    {
      "entropy": 6.032597446441651,
      "epoch": 3.195840251682251,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039935351834590696,
      "loss": 5.3057,
      "mean_token_accuracy": 0.20612097084522246,
      "num_tokens": 74106054.0,
      "step": 36570
    },
    {
      "entropy": 6.005875301361084,
      "epoch": 3.196277200034956,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039932754774277214,
      "loss": 5.2308,
      "mean_token_accuracy": 0.207464762032032,
      "num_tokens": 74115641.0,
      "step": 36575
    },
    {
      "entropy": 6.029762077331543,
      "epoch": 3.1967141483876604,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003993015747549698,
      "loss": 5.2208,
      "mean_token_accuracy": 0.20772870928049086,
      "num_tokens": 74125370.0,
      "step": 36580
    },
    {
      "entropy": 6.048151159286499,
      "epoch": 3.1971510967403654,
      "grad_norm": 1.359375,
      "learning_rate": 0.00039927559938299814,
      "loss": 5.2196,
      "mean_token_accuracy": 0.20973674803972245,
      "num_tokens": 74134959.0,
      "step": 36585
    },
    {
      "entropy": 6.069590950012207,
      "epoch": 3.19758804509307,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003992496216273554,
      "loss": 5.3455,
      "mean_token_accuracy": 0.19803361594676971,
      "num_tokens": 74144939.0,
      "step": 36590
    },
    {
      "entropy": 6.006773614883423,
      "epoch": 3.1980249934457747,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039922364148853973,
      "loss": 5.1253,
      "mean_token_accuracy": 0.21552238762378692,
      "num_tokens": 74154336.0,
      "step": 36595
    },
    {
      "entropy": 6.014067077636719,
      "epoch": 3.198461941798479,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003991976589670496,
      "loss": 5.2388,
      "mean_token_accuracy": 0.20654570162296296,
      "num_tokens": 74164022.0,
      "step": 36600
    },
    {
      "entropy": 6.111302757263184,
      "epoch": 3.198898890151184,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003991716740633832,
      "loss": 5.2449,
      "mean_token_accuracy": 0.20574762523174286,
      "num_tokens": 74174186.0,
      "step": 36605
    },
    {
      "entropy": 6.018744707107544,
      "epoch": 3.199335838503889,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003991456867780391,
      "loss": 5.2349,
      "mean_token_accuracy": 0.21073849201202394,
      "num_tokens": 74183913.0,
      "step": 36610
    },
    {
      "entropy": 6.041465139389038,
      "epoch": 3.1997727868565935,
      "grad_norm": 1.125,
      "learning_rate": 0.0003991196971115157,
      "loss": 5.2992,
      "mean_token_accuracy": 0.20591090321540834,
      "num_tokens": 74194307.0,
      "step": 36615
    },
    {
      "entropy": 6.0634849071502686,
      "epoch": 3.2002097352092984,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039909370506431146,
      "loss": 5.2494,
      "mean_token_accuracy": 0.2086953803896904,
      "num_tokens": 74204672.0,
      "step": 36620
    },
    {
      "entropy": 6.059436798095703,
      "epoch": 3.200646683562003,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003990677106369249,
      "loss": 5.233,
      "mean_token_accuracy": 0.21008283644914627,
      "num_tokens": 74214022.0,
      "step": 36625
    },
    {
      "entropy": 5.931782341003418,
      "epoch": 3.201083631914708,
      "grad_norm": 1.328125,
      "learning_rate": 0.00039904171382985465,
      "loss": 5.1852,
      "mean_token_accuracy": 0.20840624421834947,
      "num_tokens": 74223736.0,
      "step": 36630
    },
    {
      "entropy": 5.996647310256958,
      "epoch": 3.2015205802674123,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003990157146435993,
      "loss": 5.2046,
      "mean_token_accuracy": 0.21195336878299714,
      "num_tokens": 74234742.0,
      "step": 36635
    },
    {
      "entropy": 6.063381099700928,
      "epoch": 3.201957528620117,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003989897130786577,
      "loss": 5.192,
      "mean_token_accuracy": 0.213476300239563,
      "num_tokens": 74244949.0,
      "step": 36640
    },
    {
      "entropy": 6.076003313064575,
      "epoch": 3.2023944769728216,
      "grad_norm": 1.34375,
      "learning_rate": 0.00039896370913552825,
      "loss": 5.2659,
      "mean_token_accuracy": 0.206112539768219,
      "num_tokens": 74255280.0,
      "step": 36645
    },
    {
      "entropy": 5.887169361114502,
      "epoch": 3.2028314253255266,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039893770281471005,
      "loss": 5.1382,
      "mean_token_accuracy": 0.2243207335472107,
      "num_tokens": 74265615.0,
      "step": 36650
    },
    {
      "entropy": 5.949886751174927,
      "epoch": 3.203268373678231,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003989116941167016,
      "loss": 5.2487,
      "mean_token_accuracy": 0.21404578685760497,
      "num_tokens": 74275791.0,
      "step": 36655
    },
    {
      "entropy": 6.047795200347901,
      "epoch": 3.203705322030936,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039888568304200206,
      "loss": 5.2502,
      "mean_token_accuracy": 0.2002468004822731,
      "num_tokens": 74286295.0,
      "step": 36660
    },
    {
      "entropy": 5.989293432235717,
      "epoch": 3.204142270383641,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039885966959111015,
      "loss": 5.1578,
      "mean_token_accuracy": 0.21534905582666397,
      "num_tokens": 74295163.0,
      "step": 36665
    },
    {
      "entropy": 6.025145864486694,
      "epoch": 3.2045792187363453,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003988336537645249,
      "loss": 5.2356,
      "mean_token_accuracy": 0.20813699662685395,
      "num_tokens": 74304855.0,
      "step": 36670
    },
    {
      "entropy": 6.038328552246094,
      "epoch": 3.2050161670890502,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003988076355627453,
      "loss": 5.2576,
      "mean_token_accuracy": 0.20569224357604982,
      "num_tokens": 74314584.0,
      "step": 36675
    },
    {
      "entropy": 6.013509845733642,
      "epoch": 3.2054531154417547,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003987816149862703,
      "loss": 5.2294,
      "mean_token_accuracy": 0.20763790160417556,
      "num_tokens": 74324992.0,
      "step": 36680
    },
    {
      "entropy": 6.006850671768189,
      "epoch": 3.2058900637944596,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039875559203559913,
      "loss": 5.1828,
      "mean_token_accuracy": 0.2112337291240692,
      "num_tokens": 74336180.0,
      "step": 36685
    },
    {
      "entropy": 5.98041615486145,
      "epoch": 3.206327012147164,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003987295667112309,
      "loss": 5.0995,
      "mean_token_accuracy": 0.22083953469991685,
      "num_tokens": 74346073.0,
      "step": 36690
    },
    {
      "entropy": 5.982775640487671,
      "epoch": 3.206763960499869,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003987035390136647,
      "loss": 5.1704,
      "mean_token_accuracy": 0.21245847046375274,
      "num_tokens": 74356470.0,
      "step": 36695
    },
    {
      "entropy": 5.9812459468841555,
      "epoch": 3.2072009088525735,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039867750894339974,
      "loss": 5.17,
      "mean_token_accuracy": 0.21819260120391845,
      "num_tokens": 74367035.0,
      "step": 36700
    },
    {
      "entropy": 5.982419824600219,
      "epoch": 3.2076378572052784,
      "grad_norm": 1.25,
      "learning_rate": 0.0003986514765009354,
      "loss": 5.2163,
      "mean_token_accuracy": 0.21330926716327667,
      "num_tokens": 74377264.0,
      "step": 36705
    },
    {
      "entropy": 6.068867492675781,
      "epoch": 3.208074805557983,
      "grad_norm": 1.125,
      "learning_rate": 0.00039862544168677103,
      "loss": 5.3284,
      "mean_token_accuracy": 0.1937178134918213,
      "num_tokens": 74389053.0,
      "step": 36710
    },
    {
      "entropy": 6.040244817733765,
      "epoch": 3.2085117539106878,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003985994045014058,
      "loss": 5.2238,
      "mean_token_accuracy": 0.2111205965280533,
      "num_tokens": 74399546.0,
      "step": 36715
    },
    {
      "entropy": 6.029888677597046,
      "epoch": 3.2089487022633927,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039857336494533924,
      "loss": 5.2305,
      "mean_token_accuracy": 0.2105369284749031,
      "num_tokens": 74409905.0,
      "step": 36720
    },
    {
      "entropy": 5.968280935287476,
      "epoch": 3.209385650616097,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003985473230190708,
      "loss": 5.2077,
      "mean_token_accuracy": 0.21042898148298264,
      "num_tokens": 74419893.0,
      "step": 36725
    },
    {
      "entropy": 5.965349245071411,
      "epoch": 3.209822598968802,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003985212787231,
      "loss": 5.1614,
      "mean_token_accuracy": 0.21463136970996857,
      "num_tokens": 74430286.0,
      "step": 36730
    },
    {
      "entropy": 5.982477140426636,
      "epoch": 3.2102595473215065,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003984952320579263,
      "loss": 5.1666,
      "mean_token_accuracy": 0.20927275270223616,
      "num_tokens": 74439601.0,
      "step": 36735
    },
    {
      "entropy": 6.088380479812622,
      "epoch": 3.2106964956742114,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003984691830240494,
      "loss": 5.2325,
      "mean_token_accuracy": 0.20644260346889495,
      "num_tokens": 74448763.0,
      "step": 36740
    },
    {
      "entropy": 6.130937385559082,
      "epoch": 3.211133444026916,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039844313162196876,
      "loss": 5.3373,
      "mean_token_accuracy": 0.198067544400692,
      "num_tokens": 74458357.0,
      "step": 36745
    },
    {
      "entropy": 6.038661909103394,
      "epoch": 3.211570392379621,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039841707785218426,
      "loss": 5.1622,
      "mean_token_accuracy": 0.21736419200897217,
      "num_tokens": 74468280.0,
      "step": 36750
    },
    {
      "entropy": 5.941810846328735,
      "epoch": 3.2120073407323253,
      "grad_norm": 1.25,
      "learning_rate": 0.00039839102171519556,
      "loss": 5.1255,
      "mean_token_accuracy": 0.21784884184598924,
      "num_tokens": 74478103.0,
      "step": 36755
    },
    {
      "entropy": 6.038300800323486,
      "epoch": 3.21244428908503,
      "grad_norm": 1.25,
      "learning_rate": 0.0003983649632115024,
      "loss": 5.1945,
      "mean_token_accuracy": 0.209469573199749,
      "num_tokens": 74487903.0,
      "step": 36760
    },
    {
      "entropy": 5.987263154983521,
      "epoch": 3.2128812374377347,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003983389023416045,
      "loss": 5.2251,
      "mean_token_accuracy": 0.21041646599769592,
      "num_tokens": 74497998.0,
      "step": 36765
    },
    {
      "entropy": 6.012473011016846,
      "epoch": 3.2133181857904396,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039831283910600197,
      "loss": 5.2099,
      "mean_token_accuracy": 0.20777094066143037,
      "num_tokens": 74507313.0,
      "step": 36770
    },
    {
      "entropy": 6.063389539718628,
      "epoch": 3.2137551341431445,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003982867735051945,
      "loss": 5.2228,
      "mean_token_accuracy": 0.2088652729988098,
      "num_tokens": 74516983.0,
      "step": 36775
    },
    {
      "entropy": 5.989509153366089,
      "epoch": 3.214192082495849,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003982607055396821,
      "loss": 5.1925,
      "mean_token_accuracy": 0.2124797970056534,
      "num_tokens": 74527061.0,
      "step": 36780
    },
    {
      "entropy": 5.955058145523071,
      "epoch": 3.214629030848554,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039823463520996484,
      "loss": 5.2473,
      "mean_token_accuracy": 0.21088431626558304,
      "num_tokens": 74537782.0,
      "step": 36785
    },
    {
      "entropy": 6.045482158660889,
      "epoch": 3.2150659792012584,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003982085625165426,
      "loss": 5.23,
      "mean_token_accuracy": 0.21152880787849426,
      "num_tokens": 74548082.0,
      "step": 36790
    },
    {
      "entropy": 6.034719467163086,
      "epoch": 3.2155029275539633,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003981824874599157,
      "loss": 5.2207,
      "mean_token_accuracy": 0.2056033954024315,
      "num_tokens": 74559286.0,
      "step": 36795
    },
    {
      "entropy": 6.018435049057007,
      "epoch": 3.2159398759066677,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000398156410040584,
      "loss": 5.2494,
      "mean_token_accuracy": 0.20590716004371643,
      "num_tokens": 74569306.0,
      "step": 36800
    },
    {
      "entropy": 6.056318283081055,
      "epoch": 3.2163768242593727,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039813033025904793,
      "loss": 5.3072,
      "mean_token_accuracy": 0.19860024452209474,
      "num_tokens": 74579798.0,
      "step": 36805
    },
    {
      "entropy": 6.015184831619263,
      "epoch": 3.216813772612077,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039810424811580756,
      "loss": 5.239,
      "mean_token_accuracy": 0.20982321202754975,
      "num_tokens": 74590068.0,
      "step": 36810
    },
    {
      "entropy": 6.046512269973755,
      "epoch": 3.217250720964782,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039807816361136323,
      "loss": 5.301,
      "mean_token_accuracy": 0.19690708369016646,
      "num_tokens": 74600341.0,
      "step": 36815
    },
    {
      "entropy": 5.984365177154541,
      "epoch": 3.2176876693174865,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003980520767462152,
      "loss": 5.2682,
      "mean_token_accuracy": 0.20565005242824555,
      "num_tokens": 74609548.0,
      "step": 36820
    },
    {
      "entropy": 6.044133710861206,
      "epoch": 3.2181246176701914,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003980259875208639,
      "loss": 5.1658,
      "mean_token_accuracy": 0.21463459730148315,
      "num_tokens": 74619590.0,
      "step": 36825
    },
    {
      "entropy": 6.0668113231658936,
      "epoch": 3.218561566022896,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003979998959358097,
      "loss": 5.2077,
      "mean_token_accuracy": 0.2101852536201477,
      "num_tokens": 74629664.0,
      "step": 36830
    },
    {
      "entropy": 6.0264654636383055,
      "epoch": 3.218998514375601,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003979738019915529,
      "loss": 5.2238,
      "mean_token_accuracy": 0.21695445328950883,
      "num_tokens": 74639114.0,
      "step": 36835
    },
    {
      "entropy": 5.954277420043946,
      "epoch": 3.2194354627283057,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003979477056885943,
      "loss": 5.1578,
      "mean_token_accuracy": 0.21688652634620667,
      "num_tokens": 74648730.0,
      "step": 36840
    },
    {
      "entropy": 6.0479512214660645,
      "epoch": 3.21987241108101,
      "grad_norm": 1.359375,
      "learning_rate": 0.00039792160702743426,
      "loss": 5.261,
      "mean_token_accuracy": 0.2062561556696892,
      "num_tokens": 74659102.0,
      "step": 36845
    },
    {
      "entropy": 6.071783113479614,
      "epoch": 3.220309359433715,
      "grad_norm": 1.25,
      "learning_rate": 0.00039789550600857336,
      "loss": 5.3653,
      "mean_token_accuracy": 0.1944063276052475,
      "num_tokens": 74669344.0,
      "step": 36850
    },
    {
      "entropy": 6.091378545761108,
      "epoch": 3.2207463077864196,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003978694026325123,
      "loss": 5.2319,
      "mean_token_accuracy": 0.20756894946098328,
      "num_tokens": 74678524.0,
      "step": 36855
    },
    {
      "entropy": 6.0188392162322994,
      "epoch": 3.2211832561391245,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003978432968997517,
      "loss": 5.2568,
      "mean_token_accuracy": 0.20997204631567,
      "num_tokens": 74688584.0,
      "step": 36860
    },
    {
      "entropy": 6.04990701675415,
      "epoch": 3.221620204491829,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039781718881079227,
      "loss": 5.3175,
      "mean_token_accuracy": 0.19820194095373153,
      "num_tokens": 74699233.0,
      "step": 36865
    },
    {
      "entropy": 6.09306001663208,
      "epoch": 3.222057152844534,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039779107836613475,
      "loss": 5.2571,
      "mean_token_accuracy": 0.20726380348205567,
      "num_tokens": 74710144.0,
      "step": 36870
    },
    {
      "entropy": 6.049701166152954,
      "epoch": 3.2224941011972383,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039776496556628014,
      "loss": 5.3017,
      "mean_token_accuracy": 0.19864337295293807,
      "num_tokens": 74720289.0,
      "step": 36875
    },
    {
      "entropy": 6.003176355361939,
      "epoch": 3.2229310495499432,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039773885041172907,
      "loss": 5.2806,
      "mean_token_accuracy": 0.20129211544990538,
      "num_tokens": 74732395.0,
      "step": 36880
    },
    {
      "entropy": 6.034460496902466,
      "epoch": 3.2233679979026477,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039771273290298254,
      "loss": 5.2072,
      "mean_token_accuracy": 0.21233828961849213,
      "num_tokens": 74741986.0,
      "step": 36885
    },
    {
      "entropy": 6.024227857589722,
      "epoch": 3.2238049462553526,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003976866130405415,
      "loss": 5.1639,
      "mean_token_accuracy": 0.20549917221069336,
      "num_tokens": 74751540.0,
      "step": 36890
    },
    {
      "entropy": 6.00803508758545,
      "epoch": 3.224241894608057,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039766049082490696,
      "loss": 5.2253,
      "mean_token_accuracy": 0.20300124138593673,
      "num_tokens": 74761875.0,
      "step": 36895
    },
    {
      "entropy": 5.987338876724243,
      "epoch": 3.224678842960762,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039763436625657987,
      "loss": 5.1794,
      "mean_token_accuracy": 0.21049831956624984,
      "num_tokens": 74771273.0,
      "step": 36900
    },
    {
      "entropy": 6.008587598800659,
      "epoch": 3.225115791313467,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003976082393360614,
      "loss": 5.1692,
      "mean_token_accuracy": 0.21303993016481398,
      "num_tokens": 74781176.0,
      "step": 36905
    },
    {
      "entropy": 6.052298641204834,
      "epoch": 3.2255527396661714,
      "grad_norm": 1.34375,
      "learning_rate": 0.00039758211006385273,
      "loss": 5.2467,
      "mean_token_accuracy": 0.20457683354616166,
      "num_tokens": 74790731.0,
      "step": 36910
    },
    {
      "entropy": 6.0165363311767575,
      "epoch": 3.2259896880188763,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00039755597844045486,
      "loss": 5.1691,
      "mean_token_accuracy": 0.20834168493747712,
      "num_tokens": 74799810.0,
      "step": 36915
    },
    {
      "entropy": 6.010300970077514,
      "epoch": 3.226426636371581,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003975298444663692,
      "loss": 5.2635,
      "mean_token_accuracy": 0.20710409432649612,
      "num_tokens": 74809701.0,
      "step": 36920
    },
    {
      "entropy": 6.045156717300415,
      "epoch": 3.2268635847242857,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039750370814209683,
      "loss": 5.2026,
      "mean_token_accuracy": 0.2121707856655121,
      "num_tokens": 74820061.0,
      "step": 36925
    },
    {
      "entropy": 6.0711382865905765,
      "epoch": 3.22730053307699,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003974775694681392,
      "loss": 5.165,
      "mean_token_accuracy": 0.20950412452220918,
      "num_tokens": 74830360.0,
      "step": 36930
    },
    {
      "entropy": 6.008327150344849,
      "epoch": 3.227737481429695,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003974514284449975,
      "loss": 5.2471,
      "mean_token_accuracy": 0.20960066169500352,
      "num_tokens": 74841855.0,
      "step": 36935
    },
    {
      "entropy": 5.932042980194092,
      "epoch": 3.2281744297823995,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039742528507317337,
      "loss": 5.1868,
      "mean_token_accuracy": 0.2136224925518036,
      "num_tokens": 74851955.0,
      "step": 36940
    },
    {
      "entropy": 6.032527065277099,
      "epoch": 3.2286113781351045,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003973991393531681,
      "loss": 5.2078,
      "mean_token_accuracy": 0.2179451748728752,
      "num_tokens": 74861720.0,
      "step": 36945
    },
    {
      "entropy": 6.031464433670044,
      "epoch": 3.229048326487809,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039737299128548317,
      "loss": 5.2048,
      "mean_token_accuracy": 0.2069065734744072,
      "num_tokens": 74872136.0,
      "step": 36950
    },
    {
      "entropy": 6.061466073989868,
      "epoch": 3.229485274840514,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003973468408706202,
      "loss": 5.2319,
      "mean_token_accuracy": 0.207057186961174,
      "num_tokens": 74881683.0,
      "step": 36955
    },
    {
      "entropy": 6.021606683731079,
      "epoch": 3.2299222231932188,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039732068810908067,
      "loss": 5.1736,
      "mean_token_accuracy": 0.2125946044921875,
      "num_tokens": 74892175.0,
      "step": 36960
    },
    {
      "entropy": 5.959430360794068,
      "epoch": 3.2303591715459232,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003972945330013662,
      "loss": 5.2217,
      "mean_token_accuracy": 0.20935605317354203,
      "num_tokens": 74903547.0,
      "step": 36965
    },
    {
      "entropy": 5.995765924453735,
      "epoch": 3.230796119898628,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039726837554797856,
      "loss": 5.1899,
      "mean_token_accuracy": 0.21457479000091553,
      "num_tokens": 74912990.0,
      "step": 36970
    },
    {
      "entropy": 6.063954496383667,
      "epoch": 3.2312330682513326,
      "grad_norm": 1.34375,
      "learning_rate": 0.00039724221574941943,
      "loss": 5.2189,
      "mean_token_accuracy": 0.20499557256698608,
      "num_tokens": 74922306.0,
      "step": 36975
    },
    {
      "entropy": 6.059004926681519,
      "epoch": 3.2316700166040375,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039721605360619054,
      "loss": 5.2227,
      "mean_token_accuracy": 0.20576616674661635,
      "num_tokens": 74932207.0,
      "step": 36980
    },
    {
      "entropy": 6.0064441680908205,
      "epoch": 3.232106964956742,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039718988911879367,
      "loss": 5.2456,
      "mean_token_accuracy": 0.209969924390316,
      "num_tokens": 74941933.0,
      "step": 36985
    },
    {
      "entropy": 6.047195911407471,
      "epoch": 3.232543913309447,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039716372228773073,
      "loss": 5.2852,
      "mean_token_accuracy": 0.20698400884866713,
      "num_tokens": 74953418.0,
      "step": 36990
    },
    {
      "entropy": 6.092708778381348,
      "epoch": 3.2329808616621514,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039713755311350356,
      "loss": 5.2502,
      "mean_token_accuracy": 0.20272576212882995,
      "num_tokens": 74963381.0,
      "step": 36995
    },
    {
      "entropy": 6.0514177799224855,
      "epoch": 3.2334178100148563,
      "grad_norm": 1.4375,
      "learning_rate": 0.00039711138159661413,
      "loss": 5.2253,
      "mean_token_accuracy": 0.2177494764328003,
      "num_tokens": 74974072.0,
      "step": 37000
    },
    {
      "entropy": 5.933843421936035,
      "epoch": 3.2338547583675608,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003970852077375644,
      "loss": 5.1303,
      "mean_token_accuracy": 0.22289904803037644,
      "num_tokens": 74984191.0,
      "step": 37005
    },
    {
      "entropy": 5.982804775238037,
      "epoch": 3.2342917067202657,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003970590315368564,
      "loss": 5.2293,
      "mean_token_accuracy": 0.2101096048951149,
      "num_tokens": 74994278.0,
      "step": 37010
    },
    {
      "entropy": 5.980463981628418,
      "epoch": 3.2347286550729706,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003970328529949923,
      "loss": 5.1917,
      "mean_token_accuracy": 0.21883018016815187,
      "num_tokens": 75004150.0,
      "step": 37015
    },
    {
      "entropy": 6.027348327636719,
      "epoch": 3.235165603425675,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003970066721124741,
      "loss": 5.208,
      "mean_token_accuracy": 0.21165333241224288,
      "num_tokens": 75013632.0,
      "step": 37020
    },
    {
      "entropy": 6.035063457489014,
      "epoch": 3.23560255177838,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003969804888898039,
      "loss": 5.2702,
      "mean_token_accuracy": 0.19925092160701752,
      "num_tokens": 75023217.0,
      "step": 37025
    },
    {
      "entropy": 6.031669950485229,
      "epoch": 3.2360395001310844,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003969543033274841,
      "loss": 5.2228,
      "mean_token_accuracy": 0.20717982500791549,
      "num_tokens": 75033771.0,
      "step": 37030
    },
    {
      "entropy": 6.014716243743896,
      "epoch": 3.2364764484837893,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039692811542601684,
      "loss": 5.2672,
      "mean_token_accuracy": 0.2083122029900551,
      "num_tokens": 75044151.0,
      "step": 37035
    },
    {
      "entropy": 5.978878211975098,
      "epoch": 3.236913396836494,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003969019251859045,
      "loss": 5.1598,
      "mean_token_accuracy": 0.22130398601293563,
      "num_tokens": 75054129.0,
      "step": 37040
    },
    {
      "entropy": 6.032294654846192,
      "epoch": 3.2373503451891987,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039687573260764933,
      "loss": 5.2188,
      "mean_token_accuracy": 0.20532535314559935,
      "num_tokens": 75063147.0,
      "step": 37045
    },
    {
      "entropy": 5.977395343780517,
      "epoch": 3.237787293541903,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039684953769175376,
      "loss": 5.2649,
      "mean_token_accuracy": 0.20178984105587006,
      "num_tokens": 75072389.0,
      "step": 37050
    },
    {
      "entropy": 6.002229881286621,
      "epoch": 3.238224241894608,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003968233404387201,
      "loss": 5.2245,
      "mean_token_accuracy": 0.20928641110658647,
      "num_tokens": 75081253.0,
      "step": 37055
    },
    {
      "entropy": 5.995391702651977,
      "epoch": 3.2386611902473126,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000396797140849051,
      "loss": 5.1966,
      "mean_token_accuracy": 0.21543762385845183,
      "num_tokens": 75091316.0,
      "step": 37060
    },
    {
      "entropy": 6.1085912704467775,
      "epoch": 3.2390981386000175,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00039677093892324895,
      "loss": 5.3287,
      "mean_token_accuracy": 0.20245067328214644,
      "num_tokens": 75101204.0,
      "step": 37065
    },
    {
      "entropy": 6.045768117904663,
      "epoch": 3.2395350869527224,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003967447346618164,
      "loss": 5.2134,
      "mean_token_accuracy": 0.20371493697166443,
      "num_tokens": 75111524.0,
      "step": 37070
    },
    {
      "entropy": 5.965506601333618,
      "epoch": 3.239972035305427,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039671852806525613,
      "loss": 5.1884,
      "mean_token_accuracy": 0.21010754406452178,
      "num_tokens": 75120699.0,
      "step": 37075
    },
    {
      "entropy": 6.035027313232422,
      "epoch": 3.240408983658132,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00039669231913407065,
      "loss": 5.216,
      "mean_token_accuracy": 0.20798491239547728,
      "num_tokens": 75130892.0,
      "step": 37080
    },
    {
      "entropy": 6.095092153549194,
      "epoch": 3.2408459320108363,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003966661078687628,
      "loss": 5.2987,
      "mean_token_accuracy": 0.20690613985061646,
      "num_tokens": 75140994.0,
      "step": 37085
    },
    {
      "entropy": 6.047215223312378,
      "epoch": 3.241282880363541,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003966398942698352,
      "loss": 5.315,
      "mean_token_accuracy": 0.2020121306180954,
      "num_tokens": 75151084.0,
      "step": 37090
    },
    {
      "entropy": 6.0416257858276365,
      "epoch": 3.2417198287162456,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003966136783377907,
      "loss": 5.2489,
      "mean_token_accuracy": 0.20430784970521926,
      "num_tokens": 75160815.0,
      "step": 37095
    },
    {
      "entropy": 6.050641393661499,
      "epoch": 3.2421567770689506,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003965874600731321,
      "loss": 5.2023,
      "mean_token_accuracy": 0.21370913386344909,
      "num_tokens": 75170827.0,
      "step": 37100
    },
    {
      "entropy": 6.044087171554565,
      "epoch": 3.242593725421655,
      "grad_norm": 1.375,
      "learning_rate": 0.0003965612394763623,
      "loss": 5.2418,
      "mean_token_accuracy": 0.2091803476214409,
      "num_tokens": 75180250.0,
      "step": 37105
    },
    {
      "entropy": 6.021984243392945,
      "epoch": 3.24303067377436,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003965350165479842,
      "loss": 5.2497,
      "mean_token_accuracy": 0.20822651237249373,
      "num_tokens": 75190862.0,
      "step": 37110
    },
    {
      "entropy": 6.034952831268311,
      "epoch": 3.2434676221270644,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003965087912885009,
      "loss": 5.1993,
      "mean_token_accuracy": 0.21048382967710494,
      "num_tokens": 75199632.0,
      "step": 37115
    },
    {
      "entropy": 5.974981069564819,
      "epoch": 3.2439045704797693,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039648256369841523,
      "loss": 5.1593,
      "mean_token_accuracy": 0.21314139068126678,
      "num_tokens": 75209626.0,
      "step": 37120
    },
    {
      "entropy": 6.0096532821655275,
      "epoch": 3.2443415188324742,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003964563337782305,
      "loss": 5.2526,
      "mean_token_accuracy": 0.2001126542687416,
      "num_tokens": 75220682.0,
      "step": 37125
    },
    {
      "entropy": 6.059200954437256,
      "epoch": 3.2447784671851787,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039643010152844946,
      "loss": 5.235,
      "mean_token_accuracy": 0.2112404376268387,
      "num_tokens": 75230330.0,
      "step": 37130
    },
    {
      "entropy": 6.0707415580749515,
      "epoch": 3.2452154155378836,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039640386694957553,
      "loss": 5.2404,
      "mean_token_accuracy": 0.2078678786754608,
      "num_tokens": 75240485.0,
      "step": 37135
    },
    {
      "entropy": 6.032322645187378,
      "epoch": 3.245652363890588,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039637763004211175,
      "loss": 5.2091,
      "mean_token_accuracy": 0.20464661568403245,
      "num_tokens": 75250670.0,
      "step": 37140
    },
    {
      "entropy": 6.020496034622193,
      "epoch": 3.246089312243293,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039635139080656146,
      "loss": 5.2539,
      "mean_token_accuracy": 0.21140020489692687,
      "num_tokens": 75260312.0,
      "step": 37145
    },
    {
      "entropy": 6.031261539459228,
      "epoch": 3.2465262605959975,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039632514924342794,
      "loss": 5.2325,
      "mean_token_accuracy": 0.20464251935482025,
      "num_tokens": 75269978.0,
      "step": 37150
    },
    {
      "entropy": 6.007216358184815,
      "epoch": 3.2469632089487024,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00039629890535321444,
      "loss": 5.2481,
      "mean_token_accuracy": 0.20897120237350464,
      "num_tokens": 75279781.0,
      "step": 37155
    },
    {
      "entropy": 6.087862205505371,
      "epoch": 3.247400157301407,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039627265913642435,
      "loss": 5.3056,
      "mean_token_accuracy": 0.20380737334489823,
      "num_tokens": 75290352.0,
      "step": 37160
    },
    {
      "entropy": 6.087536811828613,
      "epoch": 3.2478371056541118,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003962464105935612,
      "loss": 5.2181,
      "mean_token_accuracy": 0.20851015746593476,
      "num_tokens": 75300511.0,
      "step": 37165
    },
    {
      "entropy": 6.002131557464599,
      "epoch": 3.2482740540068162,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039622015972512833,
      "loss": 5.25,
      "mean_token_accuracy": 0.2113138109445572,
      "num_tokens": 75310717.0,
      "step": 37170
    },
    {
      "entropy": 6.022753000259399,
      "epoch": 3.248711002359521,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003961939065316293,
      "loss": 5.2459,
      "mean_token_accuracy": 0.2085636302828789,
      "num_tokens": 75322402.0,
      "step": 37175
    },
    {
      "entropy": 5.992082595825195,
      "epoch": 3.2491479507122256,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039616765101356755,
      "loss": 5.1474,
      "mean_token_accuracy": 0.21260367780923844,
      "num_tokens": 75331866.0,
      "step": 37180
    },
    {
      "entropy": 6.025499868392944,
      "epoch": 3.2495848990649305,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039614139317144676,
      "loss": 5.1584,
      "mean_token_accuracy": 0.21936851292848586,
      "num_tokens": 75341125.0,
      "step": 37185
    },
    {
      "entropy": 6.000305032730102,
      "epoch": 3.250021847417635,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003961151330057706,
      "loss": 5.2149,
      "mean_token_accuracy": 0.21166122257709502,
      "num_tokens": 75350518.0,
      "step": 37190
    },
    {
      "entropy": 6.089756965637207,
      "epoch": 3.25045879577034,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003960888705170428,
      "loss": 5.3123,
      "mean_token_accuracy": 0.19385611712932588,
      "num_tokens": 75359330.0,
      "step": 37195
    },
    {
      "entropy": 6.064840936660767,
      "epoch": 3.250895744123045,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000396062605705767,
      "loss": 5.3024,
      "mean_token_accuracy": 0.2041465550661087,
      "num_tokens": 75368771.0,
      "step": 37200
    },
    {
      "entropy": 6.082134628295899,
      "epoch": 3.2513326924757493,
      "grad_norm": 1.3125,
      "learning_rate": 0.000396036338572447,
      "loss": 5.2596,
      "mean_token_accuracy": 0.20742612481117248,
      "num_tokens": 75379656.0,
      "step": 37205
    },
    {
      "entropy": 6.030158376693725,
      "epoch": 3.251769640828454,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039601006911758646,
      "loss": 5.2489,
      "mean_token_accuracy": 0.205889792740345,
      "num_tokens": 75390248.0,
      "step": 37210
    },
    {
      "entropy": 6.0015209197998045,
      "epoch": 3.2522065891811587,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003959837973416895,
      "loss": 5.2651,
      "mean_token_accuracy": 0.20127639174461365,
      "num_tokens": 75400090.0,
      "step": 37215
    },
    {
      "entropy": 6.0401129722595215,
      "epoch": 3.2526435375338636,
      "grad_norm": 1.25,
      "learning_rate": 0.00039595752324525984,
      "loss": 5.2502,
      "mean_token_accuracy": 0.20454029589891434,
      "num_tokens": 75410380.0,
      "step": 37220
    },
    {
      "entropy": 6.089219808578491,
      "epoch": 3.253080485886568,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003959312468288016,
      "loss": 5.1961,
      "mean_token_accuracy": 0.20981297940015792,
      "num_tokens": 75421265.0,
      "step": 37225
    },
    {
      "entropy": 6.049987745285034,
      "epoch": 3.253517434239273,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003959049680928187,
      "loss": 5.2175,
      "mean_token_accuracy": 0.2142067462205887,
      "num_tokens": 75432049.0,
      "step": 37230
    },
    {
      "entropy": 6.046846389770508,
      "epoch": 3.2539543825919774,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003958786870378151,
      "loss": 5.2788,
      "mean_token_accuracy": 0.21064915657043456,
      "num_tokens": 75441364.0,
      "step": 37235
    },
    {
      "entropy": 5.996960926055908,
      "epoch": 3.2543913309446824,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003958524036642949,
      "loss": 5.2809,
      "mean_token_accuracy": 0.20307852625846862,
      "num_tokens": 75452019.0,
      "step": 37240
    },
    {
      "entropy": 6.066973924636841,
      "epoch": 3.254828279297387,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039582611797276237,
      "loss": 5.2783,
      "mean_token_accuracy": 0.20092511922121048,
      "num_tokens": 75463053.0,
      "step": 37245
    },
    {
      "entropy": 6.040705442428589,
      "epoch": 3.2552652276500917,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003957998299637215,
      "loss": 5.2418,
      "mean_token_accuracy": 0.20377640575170516,
      "num_tokens": 75472757.0,
      "step": 37250
    },
    {
      "entropy": 5.972899675369263,
      "epoch": 3.2557021760027967,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003957735396376766,
      "loss": 5.1814,
      "mean_token_accuracy": 0.21395194977521897,
      "num_tokens": 75482228.0,
      "step": 37255
    },
    {
      "entropy": 5.996842575073242,
      "epoch": 3.256139124355501,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003957472469951319,
      "loss": 5.296,
      "mean_token_accuracy": 0.2049928903579712,
      "num_tokens": 75491344.0,
      "step": 37260
    },
    {
      "entropy": 6.016765546798706,
      "epoch": 3.256576072708206,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039572095203659177,
      "loss": 5.3092,
      "mean_token_accuracy": 0.20388765037059783,
      "num_tokens": 75502042.0,
      "step": 37265
    },
    {
      "entropy": 6.016802024841309,
      "epoch": 3.2570130210609105,
      "grad_norm": 1.375,
      "learning_rate": 0.0003956946547625606,
      "loss": 5.1936,
      "mean_token_accuracy": 0.21109022498130797,
      "num_tokens": 75512116.0,
      "step": 37270
    },
    {
      "entropy": 6.019095611572266,
      "epoch": 3.2574499694136154,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039566835517354266,
      "loss": 5.1969,
      "mean_token_accuracy": 0.21074136048555375,
      "num_tokens": 75521220.0,
      "step": 37275
    },
    {
      "entropy": 6.092329216003418,
      "epoch": 3.25788691776632,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003956420532700423,
      "loss": 5.2734,
      "mean_token_accuracy": 0.20600700676441192,
      "num_tokens": 75530556.0,
      "step": 37280
    },
    {
      "entropy": 6.022590970993042,
      "epoch": 3.258323866119025,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003956157490525643,
      "loss": 5.2165,
      "mean_token_accuracy": 0.2181480795145035,
      "num_tokens": 75539843.0,
      "step": 37285
    },
    {
      "entropy": 5.974350690841675,
      "epoch": 3.2587608144717293,
      "grad_norm": 1.34375,
      "learning_rate": 0.000395589442521613,
      "loss": 5.1711,
      "mean_token_accuracy": 0.21560779064893723,
      "num_tokens": 75549945.0,
      "step": 37290
    },
    {
      "entropy": 6.015915012359619,
      "epoch": 3.259197762824434,
      "grad_norm": 1.328125,
      "learning_rate": 0.00039556313367769297,
      "loss": 5.3285,
      "mean_token_accuracy": 0.20000385493040085,
      "num_tokens": 75560808.0,
      "step": 37295
    },
    {
      "entropy": 6.122086906433106,
      "epoch": 3.2596347111771387,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039553682252130886,
      "loss": 5.2997,
      "mean_token_accuracy": 0.20287356078624724,
      "num_tokens": 75571431.0,
      "step": 37300
    },
    {
      "entropy": 6.054490661621093,
      "epoch": 3.2600716595298436,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039551050905296543,
      "loss": 5.2036,
      "mean_token_accuracy": 0.20797986537218094,
      "num_tokens": 75582543.0,
      "step": 37305
    },
    {
      "entropy": 6.002942895889282,
      "epoch": 3.2605086078825485,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039548419327316715,
      "loss": 5.2041,
      "mean_token_accuracy": 0.21249691545963287,
      "num_tokens": 75591951.0,
      "step": 37310
    },
    {
      "entropy": 5.960797500610352,
      "epoch": 3.260945556235253,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000395457875182419,
      "loss": 5.2131,
      "mean_token_accuracy": 0.21148380786180496,
      "num_tokens": 75602107.0,
      "step": 37315
    },
    {
      "entropy": 5.971871852874756,
      "epoch": 3.261382504587958,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039543155478122564,
      "loss": 5.1674,
      "mean_token_accuracy": 0.21383659690618514,
      "num_tokens": 75612449.0,
      "step": 37320
    },
    {
      "entropy": 6.033534240722656,
      "epoch": 3.2618194529406623,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000395405232070092,
      "loss": 5.3043,
      "mean_token_accuracy": 0.20004461705684662,
      "num_tokens": 75622078.0,
      "step": 37325
    },
    {
      "entropy": 6.05167179107666,
      "epoch": 3.2622564012933672,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039537890704952294,
      "loss": 5.3104,
      "mean_token_accuracy": 0.20603896379470826,
      "num_tokens": 75632672.0,
      "step": 37330
    },
    {
      "entropy": 6.095401954650879,
      "epoch": 3.2626933496460717,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00039535257972002325,
      "loss": 5.2493,
      "mean_token_accuracy": 0.2095297485589981,
      "num_tokens": 75641633.0,
      "step": 37335
    },
    {
      "entropy": 6.002948045730591,
      "epoch": 3.2631302979987766,
      "grad_norm": 1.390625,
      "learning_rate": 0.00039532625008209814,
      "loss": 5.1559,
      "mean_token_accuracy": 0.2169978842139244,
      "num_tokens": 75651109.0,
      "step": 37340
    },
    {
      "entropy": 5.9873167991638185,
      "epoch": 3.263567246351481,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003952999181362525,
      "loss": 5.1974,
      "mean_token_accuracy": 0.21301135718822478,
      "num_tokens": 75661176.0,
      "step": 37345
    },
    {
      "entropy": 6.074447250366211,
      "epoch": 3.264004194704186,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00039527358388299135,
      "loss": 5.2606,
      "mean_token_accuracy": 0.20419962853193283,
      "num_tokens": 75670838.0,
      "step": 37350
    },
    {
      "entropy": 6.025091028213501,
      "epoch": 3.2644411430568905,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003952472473228198,
      "loss": 5.2178,
      "mean_token_accuracy": 0.20778876096010207,
      "num_tokens": 75681079.0,
      "step": 37355
    },
    {
      "entropy": 6.03684892654419,
      "epoch": 3.2648780914095954,
      "grad_norm": 1.28125,
      "learning_rate": 0.000395220908456243,
      "loss": 5.1244,
      "mean_token_accuracy": 0.2204086422920227,
      "num_tokens": 75691748.0,
      "step": 37360
    },
    {
      "entropy": 6.011173725128174,
      "epoch": 3.2653150397623003,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003951945672837663,
      "loss": 5.2519,
      "mean_token_accuracy": 0.2062473937869072,
      "num_tokens": 75702751.0,
      "step": 37365
    },
    {
      "entropy": 5.98567385673523,
      "epoch": 3.265751988115005,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039516822380589476,
      "loss": 5.2338,
      "mean_token_accuracy": 0.21259106695652008,
      "num_tokens": 75712107.0,
      "step": 37370
    },
    {
      "entropy": 6.068732500076294,
      "epoch": 3.2661889364677097,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003951418780231338,
      "loss": 5.2303,
      "mean_token_accuracy": 0.20782969743013383,
      "num_tokens": 75721595.0,
      "step": 37375
    },
    {
      "entropy": 5.99558835029602,
      "epoch": 3.266625884820414,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003951155299359886,
      "loss": 5.1592,
      "mean_token_accuracy": 0.21258829683065414,
      "num_tokens": 75731332.0,
      "step": 37380
    },
    {
      "entropy": 5.997040319442749,
      "epoch": 3.267062833173119,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003950891795449646,
      "loss": 5.1737,
      "mean_token_accuracy": 0.21465655714273452,
      "num_tokens": 75740425.0,
      "step": 37385
    },
    {
      "entropy": 5.976790523529052,
      "epoch": 3.2674997815258235,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003950628268505672,
      "loss": 5.2275,
      "mean_token_accuracy": 0.21473339349031448,
      "num_tokens": 75750091.0,
      "step": 37390
    },
    {
      "entropy": 6.055781841278076,
      "epoch": 3.2679367298785285,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039503647185330185,
      "loss": 5.2982,
      "mean_token_accuracy": 0.2018134132027626,
      "num_tokens": 75760780.0,
      "step": 37395
    },
    {
      "entropy": 6.060359382629395,
      "epoch": 3.268373678231233,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039501011455367406,
      "loss": 5.2296,
      "mean_token_accuracy": 0.209532667696476,
      "num_tokens": 75771056.0,
      "step": 37400
    },
    {
      "entropy": 5.964019346237182,
      "epoch": 3.268810626583938,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003949837549521895,
      "loss": 5.2047,
      "mean_token_accuracy": 0.20721628218889238,
      "num_tokens": 75781265.0,
      "step": 37405
    },
    {
      "entropy": 6.0093179702758786,
      "epoch": 3.2692475749366423,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039495739304935357,
      "loss": 5.2595,
      "mean_token_accuracy": 0.2079733818769455,
      "num_tokens": 75791911.0,
      "step": 37410
    },
    {
      "entropy": 5.981763076782227,
      "epoch": 3.2696845232893472,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000394931028845672,
      "loss": 5.1801,
      "mean_token_accuracy": 0.2124875858426094,
      "num_tokens": 75802085.0,
      "step": 37415
    },
    {
      "entropy": 6.028497409820557,
      "epoch": 3.270121471642052,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003949046623416504,
      "loss": 5.1819,
      "mean_token_accuracy": 0.21451418250799179,
      "num_tokens": 75810740.0,
      "step": 37420
    },
    {
      "entropy": 6.055467891693115,
      "epoch": 3.2705584199947566,
      "grad_norm": 1.359375,
      "learning_rate": 0.00039487829353779464,
      "loss": 5.2006,
      "mean_token_accuracy": 0.21383293718099594,
      "num_tokens": 75820260.0,
      "step": 37425
    },
    {
      "entropy": 6.017135906219482,
      "epoch": 3.2709953683474615,
      "grad_norm": 1.25,
      "learning_rate": 0.00039485192243461035,
      "loss": 5.1777,
      "mean_token_accuracy": 0.21822525113821029,
      "num_tokens": 75830152.0,
      "step": 37430
    },
    {
      "entropy": 5.942583656311035,
      "epoch": 3.271432316700166,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003948255490326034,
      "loss": 5.1654,
      "mean_token_accuracy": 0.21673533618450164,
      "num_tokens": 75840617.0,
      "step": 37435
    },
    {
      "entropy": 5.97212061882019,
      "epoch": 3.271869265052871,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003947991733322796,
      "loss": 5.1586,
      "mean_token_accuracy": 0.2181107744574547,
      "num_tokens": 75850114.0,
      "step": 37440
    },
    {
      "entropy": 5.937120962142944,
      "epoch": 3.2723062134055754,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003947727953341449,
      "loss": 5.1787,
      "mean_token_accuracy": 0.22137044221162797,
      "num_tokens": 75861202.0,
      "step": 37445
    },
    {
      "entropy": 6.077134609222412,
      "epoch": 3.2727431617582803,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039474641503870524,
      "loss": 5.1849,
      "mean_token_accuracy": 0.20456338971853255,
      "num_tokens": 75870613.0,
      "step": 37450
    },
    {
      "entropy": 6.035810041427612,
      "epoch": 3.2731801101109848,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039472003244646656,
      "loss": 5.2009,
      "mean_token_accuracy": 0.20848208516836167,
      "num_tokens": 75880550.0,
      "step": 37455
    },
    {
      "entropy": 5.947380304336548,
      "epoch": 3.2736170584636897,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003946936475579349,
      "loss": 5.1606,
      "mean_token_accuracy": 0.2173902854323387,
      "num_tokens": 75890838.0,
      "step": 37460
    },
    {
      "entropy": 5.997327423095703,
      "epoch": 3.274054006816394,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003946672603736164,
      "loss": 5.2574,
      "mean_token_accuracy": 0.21077151596546173,
      "num_tokens": 75900834.0,
      "step": 37465
    },
    {
      "entropy": 5.987120056152344,
      "epoch": 3.274490955169099,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039464087089401714,
      "loss": 5.2657,
      "mean_token_accuracy": 0.2104163497686386,
      "num_tokens": 75911442.0,
      "step": 37470
    },
    {
      "entropy": 6.062557601928711,
      "epoch": 3.274927903521804,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039461447911964327,
      "loss": 5.2832,
      "mean_token_accuracy": 0.20167320519685744,
      "num_tokens": 75922824.0,
      "step": 37475
    },
    {
      "entropy": 6.026437473297119,
      "epoch": 3.2753648518745084,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003945880850510011,
      "loss": 5.2441,
      "mean_token_accuracy": 0.2098516494035721,
      "num_tokens": 75932784.0,
      "step": 37480
    },
    {
      "entropy": 5.991314315795899,
      "epoch": 3.275801800227213,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039456168868859663,
      "loss": 5.1751,
      "mean_token_accuracy": 0.21953952461481094,
      "num_tokens": 75942176.0,
      "step": 37485
    },
    {
      "entropy": 5.950683116912842,
      "epoch": 3.276238748579918,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039453529003293647,
      "loss": 5.2007,
      "mean_token_accuracy": 0.21111409813165666,
      "num_tokens": 75952602.0,
      "step": 37490
    },
    {
      "entropy": 5.947474336624145,
      "epoch": 3.2766756969326227,
      "grad_norm": 1.359375,
      "learning_rate": 0.00039450888908452677,
      "loss": 5.2027,
      "mean_token_accuracy": 0.2077797159552574,
      "num_tokens": 75962971.0,
      "step": 37495
    },
    {
      "entropy": 6.056244325637818,
      "epoch": 3.277112645285327,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003944824858438739,
      "loss": 5.3043,
      "mean_token_accuracy": 0.20566579848527908,
      "num_tokens": 75972751.0,
      "step": 37500
    },
    {
      "entropy": 6.08010778427124,
      "epoch": 3.277549593638032,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003944560803114845,
      "loss": 5.3683,
      "mean_token_accuracy": 0.2002345561981201,
      "num_tokens": 75984653.0,
      "step": 37505
    },
    {
      "entropy": 6.077767610549927,
      "epoch": 3.2779865419907366,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003944296724878647,
      "loss": 5.2445,
      "mean_token_accuracy": 0.21045714914798735,
      "num_tokens": 75994211.0,
      "step": 37510
    },
    {
      "entropy": 5.98509726524353,
      "epoch": 3.2784234903434415,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003944032623735213,
      "loss": 5.1797,
      "mean_token_accuracy": 0.21151013374328614,
      "num_tokens": 76003891.0,
      "step": 37515
    },
    {
      "entropy": 6.007781600952148,
      "epoch": 3.278860438696146,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003943768499689607,
      "loss": 5.2239,
      "mean_token_accuracy": 0.20857033431529998,
      "num_tokens": 76014625.0,
      "step": 37520
    },
    {
      "entropy": 6.006893825531006,
      "epoch": 3.279297387048851,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039435043527468963,
      "loss": 5.1895,
      "mean_token_accuracy": 0.20880076736211778,
      "num_tokens": 76024773.0,
      "step": 37525
    },
    {
      "entropy": 5.945037841796875,
      "epoch": 3.2797343354015553,
      "grad_norm": 1.5,
      "learning_rate": 0.00039432401829121467,
      "loss": 5.0829,
      "mean_token_accuracy": 0.22144752144813537,
      "num_tokens": 76035085.0,
      "step": 37530
    },
    {
      "entropy": 6.014023113250732,
      "epoch": 3.2801712837542603,
      "grad_norm": 1.25,
      "learning_rate": 0.0003942975990190425,
      "loss": 5.2791,
      "mean_token_accuracy": 0.20375100523233414,
      "num_tokens": 76046055.0,
      "step": 37535
    },
    {
      "entropy": 5.966177558898925,
      "epoch": 3.2806082321069647,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003942711774586799,
      "loss": 5.154,
      "mean_token_accuracy": 0.22104986608028412,
      "num_tokens": 76055761.0,
      "step": 37540
    },
    {
      "entropy": 6.033720445632935,
      "epoch": 3.2810451804596696,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003942447536106335,
      "loss": 5.3424,
      "mean_token_accuracy": 0.19813776165246963,
      "num_tokens": 76065844.0,
      "step": 37545
    },
    {
      "entropy": 6.077080917358399,
      "epoch": 3.2814821288123746,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039421832747541034,
      "loss": 5.2207,
      "mean_token_accuracy": 0.2086271107196808,
      "num_tokens": 76076048.0,
      "step": 37550
    },
    {
      "entropy": 6.025246858596802,
      "epoch": 3.281919077165079,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003941918990535172,
      "loss": 5.1894,
      "mean_token_accuracy": 0.21143790781497956,
      "num_tokens": 76086775.0,
      "step": 37555
    },
    {
      "entropy": 6.006286907196045,
      "epoch": 3.282356025517784,
      "grad_norm": 1.25,
      "learning_rate": 0.0003941654683454609,
      "loss": 5.2594,
      "mean_token_accuracy": 0.2056177005171776,
      "num_tokens": 76097492.0,
      "step": 37560
    },
    {
      "entropy": 6.02746901512146,
      "epoch": 3.2827929738704884,
      "grad_norm": 1.25,
      "learning_rate": 0.00039413903535174855,
      "loss": 5.2334,
      "mean_token_accuracy": 0.20456754714250563,
      "num_tokens": 76106709.0,
      "step": 37565
    },
    {
      "entropy": 6.028074550628662,
      "epoch": 3.2832299222231933,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000394112600072887,
      "loss": 5.2179,
      "mean_token_accuracy": 0.2092749536037445,
      "num_tokens": 76117189.0,
      "step": 37570
    },
    {
      "entropy": 6.068081569671631,
      "epoch": 3.283666870575898,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039408616250938353,
      "loss": 5.2972,
      "mean_token_accuracy": 0.20584893822669983,
      "num_tokens": 76127624.0,
      "step": 37575
    },
    {
      "entropy": 6.055733919143677,
      "epoch": 3.2841038189286027,
      "grad_norm": 1.125,
      "learning_rate": 0.00039405972266174494,
      "loss": 5.2364,
      "mean_token_accuracy": 0.20669031739234925,
      "num_tokens": 76138138.0,
      "step": 37580
    },
    {
      "entropy": 6.001974630355835,
      "epoch": 3.284540767281307,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039403328053047847,
      "loss": 5.1022,
      "mean_token_accuracy": 0.22325088679790497,
      "num_tokens": 76148780.0,
      "step": 37585
    },
    {
      "entropy": 5.9811906814575195,
      "epoch": 3.284977715634012,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003940068361160913,
      "loss": 5.1726,
      "mean_token_accuracy": 0.21933709233999252,
      "num_tokens": 76158468.0,
      "step": 37590
    },
    {
      "entropy": 6.054929399490357,
      "epoch": 3.2854146639867166,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039398038941909066,
      "loss": 5.2869,
      "mean_token_accuracy": 0.2030933067202568,
      "num_tokens": 76168789.0,
      "step": 37595
    },
    {
      "entropy": 6.033947849273682,
      "epoch": 3.2858516123394215,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003939539404399839,
      "loss": 5.2023,
      "mean_token_accuracy": 0.20735421776771545,
      "num_tokens": 76179355.0,
      "step": 37600
    },
    {
      "entropy": 6.121558380126953,
      "epoch": 3.2862885606921264,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003939274891792781,
      "loss": 5.2773,
      "mean_token_accuracy": 0.20421785563230516,
      "num_tokens": 76189235.0,
      "step": 37605
    },
    {
      "entropy": 5.995529270172119,
      "epoch": 3.286725509044831,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039390103563748086,
      "loss": 5.1837,
      "mean_token_accuracy": 0.21344025582075118,
      "num_tokens": 76200146.0,
      "step": 37610
    },
    {
      "entropy": 5.997178030014038,
      "epoch": 3.2871624573975358,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039387457981509937,
      "loss": 5.2286,
      "mean_token_accuracy": 0.21135546118021012,
      "num_tokens": 76209519.0,
      "step": 37615
    },
    {
      "entropy": 6.010632705688477,
      "epoch": 3.2875994057502402,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003938481217126412,
      "loss": 5.2071,
      "mean_token_accuracy": 0.21059483289718628,
      "num_tokens": 76219316.0,
      "step": 37620
    },
    {
      "entropy": 5.990713262557984,
      "epoch": 3.288036354102945,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003938216613306138,
      "loss": 5.1981,
      "mean_token_accuracy": 0.21193728297948838,
      "num_tokens": 76230707.0,
      "step": 37625
    },
    {
      "entropy": 6.070076274871826,
      "epoch": 3.2884733024556496,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039379519866952463,
      "loss": 5.3028,
      "mean_token_accuracy": 0.20138420015573502,
      "num_tokens": 76240197.0,
      "step": 37630
    },
    {
      "entropy": 6.068628406524658,
      "epoch": 3.2889102508083545,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039376873372988123,
      "loss": 5.2671,
      "mean_token_accuracy": 0.20495824068784713,
      "num_tokens": 76249292.0,
      "step": 37635
    },
    {
      "entropy": 6.002909803390503,
      "epoch": 3.289347199161059,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039374226651219137,
      "loss": 5.1629,
      "mean_token_accuracy": 0.2175462394952774,
      "num_tokens": 76259598.0,
      "step": 37640
    },
    {
      "entropy": 6.0300017356872555,
      "epoch": 3.289784147513764,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003937157970169626,
      "loss": 5.2544,
      "mean_token_accuracy": 0.2107554107904434,
      "num_tokens": 76269624.0,
      "step": 37645
    },
    {
      "entropy": 6.020542621612549,
      "epoch": 3.2902210958664684,
      "grad_norm": 1.9375,
      "learning_rate": 0.0003936893252447026,
      "loss": 5.2236,
      "mean_token_accuracy": 0.207684288918972,
      "num_tokens": 76280662.0,
      "step": 37650
    },
    {
      "entropy": 6.014252614974976,
      "epoch": 3.2906580442191733,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003936628511959193,
      "loss": 5.2056,
      "mean_token_accuracy": 0.20949365347623825,
      "num_tokens": 76291194.0,
      "step": 37655
    },
    {
      "entropy": 5.982959508895874,
      "epoch": 3.291094992571878,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003936363748711202,
      "loss": 5.186,
      "mean_token_accuracy": 0.21796730905771255,
      "num_tokens": 76301926.0,
      "step": 37660
    },
    {
      "entropy": 6.0572840690612795,
      "epoch": 3.2915319409245827,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003936098962708132,
      "loss": 5.2617,
      "mean_token_accuracy": 0.2121116816997528,
      "num_tokens": 76313126.0,
      "step": 37665
    },
    {
      "entropy": 6.04497799873352,
      "epoch": 3.2919688892772876,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003935834153955064,
      "loss": 5.1886,
      "mean_token_accuracy": 0.21165499240159988,
      "num_tokens": 76322133.0,
      "step": 37670
    },
    {
      "entropy": 5.972199869155884,
      "epoch": 3.292405837629992,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003935569322457075,
      "loss": 5.1532,
      "mean_token_accuracy": 0.2225572034716606,
      "num_tokens": 76332459.0,
      "step": 37675
    },
    {
      "entropy": 5.978463745117187,
      "epoch": 3.292842785982697,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003935304468219245,
      "loss": 5.2248,
      "mean_token_accuracy": 0.20985858738422394,
      "num_tokens": 76342510.0,
      "step": 37680
    },
    {
      "entropy": 5.985598134994507,
      "epoch": 3.2932797343354014,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003935039591246654,
      "loss": 5.2525,
      "mean_token_accuracy": 0.20689485222101212,
      "num_tokens": 76352164.0,
      "step": 37685
    },
    {
      "entropy": 6.015766048431397,
      "epoch": 3.2937166826881064,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003934774691544383,
      "loss": 5.2255,
      "mean_token_accuracy": 0.20951447039842605,
      "num_tokens": 76362141.0,
      "step": 37690
    },
    {
      "entropy": 6.072568607330322,
      "epoch": 3.294153631040811,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003934509769117513,
      "loss": 5.2956,
      "mean_token_accuracy": 0.20640805065631868,
      "num_tokens": 76372152.0,
      "step": 37695
    },
    {
      "entropy": 6.054835605621338,
      "epoch": 3.2945905793935157,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003934244823971124,
      "loss": 5.2397,
      "mean_token_accuracy": 0.2096678078174591,
      "num_tokens": 76383033.0,
      "step": 37700
    },
    {
      "entropy": 6.083175182342529,
      "epoch": 3.29502752774622,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039339798561102994,
      "loss": 5.2371,
      "mean_token_accuracy": 0.20593275278806686,
      "num_tokens": 76393552.0,
      "step": 37705
    },
    {
      "entropy": 5.963424730300903,
      "epoch": 3.295464476098925,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003933714865540121,
      "loss": 5.1506,
      "mean_token_accuracy": 0.21285925954580306,
      "num_tokens": 76404363.0,
      "step": 37710
    },
    {
      "entropy": 6.001213455200196,
      "epoch": 3.29590142445163,
      "grad_norm": 1.34375,
      "learning_rate": 0.00039334498522656706,
      "loss": 5.185,
      "mean_token_accuracy": 0.20794504433870314,
      "num_tokens": 76413483.0,
      "step": 37715
    },
    {
      "entropy": 5.987973690032959,
      "epoch": 3.2963383728043345,
      "grad_norm": 1.25,
      "learning_rate": 0.0003933184816292032,
      "loss": 5.2452,
      "mean_token_accuracy": 0.20616123527288438,
      "num_tokens": 76424786.0,
      "step": 37720
    },
    {
      "entropy": 5.995103979110718,
      "epoch": 3.2967753211570394,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003932919757624289,
      "loss": 5.1586,
      "mean_token_accuracy": 0.21577529907226561,
      "num_tokens": 76434923.0,
      "step": 37725
    },
    {
      "entropy": 6.059745645523071,
      "epoch": 3.297212269509744,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039326546762675254,
      "loss": 5.2198,
      "mean_token_accuracy": 0.2125896319746971,
      "num_tokens": 76444999.0,
      "step": 37730
    },
    {
      "entropy": 5.973453140258789,
      "epoch": 3.297649217862449,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00039323895722268243,
      "loss": 5.217,
      "mean_token_accuracy": 0.20775442719459533,
      "num_tokens": 76454869.0,
      "step": 37735
    },
    {
      "entropy": 6.016425466537475,
      "epoch": 3.2980861662151533,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003932124445507273,
      "loss": 5.2337,
      "mean_token_accuracy": 0.2046534851193428,
      "num_tokens": 76465336.0,
      "step": 37740
    },
    {
      "entropy": 6.029199314117432,
      "epoch": 3.298523114567858,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039318592961139545,
      "loss": 5.2257,
      "mean_token_accuracy": 0.21268371790647506,
      "num_tokens": 76476142.0,
      "step": 37745
    },
    {
      "entropy": 6.030785083770752,
      "epoch": 3.2989600629205627,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00039315941240519557,
      "loss": 5.2412,
      "mean_token_accuracy": 0.21906449645757675,
      "num_tokens": 76486194.0,
      "step": 37750
    },
    {
      "entropy": 6.056823825836181,
      "epoch": 3.2993970112732676,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003931328929326363,
      "loss": 5.3257,
      "mean_token_accuracy": 0.19976802170276642,
      "num_tokens": 76496243.0,
      "step": 37755
    },
    {
      "entropy": 5.997016096115113,
      "epoch": 3.299833959625972,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039310637119422607,
      "loss": 5.2187,
      "mean_token_accuracy": 0.20992475003004074,
      "num_tokens": 76506661.0,
      "step": 37760
    },
    {
      "entropy": 6.029164505004883,
      "epoch": 3.300270907978677,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000393079847190474,
      "loss": 5.2354,
      "mean_token_accuracy": 0.21117227524518967,
      "num_tokens": 76516812.0,
      "step": 37765
    },
    {
      "entropy": 5.998880386352539,
      "epoch": 3.300707856331382,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003930533209218884,
      "loss": 5.2676,
      "mean_token_accuracy": 0.20690028369426727,
      "num_tokens": 76527280.0,
      "step": 37770
    },
    {
      "entropy": 5.929000902175903,
      "epoch": 3.3011448046840863,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039302679238897826,
      "loss": 5.1158,
      "mean_token_accuracy": 0.22383293509483337,
      "num_tokens": 76536706.0,
      "step": 37775
    },
    {
      "entropy": 6.01091570854187,
      "epoch": 3.3015817530367912,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003930002615922525,
      "loss": 5.1537,
      "mean_token_accuracy": 0.21011918485164643,
      "num_tokens": 76546574.0,
      "step": 37780
    },
    {
      "entropy": 6.0497277736663815,
      "epoch": 3.3020187013894957,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003929737285322198,
      "loss": 5.2318,
      "mean_token_accuracy": 0.21524944007396699,
      "num_tokens": 76556315.0,
      "step": 37785
    },
    {
      "entropy": 6.0158604145050045,
      "epoch": 3.3024556497422006,
      "grad_norm": 1.34375,
      "learning_rate": 0.00039294719320938923,
      "loss": 5.3022,
      "mean_token_accuracy": 0.1985819786787033,
      "num_tokens": 76565721.0,
      "step": 37790
    },
    {
      "entropy": 6.072811651229858,
      "epoch": 3.302892598094905,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039292065562426966,
      "loss": 5.2936,
      "mean_token_accuracy": 0.20152398943901062,
      "num_tokens": 76575524.0,
      "step": 37795
    },
    {
      "entropy": 6.1039492130279545,
      "epoch": 3.30332954644761,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039289411577737014,
      "loss": 5.2521,
      "mean_token_accuracy": 0.20765907615423201,
      "num_tokens": 76586653.0,
      "step": 37800
    },
    {
      "entropy": 6.045510005950928,
      "epoch": 3.3037664948003145,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039286757366919963,
      "loss": 5.2079,
      "mean_token_accuracy": 0.21137968003749846,
      "num_tokens": 76597176.0,
      "step": 37805
    },
    {
      "entropy": 5.981727933883667,
      "epoch": 3.3042034431530194,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003928410293002674,
      "loss": 5.2268,
      "mean_token_accuracy": 0.20828861594200135,
      "num_tokens": 76608010.0,
      "step": 37810
    },
    {
      "entropy": 5.98031759262085,
      "epoch": 3.304640391505724,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003928144826710824,
      "loss": 5.1132,
      "mean_token_accuracy": 0.21870255023241042,
      "num_tokens": 76618029.0,
      "step": 37815
    },
    {
      "entropy": 6.066015434265137,
      "epoch": 3.3050773398584288,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039278793378215394,
      "loss": 5.2689,
      "mean_token_accuracy": 0.20636983215808868,
      "num_tokens": 76629529.0,
      "step": 37820
    },
    {
      "entropy": 6.064904022216797,
      "epoch": 3.3055142882111337,
      "grad_norm": 1.375,
      "learning_rate": 0.0003927613826339912,
      "loss": 5.3067,
      "mean_token_accuracy": 0.20704226940870285,
      "num_tokens": 76640488.0,
      "step": 37825
    },
    {
      "entropy": 5.997753572463989,
      "epoch": 3.305951236563838,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039273482922710343,
      "loss": 5.0964,
      "mean_token_accuracy": 0.21931450366973876,
      "num_tokens": 76650271.0,
      "step": 37830
    },
    {
      "entropy": 5.986335706710816,
      "epoch": 3.3063881849165426,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003927082735619999,
      "loss": 5.287,
      "mean_token_accuracy": 0.2048288941383362,
      "num_tokens": 76661413.0,
      "step": 37835
    },
    {
      "entropy": 5.965221786499024,
      "epoch": 3.3068251332692475,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039268171563919,
      "loss": 5.251,
      "mean_token_accuracy": 0.2062004342675209,
      "num_tokens": 76671055.0,
      "step": 37840
    },
    {
      "entropy": 6.122143888473511,
      "epoch": 3.3072620816219525,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003926551554591832,
      "loss": 5.3629,
      "mean_token_accuracy": 0.20107874423265457,
      "num_tokens": 76683472.0,
      "step": 37845
    },
    {
      "entropy": 6.064175081253052,
      "epoch": 3.307699029974657,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039262859302248876,
      "loss": 5.2092,
      "mean_token_accuracy": 0.21236609667539597,
      "num_tokens": 76694641.0,
      "step": 37850
    },
    {
      "entropy": 6.036947631835938,
      "epoch": 3.308135978327362,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003926020283296164,
      "loss": 5.2053,
      "mean_token_accuracy": 0.21216341853141785,
      "num_tokens": 76704737.0,
      "step": 37855
    },
    {
      "entropy": 5.941871452331543,
      "epoch": 3.3085729266800663,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003925754613810754,
      "loss": 5.2076,
      "mean_token_accuracy": 0.21077127903699874,
      "num_tokens": 76714157.0,
      "step": 37860
    },
    {
      "entropy": 5.98871078491211,
      "epoch": 3.309009875032771,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039254889217737554,
      "loss": 5.2111,
      "mean_token_accuracy": 0.2125512182712555,
      "num_tokens": 76724488.0,
      "step": 37865
    },
    {
      "entropy": 6.068892240524292,
      "epoch": 3.3094468233854757,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003925223207190263,
      "loss": 5.304,
      "mean_token_accuracy": 0.20689622461795806,
      "num_tokens": 76734700.0,
      "step": 37870
    },
    {
      "entropy": 6.054723072052002,
      "epoch": 3.3098837717381806,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003924957470065373,
      "loss": 5.2227,
      "mean_token_accuracy": 0.21106523424386978,
      "num_tokens": 76744578.0,
      "step": 37875
    },
    {
      "entropy": 5.993359184265136,
      "epoch": 3.310320720090885,
      "grad_norm": 1.4375,
      "learning_rate": 0.00039246917104041843,
      "loss": 5.203,
      "mean_token_accuracy": 0.2146749898791313,
      "num_tokens": 76754690.0,
      "step": 37880
    },
    {
      "entropy": 6.113721609115601,
      "epoch": 3.31075766844359,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003924425928211792,
      "loss": 5.2864,
      "mean_token_accuracy": 0.20350264757871628,
      "num_tokens": 76765197.0,
      "step": 37885
    },
    {
      "entropy": 6.03808708190918,
      "epoch": 3.3111946167962945,
      "grad_norm": 1.40625,
      "learning_rate": 0.00039241601234932954,
      "loss": 5.2258,
      "mean_token_accuracy": 0.21413165777921678,
      "num_tokens": 76774985.0,
      "step": 37890
    },
    {
      "entropy": 6.036290788650513,
      "epoch": 3.3116315651489994,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003923894296253792,
      "loss": 5.2463,
      "mean_token_accuracy": 0.21372751295566558,
      "num_tokens": 76784033.0,
      "step": 37895
    },
    {
      "entropy": 5.998266744613647,
      "epoch": 3.3120685135017043,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003923628446498382,
      "loss": 5.2486,
      "mean_token_accuracy": 0.20927799344062806,
      "num_tokens": 76794826.0,
      "step": 37900
    },
    {
      "entropy": 5.996213912963867,
      "epoch": 3.3125054618544088,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003923362574232163,
      "loss": 5.1769,
      "mean_token_accuracy": 0.21139649003744126,
      "num_tokens": 76805429.0,
      "step": 37905
    },
    {
      "entropy": 6.083654451370239,
      "epoch": 3.3129424102071137,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003923096679460234,
      "loss": 5.2503,
      "mean_token_accuracy": 0.2028746634721756,
      "num_tokens": 76814880.0,
      "step": 37910
    },
    {
      "entropy": 6.007679414749146,
      "epoch": 3.313379358559818,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003922830762187697,
      "loss": 5.2329,
      "mean_token_accuracy": 0.2121048778295517,
      "num_tokens": 76824828.0,
      "step": 37915
    },
    {
      "entropy": 6.070845651626587,
      "epoch": 3.313816306912523,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000392256482241965,
      "loss": 5.3612,
      "mean_token_accuracy": 0.19823428243398666,
      "num_tokens": 76835140.0,
      "step": 37920
    },
    {
      "entropy": 6.073997831344604,
      "epoch": 3.3142532552652275,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003922298860161197,
      "loss": 5.2585,
      "mean_token_accuracy": 0.20889368802309036,
      "num_tokens": 76844718.0,
      "step": 37925
    },
    {
      "entropy": 6.112758779525757,
      "epoch": 3.3146902036179324,
      "grad_norm": 1.34375,
      "learning_rate": 0.00039220328754174376,
      "loss": 5.3368,
      "mean_token_accuracy": 0.2056504890322685,
      "num_tokens": 76854738.0,
      "step": 37930
    },
    {
      "entropy": 6.0468658924102785,
      "epoch": 3.315127151970637,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039217668681934727,
      "loss": 5.2699,
      "mean_token_accuracy": 0.20731210559606553,
      "num_tokens": 76864939.0,
      "step": 37935
    },
    {
      "entropy": 6.0084085941314695,
      "epoch": 3.315564100323342,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003921500838494405,
      "loss": 5.2676,
      "mean_token_accuracy": 0.2059336856007576,
      "num_tokens": 76875895.0,
      "step": 37940
    },
    {
      "entropy": 6.011057138442993,
      "epoch": 3.3160010486760463,
      "grad_norm": 1.25,
      "learning_rate": 0.00039212347863253386,
      "loss": 5.2214,
      "mean_token_accuracy": 0.2089844435453415,
      "num_tokens": 76885415.0,
      "step": 37945
    },
    {
      "entropy": 5.998034811019897,
      "epoch": 3.316437997028751,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003920968711691374,
      "loss": 5.2533,
      "mean_token_accuracy": 0.20775964111089706,
      "num_tokens": 76896961.0,
      "step": 37950
    },
    {
      "entropy": 6.043392086029053,
      "epoch": 3.316874945381456,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039207026145976167,
      "loss": 5.2524,
      "mean_token_accuracy": 0.21136167049407958,
      "num_tokens": 76906521.0,
      "step": 37955
    },
    {
      "entropy": 6.0508904457092285,
      "epoch": 3.3173118937341606,
      "grad_norm": 1.71875,
      "learning_rate": 0.00039204364950491693,
      "loss": 5.2017,
      "mean_token_accuracy": 0.2065824180841446,
      "num_tokens": 76916971.0,
      "step": 37960
    },
    {
      "entropy": 6.037310171127319,
      "epoch": 3.3177488420868655,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003920170353051137,
      "loss": 5.2434,
      "mean_token_accuracy": 0.2107202097773552,
      "num_tokens": 76928070.0,
      "step": 37965
    },
    {
      "entropy": 6.018146562576294,
      "epoch": 3.31818579043957,
      "grad_norm": 1.328125,
      "learning_rate": 0.00039199041886086233,
      "loss": 5.2729,
      "mean_token_accuracy": 0.20533475279808044,
      "num_tokens": 76938020.0,
      "step": 37970
    },
    {
      "entropy": 6.029195070266724,
      "epoch": 3.318622738792275,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003919638001726735,
      "loss": 5.2605,
      "mean_token_accuracy": 0.2130868226289749,
      "num_tokens": 76949371.0,
      "step": 37975
    },
    {
      "entropy": 6.07093071937561,
      "epoch": 3.3190596871449793,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003919371792410578,
      "loss": 5.1784,
      "mean_token_accuracy": 0.2118965968489647,
      "num_tokens": 76959120.0,
      "step": 37980
    },
    {
      "entropy": 5.997877359390259,
      "epoch": 3.3194966354976843,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039191055606652554,
      "loss": 5.1881,
      "mean_token_accuracy": 0.21719214022159578,
      "num_tokens": 76969493.0,
      "step": 37985
    },
    {
      "entropy": 5.997804975509643,
      "epoch": 3.3199335838503887,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039188393064958766,
      "loss": 5.2372,
      "mean_token_accuracy": 0.20735471248626708,
      "num_tokens": 76979732.0,
      "step": 37990
    },
    {
      "entropy": 6.050695896148682,
      "epoch": 3.3203705322030936,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039185730299075473,
      "loss": 5.2642,
      "mean_token_accuracy": 0.20672716796398163,
      "num_tokens": 76990354.0,
      "step": 37995
    },
    {
      "entropy": 6.058109092712402,
      "epoch": 3.320807480555798,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039183067309053743,
      "loss": 5.199,
      "mean_token_accuracy": 0.21390963196754456,
      "num_tokens": 76999469.0,
      "step": 38000
    },
    {
      "entropy": 6.035577869415283,
      "epoch": 3.321244428908503,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003918040409494467,
      "loss": 5.2072,
      "mean_token_accuracy": 0.21691007763147355,
      "num_tokens": 77009499.0,
      "step": 38005
    },
    {
      "entropy": 6.014244508743286,
      "epoch": 3.321681377261208,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039177740656799314,
      "loss": 5.2891,
      "mean_token_accuracy": 0.19553805142641068,
      "num_tokens": 77019426.0,
      "step": 38010
    },
    {
      "entropy": 6.080189943313599,
      "epoch": 3.3221183256139124,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003917507699466879,
      "loss": 5.2358,
      "mean_token_accuracy": 0.19895967990159988,
      "num_tokens": 77028274.0,
      "step": 38015
    },
    {
      "entropy": 6.006098651885987,
      "epoch": 3.3225552739666173,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003917241310860416,
      "loss": 5.1837,
      "mean_token_accuracy": 0.21298268288373948,
      "num_tokens": 77038790.0,
      "step": 38020
    },
    {
      "entropy": 6.059634447097778,
      "epoch": 3.322992222319322,
      "grad_norm": 1.4375,
      "learning_rate": 0.00039169748998656536,
      "loss": 5.217,
      "mean_token_accuracy": 0.21664568781852722,
      "num_tokens": 77047799.0,
      "step": 38025
    },
    {
      "entropy": 6.034886741638184,
      "epoch": 3.3234291706720267,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003916708466487701,
      "loss": 5.313,
      "mean_token_accuracy": 0.19422175586223603,
      "num_tokens": 77057080.0,
      "step": 38030
    },
    {
      "entropy": 6.065567350387573,
      "epoch": 3.323866119024731,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039164420107316676,
      "loss": 5.2637,
      "mean_token_accuracy": 0.21033950597047807,
      "num_tokens": 77066689.0,
      "step": 38035
    },
    {
      "entropy": 6.117749214172363,
      "epoch": 3.324303067377436,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003916175532602666,
      "loss": 5.2796,
      "mean_token_accuracy": 0.20428736358880997,
      "num_tokens": 77076366.0,
      "step": 38040
    },
    {
      "entropy": 5.982742977142334,
      "epoch": 3.3247400157301406,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003915909032105808,
      "loss": 5.1573,
      "mean_token_accuracy": 0.22331125438213348,
      "num_tokens": 77085825.0,
      "step": 38045
    },
    {
      "entropy": 6.043070316314697,
      "epoch": 3.3251769640828455,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003915642509246202,
      "loss": 5.1717,
      "mean_token_accuracy": 0.21710287183523178,
      "num_tokens": 77095249.0,
      "step": 38050
    },
    {
      "entropy": 5.9645936489105225,
      "epoch": 3.32561391243555,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039153759640289627,
      "loss": 5.1753,
      "mean_token_accuracy": 0.2111981138586998,
      "num_tokens": 77105323.0,
      "step": 38055
    },
    {
      "entropy": 6.040085363388061,
      "epoch": 3.326050860788255,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039151093964592007,
      "loss": 5.2638,
      "mean_token_accuracy": 0.20444135665893554,
      "num_tokens": 77116055.0,
      "step": 38060
    },
    {
      "entropy": 6.049335718154907,
      "epoch": 3.3264878091409598,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003914842806542031,
      "loss": 5.2553,
      "mean_token_accuracy": 0.209307436645031,
      "num_tokens": 77127293.0,
      "step": 38065
    },
    {
      "entropy": 6.050296068191528,
      "epoch": 3.3269247574936642,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003914576194282566,
      "loss": 5.2321,
      "mean_token_accuracy": 0.20282004326581954,
      "num_tokens": 77136872.0,
      "step": 38070
    },
    {
      "entropy": 6.07238278388977,
      "epoch": 3.327361705846369,
      "grad_norm": 1.390625,
      "learning_rate": 0.00039143095596859184,
      "loss": 5.2586,
      "mean_token_accuracy": 0.2053637221455574,
      "num_tokens": 77146561.0,
      "step": 38075
    },
    {
      "entropy": 6.038881063461304,
      "epoch": 3.3277986541990736,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039140429027572043,
      "loss": 5.2085,
      "mean_token_accuracy": 0.2159213587641716,
      "num_tokens": 77156043.0,
      "step": 38080
    },
    {
      "entropy": 6.034265375137329,
      "epoch": 3.3282356025517785,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003913776223501537,
      "loss": 5.2355,
      "mean_token_accuracy": 0.20298729389905928,
      "num_tokens": 77165578.0,
      "step": 38085
    },
    {
      "entropy": 6.021491813659668,
      "epoch": 3.328672550904483,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003913509521924032,
      "loss": 5.2747,
      "mean_token_accuracy": 0.20686433911323548,
      "num_tokens": 77175105.0,
      "step": 38090
    },
    {
      "entropy": 6.013903284072876,
      "epoch": 3.329109499257188,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003913242798029805,
      "loss": 5.2648,
      "mean_token_accuracy": 0.2013137996196747,
      "num_tokens": 77185821.0,
      "step": 38095
    },
    {
      "entropy": 5.979293346405029,
      "epoch": 3.3295464476098924,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003912976051823971,
      "loss": 5.1784,
      "mean_token_accuracy": 0.21035934686660768,
      "num_tokens": 77195028.0,
      "step": 38100
    },
    {
      "entropy": 6.05112361907959,
      "epoch": 3.3299833959625973,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003912709283311648,
      "loss": 5.1975,
      "mean_token_accuracy": 0.21030496805906296,
      "num_tokens": 77206317.0,
      "step": 38105
    },
    {
      "entropy": 5.981480503082276,
      "epoch": 3.3304203443153018,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003912442492497951,
      "loss": 5.1611,
      "mean_token_accuracy": 0.2136104092001915,
      "num_tokens": 77216485.0,
      "step": 38110
    },
    {
      "entropy": 5.978899431228638,
      "epoch": 3.3308572926680067,
      "grad_norm": 1.375,
      "learning_rate": 0.0003912175679387997,
      "loss": 5.2119,
      "mean_token_accuracy": 0.21968381404876708,
      "num_tokens": 77226457.0,
      "step": 38115
    },
    {
      "entropy": 5.95964412689209,
      "epoch": 3.3312942410207116,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003911908843986906,
      "loss": 5.1763,
      "mean_token_accuracy": 0.21295909136533736,
      "num_tokens": 77237673.0,
      "step": 38120
    },
    {
      "entropy": 6.035247373580932,
      "epoch": 3.331731189373416,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003911641986299793,
      "loss": 5.1863,
      "mean_token_accuracy": 0.21351920068264008,
      "num_tokens": 77247620.0,
      "step": 38125
    },
    {
      "entropy": 6.08080530166626,
      "epoch": 3.3321681377261205,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039113751063317795,
      "loss": 5.1854,
      "mean_token_accuracy": 0.21142343580722808,
      "num_tokens": 77257739.0,
      "step": 38130
    },
    {
      "entropy": 6.020953941345215,
      "epoch": 3.3326050860788254,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003911108204087982,
      "loss": 5.2681,
      "mean_token_accuracy": 0.2044183388352394,
      "num_tokens": 77268917.0,
      "step": 38135
    },
    {
      "entropy": 6.035220289230347,
      "epoch": 3.3330420344315304,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003910841279573521,
      "loss": 5.277,
      "mean_token_accuracy": 0.2062713995575905,
      "num_tokens": 77279225.0,
      "step": 38140
    },
    {
      "entropy": 6.020593690872192,
      "epoch": 3.333478982784235,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003910574332793516,
      "loss": 5.2096,
      "mean_token_accuracy": 0.21484820246696473,
      "num_tokens": 77289504.0,
      "step": 38145
    },
    {
      "entropy": 6.034152555465698,
      "epoch": 3.3339159311369397,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039103073637530877,
      "loss": 5.2402,
      "mean_token_accuracy": 0.2078482836484909,
      "num_tokens": 77299663.0,
      "step": 38150
    },
    {
      "entropy": 6.051241874694824,
      "epoch": 3.334352879489644,
      "grad_norm": 1.390625,
      "learning_rate": 0.00039100403724573555,
      "loss": 5.2047,
      "mean_token_accuracy": 0.21051164865493774,
      "num_tokens": 77308755.0,
      "step": 38155
    },
    {
      "entropy": 5.990688753128052,
      "epoch": 3.334789827842349,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003909773358911441,
      "loss": 5.1868,
      "mean_token_accuracy": 0.20662893801927568,
      "num_tokens": 77318351.0,
      "step": 38160
    },
    {
      "entropy": 6.018022584915161,
      "epoch": 3.3352267761950536,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039095063231204657,
      "loss": 5.0976,
      "mean_token_accuracy": 0.21859404742717742,
      "num_tokens": 77328091.0,
      "step": 38165
    },
    {
      "entropy": 5.987682342529297,
      "epoch": 3.3356637245477585,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003909239265089552,
      "loss": 5.2168,
      "mean_token_accuracy": 0.2112663820385933,
      "num_tokens": 77338209.0,
      "step": 38170
    },
    {
      "entropy": 6.000923156738281,
      "epoch": 3.336100672900463,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039089721848238216,
      "loss": 5.174,
      "mean_token_accuracy": 0.21118707060813904,
      "num_tokens": 77349811.0,
      "step": 38175
    },
    {
      "entropy": 6.080311679840088,
      "epoch": 3.336537621253168,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003908705082328397,
      "loss": 5.311,
      "mean_token_accuracy": 0.19990878701210021,
      "num_tokens": 77359995.0,
      "step": 38180
    },
    {
      "entropy": 6.011247873306274,
      "epoch": 3.3369745696058724,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003908437957608403,
      "loss": 5.1762,
      "mean_token_accuracy": 0.2138134643435478,
      "num_tokens": 77371149.0,
      "step": 38185
    },
    {
      "entropy": 6.024988079071045,
      "epoch": 3.3374115179585773,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003908170810668961,
      "loss": 5.2354,
      "mean_token_accuracy": 0.20940835177898406,
      "num_tokens": 77381107.0,
      "step": 38190
    },
    {
      "entropy": 6.020918178558349,
      "epoch": 3.337848466311282,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039079036415151967,
      "loss": 5.3143,
      "mean_token_accuracy": 0.1980585753917694,
      "num_tokens": 77391982.0,
      "step": 38195
    },
    {
      "entropy": 5.98198676109314,
      "epoch": 3.3382854146639866,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003907636450152232,
      "loss": 5.2097,
      "mean_token_accuracy": 0.20875026285648346,
      "num_tokens": 77403129.0,
      "step": 38200
    },
    {
      "entropy": 6.043543434143066,
      "epoch": 3.3387223630166916,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003907369236585195,
      "loss": 5.293,
      "mean_token_accuracy": 0.21071566492319108,
      "num_tokens": 77412562.0,
      "step": 38205
    },
    {
      "entropy": 6.043312358856201,
      "epoch": 3.339159311369396,
      "grad_norm": 1.25,
      "learning_rate": 0.000390710200081921,
      "loss": 5.2481,
      "mean_token_accuracy": 0.20817480385303497,
      "num_tokens": 77422693.0,
      "step": 38210
    },
    {
      "entropy": 6.068756818771362,
      "epoch": 3.339596259722101,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003906834742859402,
      "loss": 5.2255,
      "mean_token_accuracy": 0.21032170504331588,
      "num_tokens": 77432455.0,
      "step": 38215
    },
    {
      "entropy": 6.029740476608277,
      "epoch": 3.3400332080748054,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003906567462710897,
      "loss": 5.2356,
      "mean_token_accuracy": 0.20646293312311173,
      "num_tokens": 77442785.0,
      "step": 38220
    },
    {
      "entropy": 6.0121245861053465,
      "epoch": 3.3404701564275103,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039063001603788217,
      "loss": 5.2246,
      "mean_token_accuracy": 0.20895712524652482,
      "num_tokens": 77453307.0,
      "step": 38225
    },
    {
      "entropy": 6.056859731674194,
      "epoch": 3.340907104780215,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003906032835868304,
      "loss": 5.3496,
      "mean_token_accuracy": 0.2039191707968712,
      "num_tokens": 77463851.0,
      "step": 38230
    },
    {
      "entropy": 6.073389196395874,
      "epoch": 3.3413440531329197,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039057654891844716,
      "loss": 5.2762,
      "mean_token_accuracy": 0.20140997618436812,
      "num_tokens": 77473391.0,
      "step": 38235
    },
    {
      "entropy": 6.08754711151123,
      "epoch": 3.341781001485624,
      "grad_norm": 1.375,
      "learning_rate": 0.000390549812033245,
      "loss": 5.2651,
      "mean_token_accuracy": 0.20614902824163436,
      "num_tokens": 77482750.0,
      "step": 38240
    },
    {
      "entropy": 6.00233826637268,
      "epoch": 3.342217949838329,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000390523072931737,
      "loss": 5.2165,
      "mean_token_accuracy": 0.21050355583429337,
      "num_tokens": 77492637.0,
      "step": 38245
    },
    {
      "entropy": 6.033261394500732,
      "epoch": 3.342654898191034,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003904963316144359,
      "loss": 5.3033,
      "mean_token_accuracy": 0.20158137530088424,
      "num_tokens": 77502877.0,
      "step": 38250
    },
    {
      "entropy": 6.003977203369141,
      "epoch": 3.3430918465437385,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039046958808185454,
      "loss": 5.231,
      "mean_token_accuracy": 0.2145196780562401,
      "num_tokens": 77513630.0,
      "step": 38255
    },
    {
      "entropy": 6.029619216918945,
      "epoch": 3.3435287948964434,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00039044284233450614,
      "loss": 5.1928,
      "mean_token_accuracy": 0.2156426876783371,
      "num_tokens": 77523801.0,
      "step": 38260
    },
    {
      "entropy": 6.000625419616699,
      "epoch": 3.343965743249148,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039041609437290346,
      "loss": 5.2352,
      "mean_token_accuracy": 0.20975444763898848,
      "num_tokens": 77533985.0,
      "step": 38265
    },
    {
      "entropy": 6.012223434448242,
      "epoch": 3.3444026916018528,
      "grad_norm": 1.25,
      "learning_rate": 0.0003903893441975596,
      "loss": 5.1671,
      "mean_token_accuracy": 0.22061680406332015,
      "num_tokens": 77543713.0,
      "step": 38270
    },
    {
      "entropy": 6.012998485565186,
      "epoch": 3.3448396399545572,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003903625918089877,
      "loss": 5.2474,
      "mean_token_accuracy": 0.20705081671476364,
      "num_tokens": 77553945.0,
      "step": 38275
    },
    {
      "entropy": 5.9676941394805905,
      "epoch": 3.345276588307262,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003903358372077008,
      "loss": 5.1836,
      "mean_token_accuracy": 0.21235837936401367,
      "num_tokens": 77563615.0,
      "step": 38280
    },
    {
      "entropy": 5.989978075027466,
      "epoch": 3.3457135366599666,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003903090803942121,
      "loss": 5.2118,
      "mean_token_accuracy": 0.2091172769665718,
      "num_tokens": 77573886.0,
      "step": 38285
    },
    {
      "entropy": 6.050515747070312,
      "epoch": 3.3461504850126715,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039028232136903483,
      "loss": 5.2701,
      "mean_token_accuracy": 0.19977811723947525,
      "num_tokens": 77585017.0,
      "step": 38290
    },
    {
      "entropy": 6.039867496490478,
      "epoch": 3.346587433365376,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039025556013268226,
      "loss": 5.2225,
      "mean_token_accuracy": 0.21179085820913315,
      "num_tokens": 77594605.0,
      "step": 38295
    },
    {
      "entropy": 6.066770458221436,
      "epoch": 3.347024381718081,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039022879668566764,
      "loss": 5.1732,
      "mean_token_accuracy": 0.21016707569360732,
      "num_tokens": 77603262.0,
      "step": 38300
    },
    {
      "entropy": 5.98300724029541,
      "epoch": 3.347461330070786,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003902020310285043,
      "loss": 5.2003,
      "mean_token_accuracy": 0.21008844077587127,
      "num_tokens": 77612963.0,
      "step": 38305
    },
    {
      "entropy": 6.0115478515625,
      "epoch": 3.3478982784234903,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003901752631617056,
      "loss": 5.2279,
      "mean_token_accuracy": 0.20717607587575912,
      "num_tokens": 77623237.0,
      "step": 38310
    },
    {
      "entropy": 6.0077918529510494,
      "epoch": 3.348335226776195,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003901484930857851,
      "loss": 5.1935,
      "mean_token_accuracy": 0.20605577677488326,
      "num_tokens": 77632513.0,
      "step": 38315
    },
    {
      "entropy": 5.983092498779297,
      "epoch": 3.3487721751288997,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003901217208012561,
      "loss": 5.1688,
      "mean_token_accuracy": 0.21680374294519425,
      "num_tokens": 77643160.0,
      "step": 38320
    },
    {
      "entropy": 6.046661806106568,
      "epoch": 3.3492091234816046,
      "grad_norm": 1.328125,
      "learning_rate": 0.00039009494630863216,
      "loss": 5.236,
      "mean_token_accuracy": 0.2107217639684677,
      "num_tokens": 77653311.0,
      "step": 38325
    },
    {
      "entropy": 6.008177471160889,
      "epoch": 3.349646071834309,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039006816960842695,
      "loss": 5.2275,
      "mean_token_accuracy": 0.2097381204366684,
      "num_tokens": 77663142.0,
      "step": 38330
    },
    {
      "entropy": 6.011527967453003,
      "epoch": 3.350083020187014,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003900413907011538,
      "loss": 5.2681,
      "mean_token_accuracy": 0.20665521323680877,
      "num_tokens": 77675143.0,
      "step": 38335
    },
    {
      "entropy": 6.027296161651611,
      "epoch": 3.3505199685397185,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003900146095873266,
      "loss": 5.2292,
      "mean_token_accuracy": 0.21258078813552855,
      "num_tokens": 77685464.0,
      "step": 38340
    },
    {
      "entropy": 6.101980495452881,
      "epoch": 3.3509569168924234,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003899878262674588,
      "loss": 5.24,
      "mean_token_accuracy": 0.20841536819934844,
      "num_tokens": 77695749.0,
      "step": 38345
    },
    {
      "entropy": 6.029736518859863,
      "epoch": 3.351393865245128,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003899610407420643,
      "loss": 5.1995,
      "mean_token_accuracy": 0.20650251358747482,
      "num_tokens": 77705678.0,
      "step": 38350
    },
    {
      "entropy": 6.031748056411743,
      "epoch": 3.3518308135978327,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003899342530116569,
      "loss": 5.2442,
      "mean_token_accuracy": 0.20290901511907578,
      "num_tokens": 77716481.0,
      "step": 38355
    },
    {
      "entropy": 6.027977085113525,
      "epoch": 3.3522677619505377,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003899074630767502,
      "loss": 5.269,
      "mean_token_accuracy": 0.2051917478442192,
      "num_tokens": 77726977.0,
      "step": 38360
    },
    {
      "entropy": 5.982347059249878,
      "epoch": 3.352704710303242,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038988067093785817,
      "loss": 5.1778,
      "mean_token_accuracy": 0.2102099433541298,
      "num_tokens": 77736624.0,
      "step": 38365
    },
    {
      "entropy": 6.0462688446044925,
      "epoch": 3.353141658655947,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038985387659549466,
      "loss": 5.2166,
      "mean_token_accuracy": 0.21951236426830292,
      "num_tokens": 77745964.0,
      "step": 38370
    },
    {
      "entropy": 6.056820917129516,
      "epoch": 3.3535786070086515,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038982708005017366,
      "loss": 5.2355,
      "mean_token_accuracy": 0.20885560065507888,
      "num_tokens": 77755801.0,
      "step": 38375
    },
    {
      "entropy": 6.044745588302613,
      "epoch": 3.3540155553613564,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003898002813024091,
      "loss": 5.2781,
      "mean_token_accuracy": 0.1992132008075714,
      "num_tokens": 77765604.0,
      "step": 38380
    },
    {
      "entropy": 6.0805741310119625,
      "epoch": 3.354452503714061,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003897734803527149,
      "loss": 5.2788,
      "mean_token_accuracy": 0.20559779703617095,
      "num_tokens": 77775754.0,
      "step": 38385
    },
    {
      "entropy": 6.100669717788696,
      "epoch": 3.354889452066766,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038974667720160534,
      "loss": 5.3873,
      "mean_token_accuracy": 0.20147249847650528,
      "num_tokens": 77786063.0,
      "step": 38390
    },
    {
      "entropy": 5.985177755355835,
      "epoch": 3.3553264004194703,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038971987184959427,
      "loss": 5.1799,
      "mean_token_accuracy": 0.21442999839782714,
      "num_tokens": 77795217.0,
      "step": 38395
    },
    {
      "entropy": 5.996312665939331,
      "epoch": 3.355763348772175,
      "grad_norm": 1.296875,
      "learning_rate": 0.000389693064297196,
      "loss": 5.2748,
      "mean_token_accuracy": 0.20577509254217147,
      "num_tokens": 77804637.0,
      "step": 38400
    },
    {
      "entropy": 6.0297950267791744,
      "epoch": 3.3562002971248797,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003896662545449247,
      "loss": 5.26,
      "mean_token_accuracy": 0.207174414396286,
      "num_tokens": 77816048.0,
      "step": 38405
    },
    {
      "entropy": 6.046404790878296,
      "epoch": 3.3566372454775846,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003896394425932944,
      "loss": 5.1735,
      "mean_token_accuracy": 0.21424249112606047,
      "num_tokens": 77826385.0,
      "step": 38410
    },
    {
      "entropy": 6.020476007461548,
      "epoch": 3.3570741938302895,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00038961262844281966,
      "loss": 5.2687,
      "mean_token_accuracy": 0.20445608347654343,
      "num_tokens": 77835744.0,
      "step": 38415
    },
    {
      "entropy": 6.043231201171875,
      "epoch": 3.357511142182994,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003895858120940146,
      "loss": 5.2525,
      "mean_token_accuracy": 0.21136813312768937,
      "num_tokens": 77845883.0,
      "step": 38420
    },
    {
      "entropy": 6.022723865509033,
      "epoch": 3.357948090535699,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038955899354739364,
      "loss": 5.2192,
      "mean_token_accuracy": 0.20330884754657746,
      "num_tokens": 77856213.0,
      "step": 38425
    },
    {
      "entropy": 6.058835220336914,
      "epoch": 3.3583850388884033,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038953217280347113,
      "loss": 5.2972,
      "mean_token_accuracy": 0.20434581190347673,
      "num_tokens": 77867266.0,
      "step": 38430
    },
    {
      "entropy": 6.085478782653809,
      "epoch": 3.3588219872411083,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003895053498627616,
      "loss": 5.3187,
      "mean_token_accuracy": 0.21090105921030045,
      "num_tokens": 77878358.0,
      "step": 38435
    },
    {
      "entropy": 6.086871862411499,
      "epoch": 3.3592589355938127,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003894785247257794,
      "loss": 5.3153,
      "mean_token_accuracy": 0.20540694892406464,
      "num_tokens": 77889191.0,
      "step": 38440
    },
    {
      "entropy": 6.055378103256226,
      "epoch": 3.3596958839465176,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003894516973930392,
      "loss": 5.2719,
      "mean_token_accuracy": 0.2074907138943672,
      "num_tokens": 77898937.0,
      "step": 38445
    },
    {
      "entropy": 6.011759233474732,
      "epoch": 3.360132832299222,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003894248678650553,
      "loss": 5.1871,
      "mean_token_accuracy": 0.2118406280875206,
      "num_tokens": 77907440.0,
      "step": 38450
    },
    {
      "entropy": 6.00727014541626,
      "epoch": 3.360569780651927,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003893980361423426,
      "loss": 5.2363,
      "mean_token_accuracy": 0.21090921014547348,
      "num_tokens": 77917471.0,
      "step": 38455
    },
    {
      "entropy": 6.077342796325683,
      "epoch": 3.3610067290046315,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003893712022254156,
      "loss": 5.1792,
      "mean_token_accuracy": 0.21293700188398362,
      "num_tokens": 77927292.0,
      "step": 38460
    },
    {
      "entropy": 6.05226411819458,
      "epoch": 3.3614436773573364,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00038934436611478895,
      "loss": 5.2993,
      "mean_token_accuracy": 0.20906385332345961,
      "num_tokens": 77937870.0,
      "step": 38465
    },
    {
      "entropy": 6.058169317245484,
      "epoch": 3.3618806257100413,
      "grad_norm": 1.359375,
      "learning_rate": 0.00038931752781097757,
      "loss": 5.215,
      "mean_token_accuracy": 0.20844455510377885,
      "num_tokens": 77947369.0,
      "step": 38470
    },
    {
      "entropy": 6.057256412506104,
      "epoch": 3.362317574062746,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038929068731449604,
      "loss": 5.2923,
      "mean_token_accuracy": 0.20610660165548325,
      "num_tokens": 77958314.0,
      "step": 38475
    },
    {
      "entropy": 6.124323940277099,
      "epoch": 3.3627545224154503,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003892638446258592,
      "loss": 5.3421,
      "mean_token_accuracy": 0.1987604945898056,
      "num_tokens": 77968734.0,
      "step": 38480
    },
    {
      "entropy": 6.0692342758178714,
      "epoch": 3.363191470768155,
      "grad_norm": 1.265625,
      "learning_rate": 0.000389236999745582,
      "loss": 5.2547,
      "mean_token_accuracy": 0.20837627053260804,
      "num_tokens": 77978964.0,
      "step": 38485
    },
    {
      "entropy": 5.987607145309449,
      "epoch": 3.36362841912086,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003892101526741792,
      "loss": 5.1945,
      "mean_token_accuracy": 0.21378018260002135,
      "num_tokens": 77989530.0,
      "step": 38490
    },
    {
      "entropy": 6.028872919082642,
      "epoch": 3.3640653674735645,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000389183303412166,
      "loss": 5.2227,
      "mean_token_accuracy": 0.21043815314769745,
      "num_tokens": 77999870.0,
      "step": 38495
    },
    {
      "entropy": 6.050806522369385,
      "epoch": 3.3645023158262695,
      "grad_norm": 1.375,
      "learning_rate": 0.00038915645196005704,
      "loss": 5.2407,
      "mean_token_accuracy": 0.2089338019490242,
      "num_tokens": 78010047.0,
      "step": 38500
    },
    {
      "entropy": 6.057279634475708,
      "epoch": 3.364939264178974,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003891295983183676,
      "loss": 5.1735,
      "mean_token_accuracy": 0.2088167831301689,
      "num_tokens": 78019834.0,
      "step": 38505
    },
    {
      "entropy": 6.088221836090088,
      "epoch": 3.365376212531679,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038910274248761263,
      "loss": 5.2251,
      "mean_token_accuracy": 0.20211583524942398,
      "num_tokens": 78029639.0,
      "step": 38510
    },
    {
      "entropy": 6.05239691734314,
      "epoch": 3.3658131608843833,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003890758844683072,
      "loss": 5.2062,
      "mean_token_accuracy": 0.2110273689031601,
      "num_tokens": 78039638.0,
      "step": 38515
    },
    {
      "entropy": 5.979272747039795,
      "epoch": 3.3662501092370882,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038904902426096664,
      "loss": 5.2178,
      "mean_token_accuracy": 0.21079452484846115,
      "num_tokens": 78050666.0,
      "step": 38520
    },
    {
      "entropy": 6.001174020767212,
      "epoch": 3.3666870575897927,
      "grad_norm": 1.375,
      "learning_rate": 0.000389022161866106,
      "loss": 5.1301,
      "mean_token_accuracy": 0.22157542556524276,
      "num_tokens": 78060292.0,
      "step": 38525
    },
    {
      "entropy": 6.034449529647827,
      "epoch": 3.3671240059424976,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038899529728424057,
      "loss": 5.2555,
      "mean_token_accuracy": 0.20424299240112304,
      "num_tokens": 78069900.0,
      "step": 38530
    },
    {
      "entropy": 6.00076813697815,
      "epoch": 3.367560954295202,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003889684305158856,
      "loss": 5.1977,
      "mean_token_accuracy": 0.20705907344818114,
      "num_tokens": 78078788.0,
      "step": 38535
    },
    {
      "entropy": 6.030644130706787,
      "epoch": 3.367997902647907,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003889415615615564,
      "loss": 5.2451,
      "mean_token_accuracy": 0.2055851474404335,
      "num_tokens": 78087612.0,
      "step": 38540
    },
    {
      "entropy": 6.039441013336182,
      "epoch": 3.368434851000612,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038891469042176834,
      "loss": 5.2006,
      "mean_token_accuracy": 0.21161784678697587,
      "num_tokens": 78098399.0,
      "step": 38545
    },
    {
      "entropy": 6.095770788192749,
      "epoch": 3.3688717993533164,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003888878170970369,
      "loss": 5.2478,
      "mean_token_accuracy": 0.20111464709043503,
      "num_tokens": 78107956.0,
      "step": 38550
    },
    {
      "entropy": 6.047022485733033,
      "epoch": 3.3693087477060213,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038886094158787726,
      "loss": 5.2767,
      "mean_token_accuracy": 0.20399336665868759,
      "num_tokens": 78119613.0,
      "step": 38555
    },
    {
      "entropy": 6.030018138885498,
      "epoch": 3.3697456960587258,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003888340638948052,
      "loss": 5.2454,
      "mean_token_accuracy": 0.21230384856462478,
      "num_tokens": 78128986.0,
      "step": 38560
    },
    {
      "entropy": 6.053660678863525,
      "epoch": 3.3701826444114307,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038880718401833623,
      "loss": 5.2517,
      "mean_token_accuracy": 0.20866907238960267,
      "num_tokens": 78139085.0,
      "step": 38565
    },
    {
      "entropy": 5.990028953552246,
      "epoch": 3.370619592764135,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003887803019589858,
      "loss": 5.1431,
      "mean_token_accuracy": 0.21738651543855667,
      "num_tokens": 78148780.0,
      "step": 38570
    },
    {
      "entropy": 6.011289453506469,
      "epoch": 3.37105654111684,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003887534177172695,
      "loss": 5.2367,
      "mean_token_accuracy": 0.21478302031755447,
      "num_tokens": 78159021.0,
      "step": 38575
    },
    {
      "entropy": 6.061536598205566,
      "epoch": 3.3714934894695445,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038872653129370303,
      "loss": 5.2411,
      "mean_token_accuracy": 0.2105087772011757,
      "num_tokens": 78169553.0,
      "step": 38580
    },
    {
      "entropy": 5.965493965148926,
      "epoch": 3.3719304378222494,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038869964268880224,
      "loss": 5.141,
      "mean_token_accuracy": 0.21793848425149917,
      "num_tokens": 78179631.0,
      "step": 38585
    },
    {
      "entropy": 6.0446288108825685,
      "epoch": 3.372367386174954,
      "grad_norm": 1.375,
      "learning_rate": 0.0003886727519030826,
      "loss": 5.1949,
      "mean_token_accuracy": 0.20759467035531998,
      "num_tokens": 78189497.0,
      "step": 38590
    },
    {
      "entropy": 5.959914684295654,
      "epoch": 3.372804334527659,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038864585893706006,
      "loss": 5.2148,
      "mean_token_accuracy": 0.21234899014234543,
      "num_tokens": 78200786.0,
      "step": 38595
    },
    {
      "entropy": 6.054821348190307,
      "epoch": 3.3732412828803637,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00038861896379125034,
      "loss": 5.2103,
      "mean_token_accuracy": 0.20987822264432907,
      "num_tokens": 78209783.0,
      "step": 38600
    },
    {
      "entropy": 5.996183776855469,
      "epoch": 3.373678231233068,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003885920664661695,
      "loss": 5.2075,
      "mean_token_accuracy": 0.21553042978048326,
      "num_tokens": 78218988.0,
      "step": 38605
    },
    {
      "entropy": 6.014906406402588,
      "epoch": 3.374115179585773,
      "grad_norm": 1.25,
      "learning_rate": 0.0003885651669623331,
      "loss": 5.2291,
      "mean_token_accuracy": 0.20503934621810913,
      "num_tokens": 78228545.0,
      "step": 38610
    },
    {
      "entropy": 6.004794073104859,
      "epoch": 3.3745521279384776,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003885382652802574,
      "loss": 5.1597,
      "mean_token_accuracy": 0.21385621875524521,
      "num_tokens": 78238167.0,
      "step": 38615
    },
    {
      "entropy": 5.998469972610474,
      "epoch": 3.3749890762911825,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003885113614204583,
      "loss": 5.1724,
      "mean_token_accuracy": 0.21864838600158693,
      "num_tokens": 78248257.0,
      "step": 38620
    },
    {
      "entropy": 6.061773633956909,
      "epoch": 3.375426024643887,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038848445538345173,
      "loss": 5.2586,
      "mean_token_accuracy": 0.20679101645946502,
      "num_tokens": 78257406.0,
      "step": 38625
    },
    {
      "entropy": 5.972565603256226,
      "epoch": 3.375862972996592,
      "grad_norm": 1.25,
      "learning_rate": 0.0003884575471697539,
      "loss": 5.1179,
      "mean_token_accuracy": 0.22044302970170976,
      "num_tokens": 78267022.0,
      "step": 38630
    },
    {
      "entropy": 6.00862603187561,
      "epoch": 3.3762999213492964,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038843063677988093,
      "loss": 5.255,
      "mean_token_accuracy": 0.20846527367830275,
      "num_tokens": 78277584.0,
      "step": 38635
    },
    {
      "entropy": 6.038939714431763,
      "epoch": 3.3767368697020013,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003884037242143488,
      "loss": 5.2118,
      "mean_token_accuracy": 0.20938608646392823,
      "num_tokens": 78288218.0,
      "step": 38640
    },
    {
      "entropy": 6.038583612442016,
      "epoch": 3.3771738180547057,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003883768094736738,
      "loss": 5.2235,
      "mean_token_accuracy": 0.21012000739574432,
      "num_tokens": 78297585.0,
      "step": 38645
    },
    {
      "entropy": 5.969464492797852,
      "epoch": 3.3776107664074106,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003883498925583723,
      "loss": 5.2026,
      "mean_token_accuracy": 0.2111121565103531,
      "num_tokens": 78308406.0,
      "step": 38650
    },
    {
      "entropy": 6.027083492279052,
      "epoch": 3.3780477147601156,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003883229734689604,
      "loss": 5.2139,
      "mean_token_accuracy": 0.20958995074033737,
      "num_tokens": 78318151.0,
      "step": 38655
    },
    {
      "entropy": 6.049335384368897,
      "epoch": 3.37848466311282,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003882960522059545,
      "loss": 5.2018,
      "mean_token_accuracy": 0.21164796203374864,
      "num_tokens": 78327316.0,
      "step": 38660
    },
    {
      "entropy": 5.998465728759766,
      "epoch": 3.378921611465525,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003882691287698709,
      "loss": 5.1939,
      "mean_token_accuracy": 0.21306172758340836,
      "num_tokens": 78337564.0,
      "step": 38665
    },
    {
      "entropy": 6.0370477676391605,
      "epoch": 3.3793585598182294,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003882422031612261,
      "loss": 5.2388,
      "mean_token_accuracy": 0.20682044476270675,
      "num_tokens": 78347796.0,
      "step": 38670
    },
    {
      "entropy": 6.107105112075805,
      "epoch": 3.3797955081709343,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003882152753805366,
      "loss": 5.28,
      "mean_token_accuracy": 0.19929276555776596,
      "num_tokens": 78357269.0,
      "step": 38675
    },
    {
      "entropy": 6.037993049621582,
      "epoch": 3.380232456523639,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038818834542831863,
      "loss": 5.1939,
      "mean_token_accuracy": 0.21519383192062377,
      "num_tokens": 78368001.0,
      "step": 38680
    },
    {
      "entropy": 5.957801055908203,
      "epoch": 3.3806694048763437,
      "grad_norm": 1.296875,
      "learning_rate": 0.000388161413305089,
      "loss": 5.1382,
      "mean_token_accuracy": 0.2173548549413681,
      "num_tokens": 78377908.0,
      "step": 38685
    },
    {
      "entropy": 6.025057077407837,
      "epoch": 3.381106353229048,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003881344790113641,
      "loss": 5.2962,
      "mean_token_accuracy": 0.2056276500225067,
      "num_tokens": 78389450.0,
      "step": 38690
    },
    {
      "entropy": 6.076132488250733,
      "epoch": 3.381543301581753,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038810754254766066,
      "loss": 5.2322,
      "mean_token_accuracy": 0.20991435199975966,
      "num_tokens": 78400397.0,
      "step": 38695
    },
    {
      "entropy": 6.027725601196289,
      "epoch": 3.3819802499344576,
      "grad_norm": 1.25,
      "learning_rate": 0.00038808060391449525,
      "loss": 5.1686,
      "mean_token_accuracy": 0.2141094520688057,
      "num_tokens": 78410322.0,
      "step": 38700
    },
    {
      "entropy": 6.0284172058105465,
      "epoch": 3.3824171982871625,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003880536631123847,
      "loss": 5.2576,
      "mean_token_accuracy": 0.19995759874582292,
      "num_tokens": 78420816.0,
      "step": 38705
    },
    {
      "entropy": 5.999194765090943,
      "epoch": 3.3828541466398674,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003880267201418455,
      "loss": 5.2784,
      "mean_token_accuracy": 0.20553128570318221,
      "num_tokens": 78430217.0,
      "step": 38710
    },
    {
      "entropy": 6.059186410903931,
      "epoch": 3.383291094992572,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003879997750033946,
      "loss": 5.2246,
      "mean_token_accuracy": 0.20974305719137193,
      "num_tokens": 78439938.0,
      "step": 38715
    },
    {
      "entropy": 6.070408296585083,
      "epoch": 3.3837280433452768,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003879728276975489,
      "loss": 5.2239,
      "mean_token_accuracy": 0.2036316514015198,
      "num_tokens": 78450805.0,
      "step": 38720
    },
    {
      "entropy": 5.99761734008789,
      "epoch": 3.3841649916979812,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038794587822482506,
      "loss": 5.2422,
      "mean_token_accuracy": 0.2076819732785225,
      "num_tokens": 78460653.0,
      "step": 38725
    },
    {
      "entropy": 6.006157398223877,
      "epoch": 3.384601940050686,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038791892658574016,
      "loss": 5.1842,
      "mean_token_accuracy": 0.21061374694108964,
      "num_tokens": 78471457.0,
      "step": 38730
    },
    {
      "entropy": 6.078844785690308,
      "epoch": 3.3850388884033906,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003878919727808111,
      "loss": 5.2563,
      "mean_token_accuracy": 0.20498303771018983,
      "num_tokens": 78480834.0,
      "step": 38735
    },
    {
      "entropy": 6.036605453491211,
      "epoch": 3.3854758367560955,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038786501681055483,
      "loss": 5.2333,
      "mean_token_accuracy": 0.21134153604507447,
      "num_tokens": 78491509.0,
      "step": 38740
    },
    {
      "entropy": 6.0759368419647215,
      "epoch": 3.3859127851088,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003878380586754884,
      "loss": 5.3176,
      "mean_token_accuracy": 0.2028556153178215,
      "num_tokens": 78502137.0,
      "step": 38745
    },
    {
      "entropy": 6.000303411483765,
      "epoch": 3.386349733461505,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038781109837612886,
      "loss": 5.1315,
      "mean_token_accuracy": 0.217672398686409,
      "num_tokens": 78511888.0,
      "step": 38750
    },
    {
      "entropy": 6.027304363250733,
      "epoch": 3.3867866818142094,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003877841359129934,
      "loss": 5.2411,
      "mean_token_accuracy": 0.2039148449897766,
      "num_tokens": 78521553.0,
      "step": 38755
    },
    {
      "entropy": 5.962284517288208,
      "epoch": 3.3872236301669143,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000387757171286599,
      "loss": 5.1335,
      "mean_token_accuracy": 0.21793924272060394,
      "num_tokens": 78531290.0,
      "step": 38760
    },
    {
      "entropy": 5.988309717178344,
      "epoch": 3.387660578519619,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003877302044974631,
      "loss": 5.1985,
      "mean_token_accuracy": 0.21366367787122725,
      "num_tokens": 78541200.0,
      "step": 38765
    },
    {
      "entropy": 6.003887701034546,
      "epoch": 3.3880975268723237,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003877032355461027,
      "loss": 5.2101,
      "mean_token_accuracy": 0.215543033182621,
      "num_tokens": 78551075.0,
      "step": 38770
    },
    {
      "entropy": 6.010976886749267,
      "epoch": 3.388534475225028,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038767626443303525,
      "loss": 5.2499,
      "mean_token_accuracy": 0.21463604569435119,
      "num_tokens": 78560935.0,
      "step": 38775
    },
    {
      "entropy": 6.031137275695801,
      "epoch": 3.388971423577733,
      "grad_norm": 1.328125,
      "learning_rate": 0.000387649291158778,
      "loss": 5.2063,
      "mean_token_accuracy": 0.21092337369918823,
      "num_tokens": 78570856.0,
      "step": 38780
    },
    {
      "entropy": 6.059831190109253,
      "epoch": 3.389408371930438,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038762231572384833,
      "loss": 5.297,
      "mean_token_accuracy": 0.20075175762176514,
      "num_tokens": 78581128.0,
      "step": 38785
    },
    {
      "entropy": 6.0414763450622555,
      "epoch": 3.3898453202831424,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003875953381287636,
      "loss": 5.2119,
      "mean_token_accuracy": 0.2075547084212303,
      "num_tokens": 78591131.0,
      "step": 38790
    },
    {
      "entropy": 6.07269868850708,
      "epoch": 3.3902822686358474,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003875683583740413,
      "loss": 5.2823,
      "mean_token_accuracy": 0.21283914595842363,
      "num_tokens": 78600470.0,
      "step": 38795
    },
    {
      "entropy": 6.032742309570312,
      "epoch": 3.390719216988552,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003875413764601989,
      "loss": 5.2537,
      "mean_token_accuracy": 0.2062015414237976,
      "num_tokens": 78610916.0,
      "step": 38800
    },
    {
      "entropy": 6.021217966079712,
      "epoch": 3.3911561653412567,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003875143923877539,
      "loss": 5.1688,
      "mean_token_accuracy": 0.20966051518917084,
      "num_tokens": 78620199.0,
      "step": 38805
    },
    {
      "entropy": 6.077747774124146,
      "epoch": 3.391593113693961,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00038748740615722395,
      "loss": 5.2531,
      "mean_token_accuracy": 0.21781585216522217,
      "num_tokens": 78629186.0,
      "step": 38810
    },
    {
      "entropy": 6.026320695877075,
      "epoch": 3.392030062046666,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00038746041776912654,
      "loss": 5.21,
      "mean_token_accuracy": 0.20359057635068895,
      "num_tokens": 78638786.0,
      "step": 38815
    },
    {
      "entropy": 6.091786098480225,
      "epoch": 3.392467010399371,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038743342722397946,
      "loss": 5.2912,
      "mean_token_accuracy": 0.20568366050720216,
      "num_tokens": 78647904.0,
      "step": 38820
    },
    {
      "entropy": 6.024830722808838,
      "epoch": 3.3929039587520755,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003874064345223003,
      "loss": 5.2373,
      "mean_token_accuracy": 0.20229964107275009,
      "num_tokens": 78657765.0,
      "step": 38825
    },
    {
      "entropy": 6.023506116867066,
      "epoch": 3.39334090710478,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038737943966460684,
      "loss": 5.2514,
      "mean_token_accuracy": 0.20241250693798066,
      "num_tokens": 78667871.0,
      "step": 38830
    },
    {
      "entropy": 5.958521556854248,
      "epoch": 3.393777855457485,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038735244265141676,
      "loss": 5.1634,
      "mean_token_accuracy": 0.21724364459514617,
      "num_tokens": 78677824.0,
      "step": 38835
    },
    {
      "entropy": 6.065642404556274,
      "epoch": 3.39421480381019,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003873254434832481,
      "loss": 5.2894,
      "mean_token_accuracy": 0.20448481440544128,
      "num_tokens": 78687498.0,
      "step": 38840
    },
    {
      "entropy": 6.05192003250122,
      "epoch": 3.3946517521628943,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038729844216061845,
      "loss": 5.2767,
      "mean_token_accuracy": 0.20793891400098802,
      "num_tokens": 78698645.0,
      "step": 38845
    },
    {
      "entropy": 6.010726881027222,
      "epoch": 3.395088700515599,
      "grad_norm": 1.421875,
      "learning_rate": 0.00038727143868404597,
      "loss": 5.1088,
      "mean_token_accuracy": 0.21921777874231338,
      "num_tokens": 78707796.0,
      "step": 38850
    },
    {
      "entropy": 5.978696870803833,
      "epoch": 3.3955256488683037,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003872444330540483,
      "loss": 5.2249,
      "mean_token_accuracy": 0.21127717941999435,
      "num_tokens": 78717838.0,
      "step": 38855
    },
    {
      "entropy": 6.064271259307861,
      "epoch": 3.3959625972210086,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003872174252711437,
      "loss": 5.2294,
      "mean_token_accuracy": 0.2072984054684639,
      "num_tokens": 78727318.0,
      "step": 38860
    },
    {
      "entropy": 6.10771484375,
      "epoch": 3.396399545573713,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038719041533585,
      "loss": 5.2647,
      "mean_token_accuracy": 0.2038218855857849,
      "num_tokens": 78737446.0,
      "step": 38865
    },
    {
      "entropy": 6.049086570739746,
      "epoch": 3.396836493926418,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003871634032486854,
      "loss": 5.2897,
      "mean_token_accuracy": 0.2029823437333107,
      "num_tokens": 78748557.0,
      "step": 38870
    },
    {
      "entropy": 6.109214353561401,
      "epoch": 3.3972734422791224,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000387136389010168,
      "loss": 5.3776,
      "mean_token_accuracy": 0.19757914990186692,
      "num_tokens": 78759482.0,
      "step": 38875
    },
    {
      "entropy": 6.000452756881714,
      "epoch": 3.3977103906318273,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038710937262081583,
      "loss": 5.1264,
      "mean_token_accuracy": 0.2194804936647415,
      "num_tokens": 78768959.0,
      "step": 38880
    },
    {
      "entropy": 5.961813163757324,
      "epoch": 3.398147338984532,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003870823540811472,
      "loss": 5.1761,
      "mean_token_accuracy": 0.21990539878606796,
      "num_tokens": 78779631.0,
      "step": 38885
    },
    {
      "entropy": 6.03257122039795,
      "epoch": 3.3985842873372367,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038705533339168023,
      "loss": 5.2335,
      "mean_token_accuracy": 0.21195680648088455,
      "num_tokens": 78789291.0,
      "step": 38890
    },
    {
      "entropy": 6.022537899017334,
      "epoch": 3.3990212356899416,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003870283105529333,
      "loss": 5.2715,
      "mean_token_accuracy": 0.2067156508564949,
      "num_tokens": 78799452.0,
      "step": 38895
    },
    {
      "entropy": 6.059751224517822,
      "epoch": 3.399458184042646,
      "grad_norm": 1.34375,
      "learning_rate": 0.00038700128556542465,
      "loss": 5.2409,
      "mean_token_accuracy": 0.20489594340324402,
      "num_tokens": 78809390.0,
      "step": 38900
    },
    {
      "entropy": 6.033454418182373,
      "epoch": 3.399895132395351,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003869742584296727,
      "loss": 5.1931,
      "mean_token_accuracy": 0.21438509970903397,
      "num_tokens": 78818979.0,
      "step": 38905
    },
    {
      "entropy": 6.040906143188477,
      "epoch": 3.4003320807480555,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003869472291461957,
      "loss": 5.2911,
      "mean_token_accuracy": 0.20590523779392242,
      "num_tokens": 78829773.0,
      "step": 38910
    },
    {
      "entropy": 6.048760509490966,
      "epoch": 3.4007690291007604,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038692019771551227,
      "loss": 5.2327,
      "mean_token_accuracy": 0.21065158843994142,
      "num_tokens": 78839664.0,
      "step": 38915
    },
    {
      "entropy": 5.98425440788269,
      "epoch": 3.401205977453465,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003868931641381408,
      "loss": 5.1963,
      "mean_token_accuracy": 0.21712955385446547,
      "num_tokens": 78849941.0,
      "step": 38920
    },
    {
      "entropy": 6.077495145797729,
      "epoch": 3.40164292580617,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038686612841459973,
      "loss": 5.2665,
      "mean_token_accuracy": 0.2066656768321991,
      "num_tokens": 78860709.0,
      "step": 38925
    },
    {
      "entropy": 6.059165620803833,
      "epoch": 3.4020798741588743,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003868390905454078,
      "loss": 5.2366,
      "mean_token_accuracy": 0.206769360601902,
      "num_tokens": 78871001.0,
      "step": 38930
    },
    {
      "entropy": 6.012314939498902,
      "epoch": 3.402516822511579,
      "grad_norm": 1.25,
      "learning_rate": 0.0003868120505310835,
      "loss": 5.2675,
      "mean_token_accuracy": 0.2137146472930908,
      "num_tokens": 78881033.0,
      "step": 38935
    },
    {
      "entropy": 5.98866376876831,
      "epoch": 3.4029537708642836,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038678500837214543,
      "loss": 5.2201,
      "mean_token_accuracy": 0.21569233685731887,
      "num_tokens": 78890078.0,
      "step": 38940
    },
    {
      "entropy": 6.050012588500977,
      "epoch": 3.4033907192169885,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038675796406911235,
      "loss": 5.2505,
      "mean_token_accuracy": 0.20833933651447295,
      "num_tokens": 78900209.0,
      "step": 38945
    },
    {
      "entropy": 6.057488107681275,
      "epoch": 3.4038276675696935,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038673091762250295,
      "loss": 5.2445,
      "mean_token_accuracy": 0.2128039062023163,
      "num_tokens": 78911431.0,
      "step": 38950
    },
    {
      "entropy": 6.060739326477051,
      "epoch": 3.404264615922398,
      "grad_norm": 1.296875,
      "learning_rate": 0.000386703869032836,
      "loss": 5.2796,
      "mean_token_accuracy": 0.2097296580672264,
      "num_tokens": 78922516.0,
      "step": 38955
    },
    {
      "entropy": 5.95376353263855,
      "epoch": 3.404701564275103,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0003866768183006303,
      "loss": 5.0895,
      "mean_token_accuracy": 0.22462569326162338,
      "num_tokens": 78933186.0,
      "step": 38960
    },
    {
      "entropy": 5.945842313766479,
      "epoch": 3.4051385126278073,
      "grad_norm": 1.25,
      "learning_rate": 0.0003866497654264048,
      "loss": 5.0988,
      "mean_token_accuracy": 0.22486559003591539,
      "num_tokens": 78944602.0,
      "step": 38965
    },
    {
      "entropy": 5.960024309158325,
      "epoch": 3.4055754609805122,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003866227104106782,
      "loss": 5.1793,
      "mean_token_accuracy": 0.21640175879001616,
      "num_tokens": 78954336.0,
      "step": 38970
    },
    {
      "entropy": 6.022485733032227,
      "epoch": 3.4060124093332167,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003865956532539695,
      "loss": 5.2437,
      "mean_token_accuracy": 0.21106385439634323,
      "num_tokens": 78965306.0,
      "step": 38975
    },
    {
      "entropy": 6.03548412322998,
      "epoch": 3.4064493576859216,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003865685939567977,
      "loss": 5.2376,
      "mean_token_accuracy": 0.20888720452785492,
      "num_tokens": 78975528.0,
      "step": 38980
    },
    {
      "entropy": 6.132811546325684,
      "epoch": 3.406886306038626,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003865415325196819,
      "loss": 5.3264,
      "mean_token_accuracy": 0.20083773583173753,
      "num_tokens": 78985402.0,
      "step": 38985
    },
    {
      "entropy": 6.033141183853149,
      "epoch": 3.407323254391331,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038651446894314107,
      "loss": 5.2499,
      "mean_token_accuracy": 0.20954209119081496,
      "num_tokens": 78996053.0,
      "step": 38990
    },
    {
      "entropy": 6.039280652999878,
      "epoch": 3.4077602027440355,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003864874032276942,
      "loss": 5.2707,
      "mean_token_accuracy": 0.20728445500135423,
      "num_tokens": 79007645.0,
      "step": 38995
    },
    {
      "entropy": 6.033050918579102,
      "epoch": 3.4081971510967404,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003864603353738605,
      "loss": 5.1722,
      "mean_token_accuracy": 0.21494061648845672,
      "num_tokens": 79018649.0,
      "step": 39000
    },
    {
      "epoch": 3.4081971510967404,
      "eval_entropy": 5.775706069526648,
      "eval_loss": 5.278534412384033,
      "eval_mean_token_accuracy": 0.21622458056198332,
      "eval_num_tokens": 79018649.0,
      "eval_runtime": 21.558,
      "eval_samples_per_second": 1716.946,
      "eval_steps_per_second": 214.63,
      "step": 39000
    },
    {
      "entropy": 5.953405523300171,
      "epoch": 3.4086340994494453,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003864332653821592,
      "loss": 5.1548,
      "mean_token_accuracy": 0.21839607805013656,
      "num_tokens": 79028549.0,
      "step": 39005
    },
    {
      "entropy": 6.000085830688477,
      "epoch": 3.4090710478021498,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003864061932531095,
      "loss": 5.2047,
      "mean_token_accuracy": 0.2081078201532364,
      "num_tokens": 79037788.0,
      "step": 39010
    },
    {
      "entropy": 6.0237242698669435,
      "epoch": 3.4095079961548547,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038637911898723067,
      "loss": 5.2528,
      "mean_token_accuracy": 0.20640605837106704,
      "num_tokens": 79048205.0,
      "step": 39015
    },
    {
      "entropy": 5.98281421661377,
      "epoch": 3.409944944507559,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003863520425850419,
      "loss": 5.1559,
      "mean_token_accuracy": 0.21019763499498367,
      "num_tokens": 79057594.0,
      "step": 39020
    },
    {
      "entropy": 6.019938611984253,
      "epoch": 3.410381892860264,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003863249640470627,
      "loss": 5.2389,
      "mean_token_accuracy": 0.21170325577259064,
      "num_tokens": 79066815.0,
      "step": 39025
    },
    {
      "entropy": 6.053456783294678,
      "epoch": 3.4108188412129685,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038629788337381227,
      "loss": 5.1873,
      "mean_token_accuracy": 0.21068113446235656,
      "num_tokens": 79077191.0,
      "step": 39030
    },
    {
      "entropy": 6.038092088699341,
      "epoch": 3.4112557895656734,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003862708005658102,
      "loss": 5.2466,
      "mean_token_accuracy": 0.20878908187150955,
      "num_tokens": 79086958.0,
      "step": 39035
    },
    {
      "entropy": 5.9953728199005125,
      "epoch": 3.411692737918378,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038624371562357573,
      "loss": 5.2691,
      "mean_token_accuracy": 0.2088411569595337,
      "num_tokens": 79098298.0,
      "step": 39040
    },
    {
      "entropy": 5.987518405914306,
      "epoch": 3.412129686271083,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038621662854762865,
      "loss": 5.1902,
      "mean_token_accuracy": 0.20948933511972428,
      "num_tokens": 79107954.0,
      "step": 39045
    },
    {
      "entropy": 6.012821292877197,
      "epoch": 3.4125666346237873,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003861895393384882,
      "loss": 5.2286,
      "mean_token_accuracy": 0.20996546596288682,
      "num_tokens": 79118062.0,
      "step": 39050
    },
    {
      "entropy": 6.043279933929443,
      "epoch": 3.413003582976492,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038616244799667427,
      "loss": 5.255,
      "mean_token_accuracy": 0.2085709646344185,
      "num_tokens": 79128358.0,
      "step": 39055
    },
    {
      "entropy": 6.079808616638184,
      "epoch": 3.413440531329197,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003861353545227063,
      "loss": 5.2584,
      "mean_token_accuracy": 0.20807139724493026,
      "num_tokens": 79139882.0,
      "step": 39060
    },
    {
      "entropy": 6.1040120124816895,
      "epoch": 3.4138774796819016,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003861082589171039,
      "loss": 5.2343,
      "mean_token_accuracy": 0.21157847195863724,
      "num_tokens": 79150748.0,
      "step": 39065
    },
    {
      "entropy": 5.985946464538574,
      "epoch": 3.4143144280346065,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00038608116118038705,
      "loss": 5.1964,
      "mean_token_accuracy": 0.21298699378967284,
      "num_tokens": 79160088.0,
      "step": 39070
    },
    {
      "entropy": 5.925455951690674,
      "epoch": 3.414751376387311,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003860540613130752,
      "loss": 5.152,
      "mean_token_accuracy": 0.2228095382452011,
      "num_tokens": 79169539.0,
      "step": 39075
    },
    {
      "entropy": 6.058934116363526,
      "epoch": 3.415188324740016,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038602695931568834,
      "loss": 5.2953,
      "mean_token_accuracy": 0.2048071339726448,
      "num_tokens": 79179505.0,
      "step": 39080
    },
    {
      "entropy": 6.010907888412476,
      "epoch": 3.4156252730927203,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038599985518874614,
      "loss": 5.2054,
      "mean_token_accuracy": 0.21485316455364228,
      "num_tokens": 79189791.0,
      "step": 39085
    },
    {
      "entropy": 6.085813331604004,
      "epoch": 3.4160622214454253,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038597274893276855,
      "loss": 5.2644,
      "mean_token_accuracy": 0.20458584129810334,
      "num_tokens": 79199708.0,
      "step": 39090
    },
    {
      "entropy": 6.0194755554199215,
      "epoch": 3.4164991697981297,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003859456405482755,
      "loss": 5.2469,
      "mean_token_accuracy": 0.20745437145233153,
      "num_tokens": 79209624.0,
      "step": 39095
    },
    {
      "entropy": 5.994833469390869,
      "epoch": 3.4169361181508346,
      "grad_norm": 1.28125,
      "learning_rate": 0.000385918530035787,
      "loss": 5.1924,
      "mean_token_accuracy": 0.21784724444150924,
      "num_tokens": 79218873.0,
      "step": 39100
    },
    {
      "entropy": 6.026698923110962,
      "epoch": 3.417373066503539,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038589141739582295,
      "loss": 5.254,
      "mean_token_accuracy": 0.2107363000512123,
      "num_tokens": 79228838.0,
      "step": 39105
    },
    {
      "entropy": 6.055309152603149,
      "epoch": 3.417810014856244,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003858643026289034,
      "loss": 5.2163,
      "mean_token_accuracy": 0.2085937887430191,
      "num_tokens": 79238185.0,
      "step": 39110
    },
    {
      "entropy": 6.060305118560791,
      "epoch": 3.418246963208949,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038583718573554845,
      "loss": 5.2294,
      "mean_token_accuracy": 0.20473540127277373,
      "num_tokens": 79248777.0,
      "step": 39115
    },
    {
      "entropy": 5.982485055923462,
      "epoch": 3.4186839115616534,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003858100667162781,
      "loss": 5.1477,
      "mean_token_accuracy": 0.21742440164089202,
      "num_tokens": 79258471.0,
      "step": 39120
    },
    {
      "entropy": 5.943774223327637,
      "epoch": 3.419120859914358,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003857829455716127,
      "loss": 5.145,
      "mean_token_accuracy": 0.21949035823345184,
      "num_tokens": 79268483.0,
      "step": 39125
    },
    {
      "entropy": 6.014630079269409,
      "epoch": 3.419557808267063,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003857558223020724,
      "loss": 5.2052,
      "mean_token_accuracy": 0.21499580144882202,
      "num_tokens": 79278497.0,
      "step": 39130
    },
    {
      "entropy": 6.038236522674561,
      "epoch": 3.4199947566197677,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003857286969081773,
      "loss": 5.207,
      "mean_token_accuracy": 0.21466685384511947,
      "num_tokens": 79288010.0,
      "step": 39135
    },
    {
      "entropy": 6.036044454574585,
      "epoch": 3.420431704972472,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038570156939044784,
      "loss": 5.2746,
      "mean_token_accuracy": 0.20782104581594468,
      "num_tokens": 79298092.0,
      "step": 39140
    },
    {
      "entropy": 6.045801162719727,
      "epoch": 3.420868653325177,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003856744397494042,
      "loss": 5.2716,
      "mean_token_accuracy": 0.2079498991370201,
      "num_tokens": 79308458.0,
      "step": 39145
    },
    {
      "entropy": 6.04626088142395,
      "epoch": 3.4213056016778816,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003856473079855668,
      "loss": 5.3079,
      "mean_token_accuracy": 0.19641191512346268,
      "num_tokens": 79318740.0,
      "step": 39150
    },
    {
      "entropy": 6.031207752227783,
      "epoch": 3.4217425500305865,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038562017409945614,
      "loss": 5.2624,
      "mean_token_accuracy": 0.21523342579603194,
      "num_tokens": 79329738.0,
      "step": 39155
    },
    {
      "entropy": 6.057099914550781,
      "epoch": 3.422179498383291,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038559303809159254,
      "loss": 5.2619,
      "mean_token_accuracy": 0.20969225466251373,
      "num_tokens": 79341137.0,
      "step": 39160
    },
    {
      "entropy": 6.022317028045654,
      "epoch": 3.422616446735996,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003855658999624965,
      "loss": 5.2311,
      "mean_token_accuracy": 0.20950252711772918,
      "num_tokens": 79351601.0,
      "step": 39165
    },
    {
      "entropy": 6.02075982093811,
      "epoch": 3.4230533950887003,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003855387597126886,
      "loss": 5.2336,
      "mean_token_accuracy": 0.2081596314907074,
      "num_tokens": 79360875.0,
      "step": 39170
    },
    {
      "entropy": 6.0828917026519775,
      "epoch": 3.4234903434414052,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038551161734268934,
      "loss": 5.2773,
      "mean_token_accuracy": 0.20202064961194993,
      "num_tokens": 79371033.0,
      "step": 39175
    },
    {
      "entropy": 6.064287090301514,
      "epoch": 3.4239272917941097,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038548447285301934,
      "loss": 5.2684,
      "mean_token_accuracy": 0.2139963060617447,
      "num_tokens": 79381237.0,
      "step": 39180
    },
    {
      "entropy": 6.082431173324585,
      "epoch": 3.4243642401468146,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003854573262441993,
      "loss": 5.2815,
      "mean_token_accuracy": 0.20097869336605073,
      "num_tokens": 79392279.0,
      "step": 39185
    },
    {
      "entropy": 6.01262264251709,
      "epoch": 3.4248011884995195,
      "grad_norm": 1.453125,
      "learning_rate": 0.00038543017751674986,
      "loss": 5.232,
      "mean_token_accuracy": 0.21051702350378038,
      "num_tokens": 79401723.0,
      "step": 39190
    },
    {
      "entropy": 5.925231027603149,
      "epoch": 3.425238136852224,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003854030266711917,
      "loss": 5.1432,
      "mean_token_accuracy": 0.2243478611111641,
      "num_tokens": 79413436.0,
      "step": 39195
    },
    {
      "entropy": 6.019675254821777,
      "epoch": 3.425675085204929,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038537587370804567,
      "loss": 5.2581,
      "mean_token_accuracy": 0.20906189680099488,
      "num_tokens": 79423923.0,
      "step": 39200
    },
    {
      "entropy": 6.085481357574463,
      "epoch": 3.4261120335576334,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003853487186278326,
      "loss": 5.2351,
      "mean_token_accuracy": 0.20712250620126724,
      "num_tokens": 79433992.0,
      "step": 39205
    },
    {
      "entropy": 6.037710952758789,
      "epoch": 3.4265489819103383,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038532156143107323,
      "loss": 5.2875,
      "mean_token_accuracy": 0.2089540809392929,
      "num_tokens": 79444564.0,
      "step": 39210
    },
    {
      "entropy": 6.063961362838745,
      "epoch": 3.4269859302630428,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003852944021182885,
      "loss": 5.2428,
      "mean_token_accuracy": 0.20515889674425125,
      "num_tokens": 79454985.0,
      "step": 39215
    },
    {
      "entropy": 6.0541379928588865,
      "epoch": 3.4274228786157477,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003852672406899993,
      "loss": 5.2692,
      "mean_token_accuracy": 0.202752186357975,
      "num_tokens": 79464959.0,
      "step": 39220
    },
    {
      "entropy": 6.053845405578613,
      "epoch": 3.427859826968452,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038524007714672674,
      "loss": 5.2323,
      "mean_token_accuracy": 0.20866752415895462,
      "num_tokens": 79474679.0,
      "step": 39225
    },
    {
      "entropy": 6.0375104427337645,
      "epoch": 3.428296775321157,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003852129114889916,
      "loss": 5.2719,
      "mean_token_accuracy": 0.21047555208206176,
      "num_tokens": 79485908.0,
      "step": 39230
    },
    {
      "entropy": 5.999696874618531,
      "epoch": 3.4287337236738615,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003851857437173152,
      "loss": 5.2638,
      "mean_token_accuracy": 0.20830924659967423,
      "num_tokens": 79496209.0,
      "step": 39235
    },
    {
      "entropy": 6.029391098022461,
      "epoch": 3.4291706720265664,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003851585738322184,
      "loss": 5.1965,
      "mean_token_accuracy": 0.21423022001981734,
      "num_tokens": 79507351.0,
      "step": 39240
    },
    {
      "entropy": 6.010392189025879,
      "epoch": 3.4296076203792714,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038513140183422256,
      "loss": 5.1998,
      "mean_token_accuracy": 0.20674879103899002,
      "num_tokens": 79517517.0,
      "step": 39245
    },
    {
      "entropy": 5.989939785003662,
      "epoch": 3.430044568731976,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003851042277238486,
      "loss": 5.1573,
      "mean_token_accuracy": 0.2162923514842987,
      "num_tokens": 79528616.0,
      "step": 39250
    },
    {
      "entropy": 6.038872289657593,
      "epoch": 3.4304815170846807,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003850770515016179,
      "loss": 5.2302,
      "mean_token_accuracy": 0.20911580920219422,
      "num_tokens": 79538424.0,
      "step": 39255
    },
    {
      "entropy": 6.059786081314087,
      "epoch": 3.430918465437385,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003850498731680517,
      "loss": 5.3069,
      "mean_token_accuracy": 0.1965608537197113,
      "num_tokens": 79548717.0,
      "step": 39260
    },
    {
      "entropy": 6.0480568408966064,
      "epoch": 3.43135541379009,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038502269272367123,
      "loss": 5.1712,
      "mean_token_accuracy": 0.21464803814888,
      "num_tokens": 79558998.0,
      "step": 39265
    },
    {
      "entropy": 6.078284692764282,
      "epoch": 3.4317923621427946,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038499551016899793,
      "loss": 5.2959,
      "mean_token_accuracy": 0.20072404742240907,
      "num_tokens": 79569424.0,
      "step": 39270
    },
    {
      "entropy": 6.027145338058472,
      "epoch": 3.4322293104954995,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038496832550455307,
      "loss": 5.2424,
      "mean_token_accuracy": 0.20526109933853148,
      "num_tokens": 79578670.0,
      "step": 39275
    },
    {
      "entropy": 6.054463863372803,
      "epoch": 3.432666258848204,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038494113873085805,
      "loss": 5.3174,
      "mean_token_accuracy": 0.2031782776117325,
      "num_tokens": 79588883.0,
      "step": 39280
    },
    {
      "entropy": 6.013948392868042,
      "epoch": 3.433103207200909,
      "grad_norm": 1.375,
      "learning_rate": 0.00038491394984843446,
      "loss": 5.1999,
      "mean_token_accuracy": 0.21065506488084793,
      "num_tokens": 79598158.0,
      "step": 39285
    },
    {
      "entropy": 5.991197490692139,
      "epoch": 3.4335401555536134,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038488675885780366,
      "loss": 5.1815,
      "mean_token_accuracy": 0.21297522485256196,
      "num_tokens": 79608699.0,
      "step": 39290
    },
    {
      "entropy": 6.049648904800415,
      "epoch": 3.4339771039063183,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038485956575948726,
      "loss": 5.1882,
      "mean_token_accuracy": 0.22034844905138015,
      "num_tokens": 79618628.0,
      "step": 39295
    },
    {
      "entropy": 6.046235275268555,
      "epoch": 3.434414052259023,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003848323705540069,
      "loss": 5.2145,
      "mean_token_accuracy": 0.20724441409111022,
      "num_tokens": 79628836.0,
      "step": 39300
    },
    {
      "entropy": 6.0350748062133786,
      "epoch": 3.4348510006117277,
      "grad_norm": 1.296875,
      "learning_rate": 0.000384805173241884,
      "loss": 5.1743,
      "mean_token_accuracy": 0.20726107358932494,
      "num_tokens": 79638306.0,
      "step": 39305
    },
    {
      "entropy": 6.009746170043945,
      "epoch": 3.4352879489644326,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038477797382364037,
      "loss": 5.2857,
      "mean_token_accuracy": 0.20859532952308654,
      "num_tokens": 79647955.0,
      "step": 39310
    },
    {
      "entropy": 6.044074773788452,
      "epoch": 3.435724897317137,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038475077229979774,
      "loss": 5.2232,
      "mean_token_accuracy": 0.21219952553510665,
      "num_tokens": 79658701.0,
      "step": 39315
    },
    {
      "entropy": 6.019621849060059,
      "epoch": 3.436161845669842,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00038472356867087776,
      "loss": 5.2657,
      "mean_token_accuracy": 0.20929087549448014,
      "num_tokens": 79669436.0,
      "step": 39320
    },
    {
      "entropy": 6.05835657119751,
      "epoch": 3.4365987940225464,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003846963629374022,
      "loss": 5.2759,
      "mean_token_accuracy": 0.20150029361248017,
      "num_tokens": 79679350.0,
      "step": 39325
    },
    {
      "entropy": 6.046243333816529,
      "epoch": 3.4370357423752513,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038466915509989297,
      "loss": 5.1504,
      "mean_token_accuracy": 0.21969217360019683,
      "num_tokens": 79689762.0,
      "step": 39330
    },
    {
      "entropy": 6.03247389793396,
      "epoch": 3.437472690727956,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038464194515887187,
      "loss": 5.1734,
      "mean_token_accuracy": 0.21192119866609574,
      "num_tokens": 79699594.0,
      "step": 39335
    },
    {
      "entropy": 5.992573881149292,
      "epoch": 3.4379096390806607,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003846147331148608,
      "loss": 5.2938,
      "mean_token_accuracy": 0.2071098953485489,
      "num_tokens": 79710403.0,
      "step": 39340
    },
    {
      "entropy": 5.930615520477295,
      "epoch": 3.438346587433365,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038458751896838176,
      "loss": 5.0883,
      "mean_token_accuracy": 0.2250412166118622,
      "num_tokens": 79721291.0,
      "step": 39345
    },
    {
      "entropy": 6.048918724060059,
      "epoch": 3.43878353578607,
      "grad_norm": 1.25,
      "learning_rate": 0.0003845603027199566,
      "loss": 5.2616,
      "mean_token_accuracy": 0.2042481020092964,
      "num_tokens": 79732505.0,
      "step": 39350
    },
    {
      "entropy": 6.049346828460694,
      "epoch": 3.439220484138775,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038453308437010747,
      "loss": 5.275,
      "mean_token_accuracy": 0.21032518595457078,
      "num_tokens": 79743707.0,
      "step": 39355
    },
    {
      "entropy": 6.096337747573853,
      "epoch": 3.4396574324914795,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038450586391935635,
      "loss": 5.2732,
      "mean_token_accuracy": 0.20142365097999573,
      "num_tokens": 79754669.0,
      "step": 39360
    },
    {
      "entropy": 6.03383059501648,
      "epoch": 3.4400943808441844,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003844786413682254,
      "loss": 5.2324,
      "mean_token_accuracy": 0.2158777266740799,
      "num_tokens": 79764444.0,
      "step": 39365
    },
    {
      "entropy": 6.032918643951416,
      "epoch": 3.440531329196889,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003844514167172368,
      "loss": 5.2744,
      "mean_token_accuracy": 0.19982291162014007,
      "num_tokens": 79774942.0,
      "step": 39370
    },
    {
      "entropy": 6.072532367706299,
      "epoch": 3.440968277549594,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003844241899669126,
      "loss": 5.2533,
      "mean_token_accuracy": 0.19877638071775436,
      "num_tokens": 79785710.0,
      "step": 39375
    },
    {
      "entropy": 6.08306999206543,
      "epoch": 3.4414052259022982,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003843969611177751,
      "loss": 5.3027,
      "mean_token_accuracy": 0.201312917470932,
      "num_tokens": 79795841.0,
      "step": 39380
    },
    {
      "entropy": 6.00464243888855,
      "epoch": 3.441842174255003,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003843697301703466,
      "loss": 5.1789,
      "mean_token_accuracy": 0.20934654474258424,
      "num_tokens": 79805400.0,
      "step": 39385
    },
    {
      "entropy": 5.979266214370727,
      "epoch": 3.4422791226077076,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003843424971251494,
      "loss": 5.2633,
      "mean_token_accuracy": 0.20867893248796462,
      "num_tokens": 79816212.0,
      "step": 39390
    },
    {
      "entropy": 6.0399268627166744,
      "epoch": 3.4427160709604125,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003843152619827058,
      "loss": 5.2031,
      "mean_token_accuracy": 0.20971101969480516,
      "num_tokens": 79827125.0,
      "step": 39395
    },
    {
      "entropy": 6.1172700881958,
      "epoch": 3.443153019313117,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038428802474353816,
      "loss": 5.275,
      "mean_token_accuracy": 0.20579421818256377,
      "num_tokens": 79837436.0,
      "step": 39400
    },
    {
      "entropy": 6.100228357315063,
      "epoch": 3.443589967665822,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003842607854081688,
      "loss": 5.3048,
      "mean_token_accuracy": 0.2027877986431122,
      "num_tokens": 79847297.0,
      "step": 39405
    },
    {
      "entropy": 6.00754680633545,
      "epoch": 3.444026916018527,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003842335439771205,
      "loss": 5.2024,
      "mean_token_accuracy": 0.20670555233955384,
      "num_tokens": 79856321.0,
      "step": 39410
    },
    {
      "entropy": 6.012883615493775,
      "epoch": 3.4444638643712313,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038420630045091556,
      "loss": 5.1958,
      "mean_token_accuracy": 0.2138250395655632,
      "num_tokens": 79866244.0,
      "step": 39415
    },
    {
      "entropy": 6.077204942703247,
      "epoch": 3.4449008127239362,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003841790548300765,
      "loss": 5.3017,
      "mean_token_accuracy": 0.20106435865163802,
      "num_tokens": 79876900.0,
      "step": 39420
    },
    {
      "entropy": 6.039185237884522,
      "epoch": 3.4453377610766407,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038415180711512604,
      "loss": 5.2746,
      "mean_token_accuracy": 0.21062643676996232,
      "num_tokens": 79887864.0,
      "step": 39425
    },
    {
      "entropy": 6.0527444839477536,
      "epoch": 3.4457747094293456,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003841245573065867,
      "loss": 5.2235,
      "mean_token_accuracy": 0.20892269164323807,
      "num_tokens": 79897625.0,
      "step": 39430
    },
    {
      "entropy": 6.091646194458008,
      "epoch": 3.44621165778205,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003840973054049812,
      "loss": 5.2431,
      "mean_token_accuracy": 0.2080795496702194,
      "num_tokens": 79907742.0,
      "step": 39435
    },
    {
      "entropy": 6.030308866500855,
      "epoch": 3.446648606134755,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038407005141083216,
      "loss": 5.2291,
      "mean_token_accuracy": 0.20686045438051223,
      "num_tokens": 79917974.0,
      "step": 39440
    },
    {
      "entropy": 6.021585988998413,
      "epoch": 3.4470855544874595,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003840427953246624,
      "loss": 5.2367,
      "mean_token_accuracy": 0.21002622097730636,
      "num_tokens": 79928089.0,
      "step": 39445
    },
    {
      "entropy": 5.9768627166748045,
      "epoch": 3.4475225028401644,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038401553714699466,
      "loss": 5.1714,
      "mean_token_accuracy": 0.21257845759391786,
      "num_tokens": 79938396.0,
      "step": 39450
    },
    {
      "entropy": 6.0760109424591064,
      "epoch": 3.447959451192869,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003839882768783518,
      "loss": 5.2568,
      "mean_token_accuracy": 0.2062744289636612,
      "num_tokens": 79948157.0,
      "step": 39455
    },
    {
      "entropy": 6.082671356201172,
      "epoch": 3.4483963995455738,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003839610145192567,
      "loss": 5.2574,
      "mean_token_accuracy": 0.2071920409798622,
      "num_tokens": 79958653.0,
      "step": 39460
    },
    {
      "entropy": 5.988761472702026,
      "epoch": 3.4488333478982787,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003839337500702323,
      "loss": 5.2593,
      "mean_token_accuracy": 0.20708468854427337,
      "num_tokens": 79969180.0,
      "step": 39465
    },
    {
      "entropy": 6.044090652465821,
      "epoch": 3.449270296250983,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038390648353180145,
      "loss": 5.2911,
      "mean_token_accuracy": 0.2084146559238434,
      "num_tokens": 79978467.0,
      "step": 39470
    },
    {
      "entropy": 6.04366660118103,
      "epoch": 3.4497072446036876,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003838792149044871,
      "loss": 5.1986,
      "mean_token_accuracy": 0.2110608071088791,
      "num_tokens": 79989071.0,
      "step": 39475
    },
    {
      "entropy": 6.054151439666748,
      "epoch": 3.4501441929563925,
      "grad_norm": 1.34375,
      "learning_rate": 0.00038385194418881247,
      "loss": 5.2156,
      "mean_token_accuracy": 0.2121250182390213,
      "num_tokens": 79999476.0,
      "step": 39480
    },
    {
      "entropy": 5.967070770263672,
      "epoch": 3.4505811413090974,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038382467138530055,
      "loss": 5.1448,
      "mean_token_accuracy": 0.22277070879936217,
      "num_tokens": 80009986.0,
      "step": 39485
    },
    {
      "entropy": 5.996435737609863,
      "epoch": 3.451018089661802,
      "grad_norm": 1.453125,
      "learning_rate": 0.00038379739649447425,
      "loss": 5.3364,
      "mean_token_accuracy": 0.2039071500301361,
      "num_tokens": 80020916.0,
      "step": 39490
    },
    {
      "entropy": 6.03862771987915,
      "epoch": 3.451455038014507,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000383770119516857,
      "loss": 5.2004,
      "mean_token_accuracy": 0.21497852355241776,
      "num_tokens": 80030769.0,
      "step": 39495
    },
    {
      "entropy": 6.07888879776001,
      "epoch": 3.4518919863672113,
      "grad_norm": 1.25,
      "learning_rate": 0.0003837428404529718,
      "loss": 5.2169,
      "mean_token_accuracy": 0.21328461468219756,
      "num_tokens": 80040838.0,
      "step": 39500
    },
    {
      "entropy": 6.08407621383667,
      "epoch": 3.452328934719916,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038371555930334194,
      "loss": 5.2593,
      "mean_token_accuracy": 0.20532893538475036,
      "num_tokens": 80050600.0,
      "step": 39505
    },
    {
      "entropy": 6.045818424224853,
      "epoch": 3.4527658830726207,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003836882760684907,
      "loss": 5.2474,
      "mean_token_accuracy": 0.20853438675403596,
      "num_tokens": 80060826.0,
      "step": 39510
    },
    {
      "entropy": 6.051508855819702,
      "epoch": 3.4532028314253256,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003836609907489414,
      "loss": 5.2789,
      "mean_token_accuracy": 0.20988459140062332,
      "num_tokens": 80070213.0,
      "step": 39515
    },
    {
      "entropy": 6.022919654846191,
      "epoch": 3.45363977977803,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003836337033452173,
      "loss": 5.1721,
      "mean_token_accuracy": 0.21175759583711623,
      "num_tokens": 80080197.0,
      "step": 39520
    },
    {
      "entropy": 6.119462347030639,
      "epoch": 3.454076728130735,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003836064138578419,
      "loss": 5.272,
      "mean_token_accuracy": 0.21149349808692933,
      "num_tokens": 80089504.0,
      "step": 39525
    },
    {
      "entropy": 6.034992933273315,
      "epoch": 3.4545136764834394,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003835791222873386,
      "loss": 5.2166,
      "mean_token_accuracy": 0.21476511061191558,
      "num_tokens": 80099621.0,
      "step": 39530
    },
    {
      "entropy": 5.969288635253906,
      "epoch": 3.4549506248361443,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038355182863423077,
      "loss": 5.2147,
      "mean_token_accuracy": 0.2122943088412285,
      "num_tokens": 80109880.0,
      "step": 39535
    },
    {
      "entropy": 5.998900604248047,
      "epoch": 3.4553875731888493,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038352453289904206,
      "loss": 5.2024,
      "mean_token_accuracy": 0.21707064062356948,
      "num_tokens": 80119236.0,
      "step": 39540
    },
    {
      "entropy": 5.9518513679504395,
      "epoch": 3.4558245215415537,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00038349723508229587,
      "loss": 5.2056,
      "mean_token_accuracy": 0.21412067264318466,
      "num_tokens": 80129598.0,
      "step": 39545
    },
    {
      "entropy": 6.123792362213135,
      "epoch": 3.4562614698942586,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00038346993518451587,
      "loss": 5.2389,
      "mean_token_accuracy": 0.19972815066576005,
      "num_tokens": 80140050.0,
      "step": 39550
    },
    {
      "entropy": 6.065082168579101,
      "epoch": 3.456698418246963,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038344263320622574,
      "loss": 5.1847,
      "mean_token_accuracy": 0.2162223771214485,
      "num_tokens": 80150101.0,
      "step": 39555
    },
    {
      "entropy": 6.073480701446533,
      "epoch": 3.457135366599668,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00038341532914794904,
      "loss": 5.2438,
      "mean_token_accuracy": 0.20285082906484603,
      "num_tokens": 80159416.0,
      "step": 39560
    },
    {
      "entropy": 5.986310577392578,
      "epoch": 3.4575723149523725,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003833880230102095,
      "loss": 5.2278,
      "mean_token_accuracy": 0.21178338676691055,
      "num_tokens": 80168979.0,
      "step": 39565
    },
    {
      "entropy": 6.015607070922852,
      "epoch": 3.4580092633050774,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038336071479353086,
      "loss": 5.2254,
      "mean_token_accuracy": 0.21214055120944977,
      "num_tokens": 80179328.0,
      "step": 39570
    },
    {
      "entropy": 6.0651562213897705,
      "epoch": 3.458446211657782,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038333340449843694,
      "loss": 5.1832,
      "mean_token_accuracy": 0.21725483387708663,
      "num_tokens": 80189151.0,
      "step": 39575
    },
    {
      "entropy": 6.047548151016235,
      "epoch": 3.458883160010487,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003833060921254517,
      "loss": 5.2677,
      "mean_token_accuracy": 0.21044930964708328,
      "num_tokens": 80199622.0,
      "step": 39580
    },
    {
      "entropy": 6.088414287567138,
      "epoch": 3.4593201083631913,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038327877767509866,
      "loss": 5.2728,
      "mean_token_accuracy": 0.20577050149440765,
      "num_tokens": 80209335.0,
      "step": 39585
    },
    {
      "entropy": 6.0516712188720705,
      "epoch": 3.459757056715896,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000383251461147902,
      "loss": 5.2541,
      "mean_token_accuracy": 0.2082236737012863,
      "num_tokens": 80220336.0,
      "step": 39590
    },
    {
      "entropy": 6.048176860809326,
      "epoch": 3.460194005068601,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038322414254438557,
      "loss": 5.2665,
      "mean_token_accuracy": 0.20365791916847228,
      "num_tokens": 80231459.0,
      "step": 39595
    },
    {
      "entropy": 6.052505874633789,
      "epoch": 3.4606309534213056,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003831968218650735,
      "loss": 5.285,
      "mean_token_accuracy": 0.209747876226902,
      "num_tokens": 80240997.0,
      "step": 39600
    },
    {
      "entropy": 6.035834932327271,
      "epoch": 3.4610679017740105,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00038316949911048953,
      "loss": 5.2477,
      "mean_token_accuracy": 0.2090597227215767,
      "num_tokens": 80251663.0,
      "step": 39605
    },
    {
      "entropy": 6.058802700042724,
      "epoch": 3.461504850126715,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038314217428115797,
      "loss": 5.2929,
      "mean_token_accuracy": 0.20445154905319213,
      "num_tokens": 80261059.0,
      "step": 39610
    },
    {
      "entropy": 5.989273118972778,
      "epoch": 3.46194179847942,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003831148473776028,
      "loss": 5.191,
      "mean_token_accuracy": 0.21499477177858353,
      "num_tokens": 80271781.0,
      "step": 39615
    },
    {
      "entropy": 6.0169126987457275,
      "epoch": 3.4623787468321243,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038308751840034826,
      "loss": 5.2611,
      "mean_token_accuracy": 0.2076971411705017,
      "num_tokens": 80281996.0,
      "step": 39620
    },
    {
      "entropy": 5.987188196182251,
      "epoch": 3.4628156951848292,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003830601873499185,
      "loss": 5.2112,
      "mean_token_accuracy": 0.21568584442138672,
      "num_tokens": 80293181.0,
      "step": 39625
    },
    {
      "entropy": 6.052859973907471,
      "epoch": 3.4632526435375337,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003830328542268377,
      "loss": 5.2379,
      "mean_token_accuracy": 0.20838479697704315,
      "num_tokens": 80303697.0,
      "step": 39630
    },
    {
      "entropy": 6.003262138366699,
      "epoch": 3.4636895918902386,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038300551903163013,
      "loss": 5.1547,
      "mean_token_accuracy": 0.21724520325660707,
      "num_tokens": 80313104.0,
      "step": 39635
    },
    {
      "entropy": 6.00202260017395,
      "epoch": 3.464126540242943,
      "grad_norm": 1.390625,
      "learning_rate": 0.00038297818176482006,
      "loss": 5.1559,
      "mean_token_accuracy": 0.21803707480430604,
      "num_tokens": 80323168.0,
      "step": 39640
    },
    {
      "entropy": 6.044218683242798,
      "epoch": 3.464563488595648,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000382950842426932,
      "loss": 5.2424,
      "mean_token_accuracy": 0.21325790882110596,
      "num_tokens": 80333064.0,
      "step": 39645
    },
    {
      "entropy": 6.01294355392456,
      "epoch": 3.465000436948353,
      "grad_norm": 1.25,
      "learning_rate": 0.00038292350101849017,
      "loss": 5.1423,
      "mean_token_accuracy": 0.2177430734038353,
      "num_tokens": 80342827.0,
      "step": 39650
    },
    {
      "entropy": 6.013788461685181,
      "epoch": 3.4654373853010574,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000382896157540019,
      "loss": 5.2769,
      "mean_token_accuracy": 0.20138885378837584,
      "num_tokens": 80352732.0,
      "step": 39655
    },
    {
      "entropy": 6.053360891342163,
      "epoch": 3.4658743336537623,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000382868811992043,
      "loss": 5.2944,
      "mean_token_accuracy": 0.20035041123628616,
      "num_tokens": 80362427.0,
      "step": 39660
    },
    {
      "entropy": 6.073028421401977,
      "epoch": 3.4663112820064668,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003828414643750867,
      "loss": 5.1966,
      "mean_token_accuracy": 0.21644284278154374,
      "num_tokens": 80372719.0,
      "step": 39665
    },
    {
      "entropy": 6.099404954910279,
      "epoch": 3.4667482303591717,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038281411468967457,
      "loss": 5.1982,
      "mean_token_accuracy": 0.20865747034549714,
      "num_tokens": 80381753.0,
      "step": 39670
    },
    {
      "entropy": 5.993002605438233,
      "epoch": 3.467185178711876,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003827867629363312,
      "loss": 5.2313,
      "mean_token_accuracy": 0.21597666144371033,
      "num_tokens": 80393401.0,
      "step": 39675
    },
    {
      "entropy": 6.075854253768921,
      "epoch": 3.467622127064581,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038275940911558135,
      "loss": 5.2769,
      "mean_token_accuracy": 0.20268284231424333,
      "num_tokens": 80403194.0,
      "step": 39680
    },
    {
      "entropy": 6.071134376525879,
      "epoch": 3.4680590754172855,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003827320532279495,
      "loss": 5.2243,
      "mean_token_accuracy": 0.20257875621318816,
      "num_tokens": 80413283.0,
      "step": 39685
    },
    {
      "entropy": 6.068140602111816,
      "epoch": 3.4684960237699904,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00038270469527396035,
      "loss": 5.2225,
      "mean_token_accuracy": 0.2096667021512985,
      "num_tokens": 80423976.0,
      "step": 39690
    },
    {
      "entropy": 6.011561155319214,
      "epoch": 3.468932972122695,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038267733525413884,
      "loss": 5.2317,
      "mean_token_accuracy": 0.21494656056165695,
      "num_tokens": 80434017.0,
      "step": 39695
    },
    {
      "entropy": 6.036246871948242,
      "epoch": 3.4693699204754,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003826499731690095,
      "loss": 5.204,
      "mean_token_accuracy": 0.21225909143686295,
      "num_tokens": 80444289.0,
      "step": 39700
    },
    {
      "entropy": 6.084712886810303,
      "epoch": 3.4698068688281047,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003826226090190974,
      "loss": 5.2592,
      "mean_token_accuracy": 0.20827787220478058,
      "num_tokens": 80453420.0,
      "step": 39705
    },
    {
      "entropy": 6.115058422088623,
      "epoch": 3.470243817180809,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038259524280492716,
      "loss": 5.2615,
      "mean_token_accuracy": 0.2078891783952713,
      "num_tokens": 80462977.0,
      "step": 39710
    },
    {
      "entropy": 6.008103609085083,
      "epoch": 3.470680765533514,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038256787452702377,
      "loss": 5.1367,
      "mean_token_accuracy": 0.21000342965126037,
      "num_tokens": 80471806.0,
      "step": 39715
    },
    {
      "entropy": 5.987054777145386,
      "epoch": 3.4711177138862186,
      "grad_norm": 1.40625,
      "learning_rate": 0.00038254050418591227,
      "loss": 5.2132,
      "mean_token_accuracy": 0.20803635120391845,
      "num_tokens": 80481510.0,
      "step": 39720
    },
    {
      "entropy": 6.062936449050904,
      "epoch": 3.4715546622389235,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003825131317821175,
      "loss": 5.2502,
      "mean_token_accuracy": 0.20371328741312028,
      "num_tokens": 80491670.0,
      "step": 39725
    },
    {
      "entropy": 6.102622127532959,
      "epoch": 3.471991610591628,
      "grad_norm": 1.359375,
      "learning_rate": 0.00038248575731616454,
      "loss": 5.2958,
      "mean_token_accuracy": 0.205492602288723,
      "num_tokens": 80501518.0,
      "step": 39730
    },
    {
      "entropy": 6.040682411193847,
      "epoch": 3.472428558944333,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038245838078857837,
      "loss": 5.2164,
      "mean_token_accuracy": 0.21201702058315278,
      "num_tokens": 80511943.0,
      "step": 39735
    },
    {
      "entropy": 5.976648426055908,
      "epoch": 3.4728655072970374,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003824310021998841,
      "loss": 5.197,
      "mean_token_accuracy": 0.2050633117556572,
      "num_tokens": 80523114.0,
      "step": 39740
    },
    {
      "entropy": 6.079890871047974,
      "epoch": 3.4733024556497423,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003824036215506071,
      "loss": 5.2994,
      "mean_token_accuracy": 0.2017757073044777,
      "num_tokens": 80534802.0,
      "step": 39745
    },
    {
      "entropy": 6.131818246841431,
      "epoch": 3.4737394040024467,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038237623884127225,
      "loss": 5.2458,
      "mean_token_accuracy": 0.20398052036762238,
      "num_tokens": 80544367.0,
      "step": 39750
    },
    {
      "entropy": 6.0316225528717045,
      "epoch": 3.4741763523551517,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038234885407240494,
      "loss": 5.2786,
      "mean_token_accuracy": 0.20953186005353927,
      "num_tokens": 80554841.0,
      "step": 39755
    },
    {
      "entropy": 6.046762180328369,
      "epoch": 3.4746133007078566,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003823214672445302,
      "loss": 5.2619,
      "mean_token_accuracy": 0.2030887708067894,
      "num_tokens": 80565076.0,
      "step": 39760
    },
    {
      "entropy": 6.088379859924316,
      "epoch": 3.475050249060561,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038229407835817366,
      "loss": 5.3284,
      "mean_token_accuracy": 0.20259055197238923,
      "num_tokens": 80575410.0,
      "step": 39765
    },
    {
      "entropy": 5.990134382247925,
      "epoch": 3.4754871974132655,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038226668741386046,
      "loss": 5.2263,
      "mean_token_accuracy": 0.21148757934570311,
      "num_tokens": 80585400.0,
      "step": 39770
    },
    {
      "entropy": 6.080948114395142,
      "epoch": 3.4759241457659704,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003822392944121159,
      "loss": 5.2254,
      "mean_token_accuracy": 0.206475330889225,
      "num_tokens": 80595429.0,
      "step": 39775
    },
    {
      "entropy": 6.058638334274292,
      "epoch": 3.4763610941186753,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003822118993534655,
      "loss": 5.2058,
      "mean_token_accuracy": 0.21013966351747512,
      "num_tokens": 80605060.0,
      "step": 39780
    },
    {
      "entropy": 6.021814680099487,
      "epoch": 3.47679804247138,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003821845022384348,
      "loss": 5.1356,
      "mean_token_accuracy": 0.21586949974298478,
      "num_tokens": 80614927.0,
      "step": 39785
    },
    {
      "entropy": 5.988208246231079,
      "epoch": 3.4772349908240847,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038215710306754906,
      "loss": 5.225,
      "mean_token_accuracy": 0.20999795943498611,
      "num_tokens": 80624748.0,
      "step": 39790
    },
    {
      "entropy": 6.081703519821167,
      "epoch": 3.477671939176789,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038212970184133395,
      "loss": 5.3241,
      "mean_token_accuracy": 0.19760188162326814,
      "num_tokens": 80635221.0,
      "step": 39795
    },
    {
      "entropy": 6.01239013671875,
      "epoch": 3.478108887529494,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038210229856031503,
      "loss": 5.2663,
      "mean_token_accuracy": 0.21120058745145798,
      "num_tokens": 80645268.0,
      "step": 39800
    },
    {
      "entropy": 6.175289678573608,
      "epoch": 3.4785458358821986,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003820748932250179,
      "loss": 5.2906,
      "mean_token_accuracy": 0.20032318979501723,
      "num_tokens": 80655209.0,
      "step": 39805
    },
    {
      "entropy": 6.072188520431519,
      "epoch": 3.4789827842349035,
      "grad_norm": 1.25,
      "learning_rate": 0.00038204748583596835,
      "loss": 5.2941,
      "mean_token_accuracy": 0.2083667829632759,
      "num_tokens": 80665652.0,
      "step": 39810
    },
    {
      "entropy": 6.007361555099488,
      "epoch": 3.4794197325876084,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003820200763936917,
      "loss": 5.2721,
      "mean_token_accuracy": 0.20217716842889785,
      "num_tokens": 80675430.0,
      "step": 39815
    },
    {
      "entropy": 6.0586442947387695,
      "epoch": 3.479856680940313,
      "grad_norm": 1.25,
      "learning_rate": 0.0003819926648987141,
      "loss": 5.2305,
      "mean_token_accuracy": 0.2095348507165909,
      "num_tokens": 80685997.0,
      "step": 39820
    },
    {
      "entropy": 6.083933544158936,
      "epoch": 3.4802936292930173,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038196525135156094,
      "loss": 5.2511,
      "mean_token_accuracy": 0.20531177669763565,
      "num_tokens": 80697470.0,
      "step": 39825
    },
    {
      "entropy": 6.054359483718872,
      "epoch": 3.4807305776457222,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003819378357527583,
      "loss": 5.1676,
      "mean_token_accuracy": 0.21316763907670974,
      "num_tokens": 80708810.0,
      "step": 39830
    },
    {
      "entropy": 6.01168041229248,
      "epoch": 3.481167525998427,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000381910418102832,
      "loss": 5.1602,
      "mean_token_accuracy": 0.22020832300186158,
      "num_tokens": 80718036.0,
      "step": 39835
    },
    {
      "entropy": 6.046662187576294,
      "epoch": 3.4816044743511316,
      "grad_norm": 1.25,
      "learning_rate": 0.0003818829984023078,
      "loss": 5.2875,
      "mean_token_accuracy": 0.20218835920095443,
      "num_tokens": 80728247.0,
      "step": 39840
    },
    {
      "entropy": 6.002243280410767,
      "epoch": 3.4820414227038365,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038185557665171174,
      "loss": 5.2588,
      "mean_token_accuracy": 0.21107019931077958,
      "num_tokens": 80738873.0,
      "step": 39845
    },
    {
      "entropy": 6.0376404285430905,
      "epoch": 3.482478371056541,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038182815285156964,
      "loss": 5.2741,
      "mean_token_accuracy": 0.20558809936046601,
      "num_tokens": 80748734.0,
      "step": 39850
    },
    {
      "entropy": 6.056715822219848,
      "epoch": 3.482915319409246,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003818007270024076,
      "loss": 5.2275,
      "mean_token_accuracy": 0.21444143652915953,
      "num_tokens": 80759256.0,
      "step": 39855
    },
    {
      "entropy": 6.018159675598144,
      "epoch": 3.4833522677619504,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003817732991047518,
      "loss": 5.1976,
      "mean_token_accuracy": 0.21424007266759873,
      "num_tokens": 80769670.0,
      "step": 39860
    },
    {
      "entropy": 5.970856475830078,
      "epoch": 3.4837892161146553,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038174586915912804,
      "loss": 5.1775,
      "mean_token_accuracy": 0.21561715602874756,
      "num_tokens": 80778937.0,
      "step": 39865
    },
    {
      "entropy": 5.985656213760376,
      "epoch": 3.4842261644673598,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038171843716606254,
      "loss": 5.2005,
      "mean_token_accuracy": 0.2101365804672241,
      "num_tokens": 80789329.0,
      "step": 39870
    },
    {
      "entropy": 6.00484037399292,
      "epoch": 3.4846631128200647,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038169100312608156,
      "loss": 5.2029,
      "mean_token_accuracy": 0.2115618959069252,
      "num_tokens": 80798943.0,
      "step": 39875
    },
    {
      "entropy": 6.085920953750611,
      "epoch": 3.485100061172769,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003816635670397112,
      "loss": 5.3086,
      "mean_token_accuracy": 0.20310102254152299,
      "num_tokens": 80808915.0,
      "step": 39880
    },
    {
      "entropy": 6.0181482315063475,
      "epoch": 3.485537009525474,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038163612890747777,
      "loss": 5.2262,
      "mean_token_accuracy": 0.2126046434044838,
      "num_tokens": 80820162.0,
      "step": 39885
    },
    {
      "entropy": 5.948827314376831,
      "epoch": 3.485973957878179,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003816086887299075,
      "loss": 5.1336,
      "mean_token_accuracy": 0.22811453193426132,
      "num_tokens": 80830179.0,
      "step": 39890
    },
    {
      "entropy": 6.01762113571167,
      "epoch": 3.4864109062308835,
      "grad_norm": 1.25,
      "learning_rate": 0.0003815812465075267,
      "loss": 5.2169,
      "mean_token_accuracy": 0.209135565161705,
      "num_tokens": 80839924.0,
      "step": 39895
    },
    {
      "entropy": 6.039445304870606,
      "epoch": 3.4868478545835884,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00038155380224086177,
      "loss": 5.2355,
      "mean_token_accuracy": 0.21409563720226288,
      "num_tokens": 80849594.0,
      "step": 39900
    },
    {
      "entropy": 6.019087076187134,
      "epoch": 3.487284802936293,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003815263559304391,
      "loss": 5.1802,
      "mean_token_accuracy": 0.2121412619948387,
      "num_tokens": 80859706.0,
      "step": 39905
    },
    {
      "entropy": 6.073727798461914,
      "epoch": 3.4877217512889978,
      "grad_norm": 1.359375,
      "learning_rate": 0.00038149890757678507,
      "loss": 5.3001,
      "mean_token_accuracy": 0.20210032612085344,
      "num_tokens": 80869121.0,
      "step": 39910
    },
    {
      "entropy": 6.065555238723755,
      "epoch": 3.488158699641702,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038147145718042626,
      "loss": 5.2798,
      "mean_token_accuracy": 0.20726739913225173,
      "num_tokens": 80879347.0,
      "step": 39915
    },
    {
      "entropy": 6.023565149307251,
      "epoch": 3.488595647994407,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000381444004741889,
      "loss": 5.1938,
      "mean_token_accuracy": 0.21844591349363326,
      "num_tokens": 80889819.0,
      "step": 39920
    },
    {
      "entropy": 6.053825187683105,
      "epoch": 3.4890325963471116,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003814165502617,
      "loss": 5.2949,
      "mean_token_accuracy": 0.20400730073451995,
      "num_tokens": 80900576.0,
      "step": 39925
    },
    {
      "entropy": 6.067576789855957,
      "epoch": 3.4894695446998165,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003813890937403858,
      "loss": 5.2178,
      "mean_token_accuracy": 0.21776459217071534,
      "num_tokens": 80910966.0,
      "step": 39930
    },
    {
      "entropy": 6.042565870285034,
      "epoch": 3.489906493052521,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038136163517847306,
      "loss": 5.2177,
      "mean_token_accuracy": 0.2016557663679123,
      "num_tokens": 80920533.0,
      "step": 39935
    },
    {
      "entropy": 6.024933290481568,
      "epoch": 3.490343441405226,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038133417457648835,
      "loss": 5.264,
      "mean_token_accuracy": 0.20718182027339935,
      "num_tokens": 80930989.0,
      "step": 39940
    },
    {
      "entropy": 6.049962091445923,
      "epoch": 3.490780389757931,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038130671193495845,
      "loss": 5.2636,
      "mean_token_accuracy": 0.20509175211191177,
      "num_tokens": 80942283.0,
      "step": 39945
    },
    {
      "entropy": 6.106241893768311,
      "epoch": 3.4912173381106353,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038127924725441024,
      "loss": 5.2226,
      "mean_token_accuracy": 0.20781994313001634,
      "num_tokens": 80953258.0,
      "step": 39950
    },
    {
      "entropy": 6.001414489746094,
      "epoch": 3.49165428646334,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003812517805353703,
      "loss": 5.1682,
      "mean_token_accuracy": 0.21585290729999543,
      "num_tokens": 80963748.0,
      "step": 39955
    },
    {
      "entropy": 6.005272388458252,
      "epoch": 3.4920912348160447,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003812243117783656,
      "loss": 5.2504,
      "mean_token_accuracy": 0.20284632891416549,
      "num_tokens": 80973677.0,
      "step": 39960
    },
    {
      "entropy": 6.040537643432617,
      "epoch": 3.4925281831687496,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00038119684098392284,
      "loss": 5.252,
      "mean_token_accuracy": 0.20850797295570372,
      "num_tokens": 80984021.0,
      "step": 39965
    },
    {
      "entropy": 5.9734625816345215,
      "epoch": 3.492965131521454,
      "grad_norm": 1.328125,
      "learning_rate": 0.000381169368152569,
      "loss": 5.0902,
      "mean_token_accuracy": 0.21770086288452148,
      "num_tokens": 80993498.0,
      "step": 39970
    },
    {
      "entropy": 6.068793869018554,
      "epoch": 3.493402079874159,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00038114189328483124,
      "loss": 5.2944,
      "mean_token_accuracy": 0.20281007885932922,
      "num_tokens": 81002693.0,
      "step": 39975
    },
    {
      "entropy": 6.066791963577271,
      "epoch": 3.4938390282268634,
      "grad_norm": 1.375,
      "learning_rate": 0.0003811144163812362,
      "loss": 5.2875,
      "mean_token_accuracy": 0.19945817738771437,
      "num_tokens": 81012715.0,
      "step": 39980
    },
    {
      "entropy": 6.064477205276489,
      "epoch": 3.4942759765795683,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00038108693744231105,
      "loss": 5.207,
      "mean_token_accuracy": 0.21333980113267897,
      "num_tokens": 81023779.0,
      "step": 39985
    },
    {
      "entropy": 6.000307607650757,
      "epoch": 3.494712924932273,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003810594564685829,
      "loss": 5.2095,
      "mean_token_accuracy": 0.22100594490766526,
      "num_tokens": 81035103.0,
      "step": 39990
    },
    {
      "entropy": 6.0325822830200195,
      "epoch": 3.4951498732849777,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003810319734605787,
      "loss": 5.25,
      "mean_token_accuracy": 0.2075882837176323,
      "num_tokens": 81045011.0,
      "step": 39995
    },
    {
      "entropy": 6.021243143081665,
      "epoch": 3.4955868216376826,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003810044884188258,
      "loss": 5.2295,
      "mean_token_accuracy": 0.20943610966205597,
      "num_tokens": 81055841.0,
      "step": 40000
    },
    {
      "entropy": 6.0264232635498045,
      "epoch": 3.496023769990387,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003809770013438512,
      "loss": 5.2522,
      "mean_token_accuracy": 0.20588289350271224,
      "num_tokens": 81067270.0,
      "step": 40005
    },
    {
      "entropy": 6.058081340789795,
      "epoch": 3.496460718343092,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038094951223618224,
      "loss": 5.2296,
      "mean_token_accuracy": 0.21340889632701873,
      "num_tokens": 81076963.0,
      "step": 40010
    },
    {
      "entropy": 6.059189748764038,
      "epoch": 3.4968976666957965,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038092202109634603,
      "loss": 5.2138,
      "mean_token_accuracy": 0.21210617423057557,
      "num_tokens": 81086146.0,
      "step": 40015
    },
    {
      "entropy": 6.0774072170257565,
      "epoch": 3.4973346150485014,
      "grad_norm": 1.359375,
      "learning_rate": 0.00038089452792487005,
      "loss": 5.2269,
      "mean_token_accuracy": 0.21111125946044923,
      "num_tokens": 81095924.0,
      "step": 40020
    },
    {
      "entropy": 5.981855344772339,
      "epoch": 3.497771563401206,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003808670327222815,
      "loss": 5.2171,
      "mean_token_accuracy": 0.20716220438480376,
      "num_tokens": 81106396.0,
      "step": 40025
    },
    {
      "entropy": 6.031017398834228,
      "epoch": 3.498208511753911,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00038083953548910776,
      "loss": 5.1751,
      "mean_token_accuracy": 0.21003814041614532,
      "num_tokens": 81117098.0,
      "step": 40030
    },
    {
      "entropy": 5.991217803955078,
      "epoch": 3.4986454601066153,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038081203622587637,
      "loss": 5.1836,
      "mean_token_accuracy": 0.20959499776363372,
      "num_tokens": 81127335.0,
      "step": 40035
    },
    {
      "entropy": 6.030367755889893,
      "epoch": 3.49908240845932,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038078453493311463,
      "loss": 5.2207,
      "mean_token_accuracy": 0.20722175240516663,
      "num_tokens": 81136608.0,
      "step": 40040
    },
    {
      "entropy": 6.100170707702636,
      "epoch": 3.4995193568120246,
      "grad_norm": 1.359375,
      "learning_rate": 0.00038075703161135,
      "loss": 5.3097,
      "mean_token_accuracy": 0.20018937289714814,
      "num_tokens": 81146690.0,
      "step": 40045
    },
    {
      "entropy": 6.028496217727661,
      "epoch": 3.4999563051647296,
      "grad_norm": 1.375,
      "learning_rate": 0.00038072952626111016,
      "loss": 5.158,
      "mean_token_accuracy": 0.21487930417060852,
      "num_tokens": 81155914.0,
      "step": 40050
    },
    {
      "entropy": 6.012443208694458,
      "epoch": 3.5003932535174345,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003807020188829226,
      "loss": 5.2046,
      "mean_token_accuracy": 0.21326897144317628,
      "num_tokens": 81166537.0,
      "step": 40055
    },
    {
      "entropy": 6.022251415252685,
      "epoch": 3.500830201870139,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003806745094773149,
      "loss": 5.2002,
      "mean_token_accuracy": 0.21527991592884063,
      "num_tokens": 81176328.0,
      "step": 40060
    },
    {
      "entropy": 6.05816330909729,
      "epoch": 3.5012671502228434,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038064699804481476,
      "loss": 5.2492,
      "mean_token_accuracy": 0.2061912402510643,
      "num_tokens": 81186345.0,
      "step": 40065
    },
    {
      "entropy": 6.03779559135437,
      "epoch": 3.5017040985755483,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003806194845859499,
      "loss": 5.2261,
      "mean_token_accuracy": 0.20590917766094208,
      "num_tokens": 81196683.0,
      "step": 40070
    },
    {
      "entropy": 6.015731382369995,
      "epoch": 3.5021410469282532,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003805919691012479,
      "loss": 5.2178,
      "mean_token_accuracy": 0.2092379793524742,
      "num_tokens": 81206381.0,
      "step": 40075
    },
    {
      "entropy": 6.03829255104065,
      "epoch": 3.5025779952809577,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038056445159123664,
      "loss": 5.296,
      "mean_token_accuracy": 0.20730481445789337,
      "num_tokens": 81217056.0,
      "step": 40080
    },
    {
      "entropy": 6.0144203186035154,
      "epoch": 3.5030149436336626,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003805369320564438,
      "loss": 5.2327,
      "mean_token_accuracy": 0.20746314376592637,
      "num_tokens": 81227803.0,
      "step": 40085
    },
    {
      "entropy": 6.031420183181763,
      "epoch": 3.503451891986367,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003805094104973974,
      "loss": 5.2091,
      "mean_token_accuracy": 0.20999953001737595,
      "num_tokens": 81237680.0,
      "step": 40090
    },
    {
      "entropy": 6.007303857803345,
      "epoch": 3.503888840339072,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003804818869146252,
      "loss": 5.2238,
      "mean_token_accuracy": 0.21503227055072785,
      "num_tokens": 81247624.0,
      "step": 40095
    },
    {
      "entropy": 6.043392753601074,
      "epoch": 3.5043257886917765,
      "grad_norm": 1.34375,
      "learning_rate": 0.00038045436130865515,
      "loss": 5.2135,
      "mean_token_accuracy": 0.2142631843686104,
      "num_tokens": 81257932.0,
      "step": 40100
    },
    {
      "entropy": 6.094358777999878,
      "epoch": 3.5047627370444814,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003804268336800151,
      "loss": 5.2512,
      "mean_token_accuracy": 0.20974400341510774,
      "num_tokens": 81268595.0,
      "step": 40105
    },
    {
      "entropy": 6.046214008331299,
      "epoch": 3.5051996853971863,
      "grad_norm": 1.4375,
      "learning_rate": 0.00038039930402923323,
      "loss": 5.1815,
      "mean_token_accuracy": 0.21713604480028154,
      "num_tokens": 81279252.0,
      "step": 40110
    },
    {
      "entropy": 5.978865385055542,
      "epoch": 3.5056366337498908,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038037177235683746,
      "loss": 5.1999,
      "mean_token_accuracy": 0.21008172035217285,
      "num_tokens": 81289897.0,
      "step": 40115
    },
    {
      "entropy": 6.001632356643677,
      "epoch": 3.5060735821025952,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038034423866335586,
      "loss": 5.216,
      "mean_token_accuracy": 0.21124597042798995,
      "num_tokens": 81300297.0,
      "step": 40120
    },
    {
      "entropy": 6.0927294254302975,
      "epoch": 3.5065105304553,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003803167029493166,
      "loss": 5.2398,
      "mean_token_accuracy": 0.20452054888010024,
      "num_tokens": 81309628.0,
      "step": 40125
    },
    {
      "entropy": 5.984418344497681,
      "epoch": 3.506947478808005,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038028916521524775,
      "loss": 5.1888,
      "mean_token_accuracy": 0.21257562935352325,
      "num_tokens": 81320511.0,
      "step": 40130
    },
    {
      "entropy": 5.999098062515259,
      "epoch": 3.5073844271607095,
      "grad_norm": 1.25,
      "learning_rate": 0.0003802616254616776,
      "loss": 5.216,
      "mean_token_accuracy": 0.21623651832342147,
      "num_tokens": 81330982.0,
      "step": 40135
    },
    {
      "entropy": 6.031703662872315,
      "epoch": 3.5078213755134144,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003802340836891343,
      "loss": 5.2385,
      "mean_token_accuracy": 0.2113550677895546,
      "num_tokens": 81339955.0,
      "step": 40140
    },
    {
      "entropy": 6.012168836593628,
      "epoch": 3.508258323866119,
      "grad_norm": 1.375,
      "learning_rate": 0.00038020653989814606,
      "loss": 5.2239,
      "mean_token_accuracy": 0.20623355805873872,
      "num_tokens": 81349151.0,
      "step": 40145
    },
    {
      "entropy": 6.0418158054351805,
      "epoch": 3.508695272218824,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003801789940892413,
      "loss": 5.1401,
      "mean_token_accuracy": 0.21586533933877944,
      "num_tokens": 81359563.0,
      "step": 40150
    },
    {
      "entropy": 6.054126214981079,
      "epoch": 3.5091322205715283,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003801514462629483,
      "loss": 5.2741,
      "mean_token_accuracy": 0.20877234637737274,
      "num_tokens": 81369669.0,
      "step": 40155
    },
    {
      "entropy": 6.021456384658814,
      "epoch": 3.509569168924233,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00038012389641979555,
      "loss": 5.2957,
      "mean_token_accuracy": 0.20438686460256578,
      "num_tokens": 81380322.0,
      "step": 40160
    },
    {
      "entropy": 6.065455961227417,
      "epoch": 3.510006117276938,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003800963445603114,
      "loss": 5.2673,
      "mean_token_accuracy": 0.20782506763935088,
      "num_tokens": 81389871.0,
      "step": 40165
    },
    {
      "entropy": 5.995428133010864,
      "epoch": 3.5104430656296426,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003800687906850242,
      "loss": 5.1379,
      "mean_token_accuracy": 0.21543887108564377,
      "num_tokens": 81399684.0,
      "step": 40170
    },
    {
      "entropy": 6.0511332035064695,
      "epoch": 3.510880013982347,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003800412347944626,
      "loss": 5.2329,
      "mean_token_accuracy": 0.2069246754050255,
      "num_tokens": 81409638.0,
      "step": 40175
    },
    {
      "entropy": 6.085084581375122,
      "epoch": 3.511316962335052,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003800136768891549,
      "loss": 5.2662,
      "mean_token_accuracy": 0.20820218026638032,
      "num_tokens": 81419963.0,
      "step": 40180
    },
    {
      "entropy": 6.012609004974365,
      "epoch": 3.511753910687757,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037998611696963,
      "loss": 5.1831,
      "mean_token_accuracy": 0.20677604228258134,
      "num_tokens": 81429747.0,
      "step": 40185
    },
    {
      "entropy": 6.037196063995362,
      "epoch": 3.5121908590404614,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003799585550364164,
      "loss": 5.1732,
      "mean_token_accuracy": 0.20550701916217803,
      "num_tokens": 81439012.0,
      "step": 40190
    },
    {
      "entropy": 6.065476560592652,
      "epoch": 3.5126278073931663,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003799309910900427,
      "loss": 5.2754,
      "mean_token_accuracy": 0.20940798223018647,
      "num_tokens": 81449713.0,
      "step": 40195
    },
    {
      "entropy": 5.992035961151123,
      "epoch": 3.5130647557458707,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003799034251310376,
      "loss": 5.1747,
      "mean_token_accuracy": 0.2159268319606781,
      "num_tokens": 81459326.0,
      "step": 40200
    },
    {
      "entropy": 6.028534412384033,
      "epoch": 3.5135017040985757,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037987585715992974,
      "loss": 5.1979,
      "mean_token_accuracy": 0.21682825386524202,
      "num_tokens": 81469419.0,
      "step": 40205
    },
    {
      "entropy": 6.0449381351470945,
      "epoch": 3.51393865245128,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037984828717724814,
      "loss": 5.2077,
      "mean_token_accuracy": 0.21687492430210115,
      "num_tokens": 81478494.0,
      "step": 40210
    },
    {
      "entropy": 6.000262117385864,
      "epoch": 3.514375600803985,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037982071518352135,
      "loss": 5.2124,
      "mean_token_accuracy": 0.21386703997850418,
      "num_tokens": 81488080.0,
      "step": 40215
    },
    {
      "entropy": 6.040250253677368,
      "epoch": 3.51481254915669,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003797931411792783,
      "loss": 5.2278,
      "mean_token_accuracy": 0.21243699938058852,
      "num_tokens": 81498075.0,
      "step": 40220
    },
    {
      "entropy": 6.066267204284668,
      "epoch": 3.5152494975093944,
      "grad_norm": 1.34375,
      "learning_rate": 0.00037976556516504796,
      "loss": 5.2383,
      "mean_token_accuracy": 0.207340769469738,
      "num_tokens": 81508407.0,
      "step": 40225
    },
    {
      "entropy": 6.1114959716796875,
      "epoch": 3.515686445862099,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037973798714135916,
      "loss": 5.218,
      "mean_token_accuracy": 0.20366239696741104,
      "num_tokens": 81518009.0,
      "step": 40230
    },
    {
      "entropy": 6.002516508102417,
      "epoch": 3.516123394214804,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003797104071087409,
      "loss": 5.2273,
      "mean_token_accuracy": 0.21614059805870056,
      "num_tokens": 81527468.0,
      "step": 40235
    },
    {
      "entropy": 6.026148080825806,
      "epoch": 3.5165603425675087,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003796828250677221,
      "loss": 5.2354,
      "mean_token_accuracy": 0.21434611976146697,
      "num_tokens": 81538133.0,
      "step": 40240
    },
    {
      "entropy": 5.98763632774353,
      "epoch": 3.516997290920213,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037965524101883196,
      "loss": 5.1681,
      "mean_token_accuracy": 0.21557241380214692,
      "num_tokens": 81548051.0,
      "step": 40245
    },
    {
      "entropy": 6.052944278717041,
      "epoch": 3.517434239272918,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003796276549625994,
      "loss": 5.2597,
      "mean_token_accuracy": 0.212579545378685,
      "num_tokens": 81557984.0,
      "step": 40250
    },
    {
      "entropy": 6.028860425949096,
      "epoch": 3.5178711876256226,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037960006689955356,
      "loss": 5.2223,
      "mean_token_accuracy": 0.2139533519744873,
      "num_tokens": 81568587.0,
      "step": 40255
    },
    {
      "entropy": 6.042753839492798,
      "epoch": 3.5183081359783275,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037957247683022375,
      "loss": 5.215,
      "mean_token_accuracy": 0.2061086192727089,
      "num_tokens": 81578085.0,
      "step": 40260
    },
    {
      "entropy": 6.079487037658692,
      "epoch": 3.518745084331032,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003795448847551389,
      "loss": 5.3393,
      "mean_token_accuracy": 0.19915023148059846,
      "num_tokens": 81588037.0,
      "step": 40265
    },
    {
      "entropy": 6.012262964248658,
      "epoch": 3.519182032683737,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037951729067482845,
      "loss": 5.2424,
      "mean_token_accuracy": 0.2117660090327263,
      "num_tokens": 81598364.0,
      "step": 40270
    },
    {
      "entropy": 6.054382944107056,
      "epoch": 3.5196189810364413,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037948969458982153,
      "loss": 5.2036,
      "mean_token_accuracy": 0.2033603236079216,
      "num_tokens": 81609428.0,
      "step": 40275
    },
    {
      "entropy": 6.020451641082763,
      "epoch": 3.5200559293891462,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037946209650064754,
      "loss": 5.2476,
      "mean_token_accuracy": 0.2024250090122223,
      "num_tokens": 81619605.0,
      "step": 40280
    },
    {
      "entropy": 6.069011306762695,
      "epoch": 3.5204928777418507,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003794344964078359,
      "loss": 5.2683,
      "mean_token_accuracy": 0.2011964052915573,
      "num_tokens": 81628987.0,
      "step": 40285
    },
    {
      "entropy": 6.081162643432617,
      "epoch": 3.5209298260945556,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003794068943119159,
      "loss": 5.2091,
      "mean_token_accuracy": 0.21629922091960907,
      "num_tokens": 81637628.0,
      "step": 40290
    },
    {
      "entropy": 5.984345293045044,
      "epoch": 3.5213667744472605,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003793792902134168,
      "loss": 5.159,
      "mean_token_accuracy": 0.2156460702419281,
      "num_tokens": 81647470.0,
      "step": 40295
    },
    {
      "entropy": 6.004925155639649,
      "epoch": 3.521803722799965,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003793516841128683,
      "loss": 5.2007,
      "mean_token_accuracy": 0.20877428948879242,
      "num_tokens": 81657882.0,
      "step": 40300
    },
    {
      "entropy": 5.977825450897217,
      "epoch": 3.5222406711526695,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037932407601079987,
      "loss": 5.2489,
      "mean_token_accuracy": 0.21039064079523087,
      "num_tokens": 81668506.0,
      "step": 40305
    },
    {
      "entropy": 6.0440386772155765,
      "epoch": 3.5226776195053744,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037929646590774093,
      "loss": 5.2506,
      "mean_token_accuracy": 0.20575089603662491,
      "num_tokens": 81678590.0,
      "step": 40310
    },
    {
      "entropy": 6.035439157485962,
      "epoch": 3.5231145678580793,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003792688538042211,
      "loss": 5.2458,
      "mean_token_accuracy": 0.22019070088863374,
      "num_tokens": 81689625.0,
      "step": 40315
    },
    {
      "entropy": 6.075700378417968,
      "epoch": 3.5235515162107838,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003792412397007701,
      "loss": 5.2904,
      "mean_token_accuracy": 0.2049048885703087,
      "num_tokens": 81699251.0,
      "step": 40320
    },
    {
      "entropy": 6.061176729202271,
      "epoch": 3.5239884645634887,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037921362359791743,
      "loss": 5.1676,
      "mean_token_accuracy": 0.21942429840564728,
      "num_tokens": 81708851.0,
      "step": 40325
    },
    {
      "entropy": 6.0473175048828125,
      "epoch": 3.524425412916193,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003791860054961929,
      "loss": 5.2444,
      "mean_token_accuracy": 0.20514658838510513,
      "num_tokens": 81718399.0,
      "step": 40330
    },
    {
      "entropy": 6.074798250198365,
      "epoch": 3.524862361268898,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00037915838539612624,
      "loss": 5.2741,
      "mean_token_accuracy": 0.21205427646636962,
      "num_tokens": 81728546.0,
      "step": 40335
    },
    {
      "entropy": 6.097346639633178,
      "epoch": 3.5252993096216025,
      "grad_norm": 1.359375,
      "learning_rate": 0.000379130763298247,
      "loss": 5.2041,
      "mean_token_accuracy": 0.21445220708847046,
      "num_tokens": 81737830.0,
      "step": 40340
    },
    {
      "entropy": 6.017520904541016,
      "epoch": 3.5257362579743075,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003791031392030853,
      "loss": 5.1596,
      "mean_token_accuracy": 0.21875840276479722,
      "num_tokens": 81747534.0,
      "step": 40345
    },
    {
      "entropy": 6.002628803253174,
      "epoch": 3.5261732063270124,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037907551311117083,
      "loss": 5.2545,
      "mean_token_accuracy": 0.20742385536432267,
      "num_tokens": 81758411.0,
      "step": 40350
    },
    {
      "entropy": 5.991137456893921,
      "epoch": 3.526610154679717,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003790478850230335,
      "loss": 5.2245,
      "mean_token_accuracy": 0.20718123614788056,
      "num_tokens": 81768441.0,
      "step": 40355
    },
    {
      "entropy": 6.057234239578247,
      "epoch": 3.5270471030324213,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037902025493920314,
      "loss": 5.2889,
      "mean_token_accuracy": 0.20650531500577926,
      "num_tokens": 81778899.0,
      "step": 40360
    },
    {
      "entropy": 6.033720588684082,
      "epoch": 3.527484051385126,
      "grad_norm": 1.25,
      "learning_rate": 0.00037899262286020984,
      "loss": 5.2147,
      "mean_token_accuracy": 0.21073468774557114,
      "num_tokens": 81788643.0,
      "step": 40365
    },
    {
      "entropy": 6.049852609634399,
      "epoch": 3.527920999737831,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037896498878658345,
      "loss": 5.239,
      "mean_token_accuracy": 0.20660446733236312,
      "num_tokens": 81798781.0,
      "step": 40370
    },
    {
      "entropy": 6.023949241638183,
      "epoch": 3.5283579480905356,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003789373527188542,
      "loss": 5.2308,
      "mean_token_accuracy": 0.21213611215353012,
      "num_tokens": 81808402.0,
      "step": 40375
    },
    {
      "entropy": 6.0315100193023685,
      "epoch": 3.5287948964432405,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003789097146575521,
      "loss": 5.2569,
      "mean_token_accuracy": 0.2081408068537712,
      "num_tokens": 81819285.0,
      "step": 40380
    },
    {
      "entropy": 6.019674396514892,
      "epoch": 3.529231844795945,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003788820746032071,
      "loss": 5.1606,
      "mean_token_accuracy": 0.21258071959018707,
      "num_tokens": 81829357.0,
      "step": 40385
    },
    {
      "entropy": 6.028132581710816,
      "epoch": 3.52966879314865,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00037885443255634955,
      "loss": 5.1588,
      "mean_token_accuracy": 0.2219076782464981,
      "num_tokens": 81838547.0,
      "step": 40390
    },
    {
      "entropy": 6.011040258407593,
      "epoch": 3.5301057415013544,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003788267885175095,
      "loss": 5.2206,
      "mean_token_accuracy": 0.2112848162651062,
      "num_tokens": 81849286.0,
      "step": 40395
    },
    {
      "entropy": 6.122584009170533,
      "epoch": 3.5305426898540593,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037879914248721737,
      "loss": 5.3087,
      "mean_token_accuracy": 0.1972619414329529,
      "num_tokens": 81859599.0,
      "step": 40400
    },
    {
      "entropy": 6.046698760986328,
      "epoch": 3.530979638206764,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003787714944660032,
      "loss": 5.2006,
      "mean_token_accuracy": 0.2167336955666542,
      "num_tokens": 81869497.0,
      "step": 40405
    },
    {
      "entropy": 6.016835308074951,
      "epoch": 3.5314165865594687,
      "grad_norm": 1.40625,
      "learning_rate": 0.00037874384445439743,
      "loss": 5.1492,
      "mean_token_accuracy": 0.2183080419898033,
      "num_tokens": 81878502.0,
      "step": 40410
    },
    {
      "entropy": 6.0061869621276855,
      "epoch": 3.531853534912173,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003787161924529303,
      "loss": 5.1683,
      "mean_token_accuracy": 0.21763334572315216,
      "num_tokens": 81888748.0,
      "step": 40415
    },
    {
      "entropy": 6.01850118637085,
      "epoch": 3.532290483264878,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003786885384621323,
      "loss": 5.1593,
      "mean_token_accuracy": 0.22318885326385499,
      "num_tokens": 81897705.0,
      "step": 40420
    },
    {
      "entropy": 6.007025909423828,
      "epoch": 3.532727431617583,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003786608824825338,
      "loss": 5.2621,
      "mean_token_accuracy": 0.20981538891792298,
      "num_tokens": 81909294.0,
      "step": 40425
    },
    {
      "entropy": 6.1078184127807615,
      "epoch": 3.5331643799702874,
      "grad_norm": 1.34375,
      "learning_rate": 0.00037863322451466523,
      "loss": 5.3254,
      "mean_token_accuracy": 0.19477395564317704,
      "num_tokens": 81919191.0,
      "step": 40430
    },
    {
      "entropy": 6.022183799743653,
      "epoch": 3.5336013283229923,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003786055645590571,
      "loss": 5.1531,
      "mean_token_accuracy": 0.21855522841215133,
      "num_tokens": 81928660.0,
      "step": 40435
    },
    {
      "entropy": 5.996410703659057,
      "epoch": 3.534038276675697,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037857790261624006,
      "loss": 5.2059,
      "mean_token_accuracy": 0.21895359754562377,
      "num_tokens": 81938626.0,
      "step": 40440
    },
    {
      "entropy": 5.990815448760986,
      "epoch": 3.5344752250284017,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003785502386867445,
      "loss": 5.1984,
      "mean_token_accuracy": 0.22166125774383544,
      "num_tokens": 81948304.0,
      "step": 40445
    },
    {
      "entropy": 6.001937770843506,
      "epoch": 3.534912173381106,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037852257277110106,
      "loss": 5.2128,
      "mean_token_accuracy": 0.21638509929180144,
      "num_tokens": 81958455.0,
      "step": 40450
    },
    {
      "entropy": 5.944555997848511,
      "epoch": 3.535349121733811,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003784949048698405,
      "loss": 5.1063,
      "mean_token_accuracy": 0.22675673365592958,
      "num_tokens": 81967809.0,
      "step": 40455
    },
    {
      "entropy": 6.010738897323608,
      "epoch": 3.535786070086516,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003784672349834934,
      "loss": 5.1795,
      "mean_token_accuracy": 0.21824503391981126,
      "num_tokens": 81979801.0,
      "step": 40460
    },
    {
      "entropy": 6.011314964294433,
      "epoch": 3.5362230184392205,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037843956311259047,
      "loss": 5.219,
      "mean_token_accuracy": 0.20903271585702896,
      "num_tokens": 81989591.0,
      "step": 40465
    },
    {
      "entropy": 6.049335241317749,
      "epoch": 3.536659966791925,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003784118892576626,
      "loss": 5.2119,
      "mean_token_accuracy": 0.2113284096121788,
      "num_tokens": 81999470.0,
      "step": 40470
    },
    {
      "entropy": 6.006860256195068,
      "epoch": 3.53709691514463,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037838421341924045,
      "loss": 5.2121,
      "mean_token_accuracy": 0.2122958108782768,
      "num_tokens": 82010307.0,
      "step": 40475
    },
    {
      "entropy": 5.969148397445679,
      "epoch": 3.537533863497335,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037835653559785496,
      "loss": 5.1414,
      "mean_token_accuracy": 0.21904342770576476,
      "num_tokens": 82019997.0,
      "step": 40480
    },
    {
      "entropy": 6.018884754180908,
      "epoch": 3.5379708118500393,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003783288557940369,
      "loss": 5.1484,
      "mean_token_accuracy": 0.21364155411720276,
      "num_tokens": 82030328.0,
      "step": 40485
    },
    {
      "entropy": 6.111373329162598,
      "epoch": 3.538407760202744,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003783011740083173,
      "loss": 5.2895,
      "mean_token_accuracy": 0.2056007832288742,
      "num_tokens": 82041153.0,
      "step": 40490
    },
    {
      "entropy": 5.929337358474731,
      "epoch": 3.5388447085554486,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0003782734902412271,
      "loss": 5.0946,
      "mean_token_accuracy": 0.22945104241371156,
      "num_tokens": 82050962.0,
      "step": 40495
    },
    {
      "entropy": 6.001006603240967,
      "epoch": 3.5392816569081536,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037824580449329716,
      "loss": 5.2315,
      "mean_token_accuracy": 0.21354152113199235,
      "num_tokens": 82061033.0,
      "step": 40500
    },
    {
      "entropy": 6.064061307907105,
      "epoch": 3.539718605260858,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037821811676505855,
      "loss": 5.2151,
      "mean_token_accuracy": 0.20935538709163665,
      "num_tokens": 82070352.0,
      "step": 40505
    },
    {
      "entropy": 6.052872228622436,
      "epoch": 3.540155553613563,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037819042705704244,
      "loss": 5.2251,
      "mean_token_accuracy": 0.21116356253623964,
      "num_tokens": 82080608.0,
      "step": 40510
    },
    {
      "entropy": 6.017598009109497,
      "epoch": 3.540592501966268,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037816273536977986,
      "loss": 5.1748,
      "mean_token_accuracy": 0.2151909813284874,
      "num_tokens": 82089935.0,
      "step": 40515
    },
    {
      "entropy": 6.0096508979797365,
      "epoch": 3.5410294503189723,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000378135041703802,
      "loss": 5.2052,
      "mean_token_accuracy": 0.21867201328277588,
      "num_tokens": 82100983.0,
      "step": 40520
    },
    {
      "entropy": 6.085442352294922,
      "epoch": 3.541466398671677,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037810734605963996,
      "loss": 5.3071,
      "mean_token_accuracy": 0.2012101948261261,
      "num_tokens": 82110791.0,
      "step": 40525
    },
    {
      "entropy": 6.047246217727661,
      "epoch": 3.5419033470243817,
      "grad_norm": 1.265625,
      "learning_rate": 0.000378079648437825,
      "loss": 5.1897,
      "mean_token_accuracy": 0.21032959818840027,
      "num_tokens": 82120330.0,
      "step": 40530
    },
    {
      "entropy": 6.116153717041016,
      "epoch": 3.5423402953770866,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037805194883888834,
      "loss": 5.2505,
      "mean_token_accuracy": 0.20602623224258423,
      "num_tokens": 82130589.0,
      "step": 40535
    },
    {
      "entropy": 6.010341930389404,
      "epoch": 3.542777243729791,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003780242472633614,
      "loss": 5.1894,
      "mean_token_accuracy": 0.21771024018526078,
      "num_tokens": 82141421.0,
      "step": 40540
    },
    {
      "entropy": 6.007891702651977,
      "epoch": 3.543214192082496,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003779965437117753,
      "loss": 5.2418,
      "mean_token_accuracy": 0.20855921357870102,
      "num_tokens": 82150688.0,
      "step": 40545
    },
    {
      "entropy": 6.106721448898315,
      "epoch": 3.5436511404352005,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037796883818466155,
      "loss": 5.2705,
      "mean_token_accuracy": 0.20472936183214188,
      "num_tokens": 82161850.0,
      "step": 40550
    },
    {
      "entropy": 6.103276157379151,
      "epoch": 3.5440880887879054,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037794113068255155,
      "loss": 5.2092,
      "mean_token_accuracy": 0.20506636202335357,
      "num_tokens": 82171492.0,
      "step": 40555
    },
    {
      "entropy": 6.040407133102417,
      "epoch": 3.54452503714061,
      "grad_norm": 1.4375,
      "learning_rate": 0.00037791342120597674,
      "loss": 5.27,
      "mean_token_accuracy": 0.20941649228334427,
      "num_tokens": 82182469.0,
      "step": 40560
    },
    {
      "entropy": 6.042173719406128,
      "epoch": 3.5449619854933148,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037788570975546854,
      "loss": 5.2045,
      "mean_token_accuracy": 0.21206069737672806,
      "num_tokens": 82192395.0,
      "step": 40565
    },
    {
      "entropy": 6.039700841903686,
      "epoch": 3.5453989338460197,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00037785799633155854,
      "loss": 5.2271,
      "mean_token_accuracy": 0.20581251680850982,
      "num_tokens": 82201555.0,
      "step": 40570
    },
    {
      "entropy": 6.022734594345093,
      "epoch": 3.545835882198724,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003778302809347782,
      "loss": 5.2184,
      "mean_token_accuracy": 0.21514514535665513,
      "num_tokens": 82211013.0,
      "step": 40575
    },
    {
      "entropy": 6.073615598678589,
      "epoch": 3.5462728305514286,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003778025635656593,
      "loss": 5.2598,
      "mean_token_accuracy": 0.20442874431610109,
      "num_tokens": 82221317.0,
      "step": 40580
    },
    {
      "entropy": 6.05484094619751,
      "epoch": 3.5467097789041335,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003777748442247333,
      "loss": 5.1756,
      "mean_token_accuracy": 0.2146112009882927,
      "num_tokens": 82230837.0,
      "step": 40585
    },
    {
      "entropy": 6.060286664962769,
      "epoch": 3.5471467272568384,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000377747122912532,
      "loss": 5.2658,
      "mean_token_accuracy": 0.20406726002693176,
      "num_tokens": 82241034.0,
      "step": 40590
    },
    {
      "entropy": 5.986660051345825,
      "epoch": 3.547583675609543,
      "grad_norm": 1.359375,
      "learning_rate": 0.000377719399629587,
      "loss": 5.2166,
      "mean_token_accuracy": 0.2108546957373619,
      "num_tokens": 82251047.0,
      "step": 40595
    },
    {
      "entropy": 6.097045373916626,
      "epoch": 3.548020623962248,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003776916743764301,
      "loss": 5.2295,
      "mean_token_accuracy": 0.20464534312486649,
      "num_tokens": 82261215.0,
      "step": 40600
    },
    {
      "entropy": 6.091945791244507,
      "epoch": 3.5484575723149523,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003776639471535931,
      "loss": 5.2529,
      "mean_token_accuracy": 0.2075262188911438,
      "num_tokens": 82271527.0,
      "step": 40605
    },
    {
      "entropy": 6.121823501586914,
      "epoch": 3.548894520667657,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003776362179616077,
      "loss": 5.3367,
      "mean_token_accuracy": 0.1974189355969429,
      "num_tokens": 82282021.0,
      "step": 40610
    },
    {
      "entropy": 6.090439891815185,
      "epoch": 3.5493314690203617,
      "grad_norm": 1.421875,
      "learning_rate": 0.00037760848680100595,
      "loss": 5.3087,
      "mean_token_accuracy": 0.20244496315717697,
      "num_tokens": 82291279.0,
      "step": 40615
    },
    {
      "entropy": 6.030266237258911,
      "epoch": 3.5497684173730666,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003775807536723196,
      "loss": 5.1707,
      "mean_token_accuracy": 0.21593862026929855,
      "num_tokens": 82300756.0,
      "step": 40620
    },
    {
      "entropy": 6.06609001159668,
      "epoch": 3.550205365725771,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003775530185760807,
      "loss": 5.2592,
      "mean_token_accuracy": 0.20549847781658173,
      "num_tokens": 82309896.0,
      "step": 40625
    },
    {
      "entropy": 5.969900894165039,
      "epoch": 3.550642314078476,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003775252815128211,
      "loss": 5.0827,
      "mean_token_accuracy": 0.22428756952285767,
      "num_tokens": 82319646.0,
      "step": 40630
    },
    {
      "entropy": 6.005008220672607,
      "epoch": 3.5510792624311804,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003774975424830729,
      "loss": 5.2461,
      "mean_token_accuracy": 0.20982466042041778,
      "num_tokens": 82329492.0,
      "step": 40635
    },
    {
      "entropy": 6.007727718353271,
      "epoch": 3.5515162107838854,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003774698014873682,
      "loss": 5.2047,
      "mean_token_accuracy": 0.21624191403388976,
      "num_tokens": 82340339.0,
      "step": 40640
    },
    {
      "entropy": 6.0598348617553714,
      "epoch": 3.5519531591365903,
      "grad_norm": 1.421875,
      "learning_rate": 0.00037744205852623895,
      "loss": 5.2034,
      "mean_token_accuracy": 0.2113136902451515,
      "num_tokens": 82351337.0,
      "step": 40645
    },
    {
      "entropy": 6.010365438461304,
      "epoch": 3.5523901074892947,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003774143136002173,
      "loss": 5.212,
      "mean_token_accuracy": 0.21555558145046233,
      "num_tokens": 82361616.0,
      "step": 40650
    },
    {
      "entropy": 6.044261074066162,
      "epoch": 3.552827055841999,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00037738656670983553,
      "loss": 5.1931,
      "mean_token_accuracy": 0.20143261253833772,
      "num_tokens": 82371359.0,
      "step": 40655
    },
    {
      "entropy": 5.98810715675354,
      "epoch": 3.553264004194704,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003773588178556258,
      "loss": 5.1226,
      "mean_token_accuracy": 0.21227218210697174,
      "num_tokens": 82381575.0,
      "step": 40660
    },
    {
      "entropy": 5.965862131118774,
      "epoch": 3.553700952547409,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003773310670381202,
      "loss": 5.2036,
      "mean_token_accuracy": 0.21206628829240798,
      "num_tokens": 82392175.0,
      "step": 40665
    },
    {
      "entropy": 6.065467643737793,
      "epoch": 3.5541379009001135,
      "grad_norm": 1.40625,
      "learning_rate": 0.00037730331425785114,
      "loss": 5.2593,
      "mean_token_accuracy": 0.21023447066545486,
      "num_tokens": 82402570.0,
      "step": 40670
    },
    {
      "entropy": 6.068017625808716,
      "epoch": 3.5545748492528184,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037727555951535093,
      "loss": 5.235,
      "mean_token_accuracy": 0.2116488426923752,
      "num_tokens": 82413433.0,
      "step": 40675
    },
    {
      "entropy": 6.090798282623291,
      "epoch": 3.555011797605523,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003772478028111519,
      "loss": 5.2625,
      "mean_token_accuracy": 0.203978331387043,
      "num_tokens": 82423144.0,
      "step": 40680
    },
    {
      "entropy": 6.005889368057251,
      "epoch": 3.555448745958228,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003772200441457864,
      "loss": 5.1792,
      "mean_token_accuracy": 0.21011148542165756,
      "num_tokens": 82433096.0,
      "step": 40685
    },
    {
      "entropy": 5.945536422729492,
      "epoch": 3.5558856943109323,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003771922835197869,
      "loss": 5.1561,
      "mean_token_accuracy": 0.22056511640548707,
      "num_tokens": 82444261.0,
      "step": 40690
    },
    {
      "entropy": 6.062177658081055,
      "epoch": 3.556322642663637,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003771645209336859,
      "loss": 5.286,
      "mean_token_accuracy": 0.2059546872973442,
      "num_tokens": 82454442.0,
      "step": 40695
    },
    {
      "entropy": 6.063695287704467,
      "epoch": 3.556759591016342,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003771367563880158,
      "loss": 5.2641,
      "mean_token_accuracy": 0.21119585484266282,
      "num_tokens": 82464589.0,
      "step": 40700
    },
    {
      "entropy": 6.060222816467285,
      "epoch": 3.5571965393690466,
      "grad_norm": 1.34375,
      "learning_rate": 0.00037710898988330924,
      "loss": 5.2679,
      "mean_token_accuracy": 0.21220873445272445,
      "num_tokens": 82475624.0,
      "step": 40705
    },
    {
      "entropy": 6.084235811233521,
      "epoch": 3.557633487721751,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003770812214200987,
      "loss": 5.2522,
      "mean_token_accuracy": 0.2113378241658211,
      "num_tokens": 82485142.0,
      "step": 40710
    },
    {
      "entropy": 6.089968729019165,
      "epoch": 3.558070436074456,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00037705345099891684,
      "loss": 5.2703,
      "mean_token_accuracy": 0.20354234129190446,
      "num_tokens": 82495165.0,
      "step": 40715
    },
    {
      "entropy": 5.959680223464966,
      "epoch": 3.558507384427161,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003770256786202964,
      "loss": 5.1521,
      "mean_token_accuracy": 0.205178901553154,
      "num_tokens": 82505867.0,
      "step": 40720
    },
    {
      "entropy": 6.071346092224121,
      "epoch": 3.5589443327798653,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037699790428476996,
      "loss": 5.2934,
      "mean_token_accuracy": 0.20904046297073364,
      "num_tokens": 82517416.0,
      "step": 40725
    },
    {
      "entropy": 6.105616474151612,
      "epoch": 3.5593812811325702,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003769701279928702,
      "loss": 5.2408,
      "mean_token_accuracy": 0.21698802411556245,
      "num_tokens": 82527192.0,
      "step": 40730
    },
    {
      "entropy": 6.09373664855957,
      "epoch": 3.5598182294852747,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00037694234974512995,
      "loss": 5.252,
      "mean_token_accuracy": 0.21200584322214128,
      "num_tokens": 82536600.0,
      "step": 40735
    },
    {
      "entropy": 6.064014863967896,
      "epoch": 3.5602551778379796,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037691456954208207,
      "loss": 5.2875,
      "mean_token_accuracy": 0.20979925841093064,
      "num_tokens": 82546899.0,
      "step": 40740
    },
    {
      "entropy": 6.008672475814819,
      "epoch": 3.560692126190684,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003768867873842593,
      "loss": 5.2632,
      "mean_token_accuracy": 0.21098148226737976,
      "num_tokens": 82557884.0,
      "step": 40745
    },
    {
      "entropy": 6.017146968841553,
      "epoch": 3.561129074543389,
      "grad_norm": 1.25,
      "learning_rate": 0.0003768590032721946,
      "loss": 5.1258,
      "mean_token_accuracy": 0.22389915883541106,
      "num_tokens": 82567726.0,
      "step": 40750
    },
    {
      "entropy": 6.0714293956756595,
      "epoch": 3.561566022896094,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037683121720642087,
      "loss": 5.2958,
      "mean_token_accuracy": 0.20451807528734206,
      "num_tokens": 82577741.0,
      "step": 40755
    },
    {
      "entropy": 6.125195598602295,
      "epoch": 3.5620029712487984,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037680342918747103,
      "loss": 5.2919,
      "mean_token_accuracy": 0.20130906105041504,
      "num_tokens": 82588591.0,
      "step": 40760
    },
    {
      "entropy": 6.024574422836304,
      "epoch": 3.562439919601503,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000376775639215878,
      "loss": 5.2035,
      "mean_token_accuracy": 0.20473978221416472,
      "num_tokens": 82598276.0,
      "step": 40765
    },
    {
      "entropy": 5.987118768692016,
      "epoch": 3.5628768679542078,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037674784729217493,
      "loss": 5.2013,
      "mean_token_accuracy": 0.21342850476503372,
      "num_tokens": 82608715.0,
      "step": 40770
    },
    {
      "entropy": 6.031762886047363,
      "epoch": 3.5633138163069127,
      "grad_norm": 1.25,
      "learning_rate": 0.00037672005341689486,
      "loss": 5.2651,
      "mean_token_accuracy": 0.21112712025642394,
      "num_tokens": 82618854.0,
      "step": 40775
    },
    {
      "entropy": 6.035682106018067,
      "epoch": 3.563750764659617,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003766922575905708,
      "loss": 5.1907,
      "mean_token_accuracy": 0.21252616345882416,
      "num_tokens": 82628391.0,
      "step": 40780
    },
    {
      "entropy": 6.0398646831512455,
      "epoch": 3.564187713012322,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000376664459813736,
      "loss": 5.3377,
      "mean_token_accuracy": 0.20170420706272124,
      "num_tokens": 82639560.0,
      "step": 40785
    },
    {
      "entropy": 6.002126836776734,
      "epoch": 3.5646246613650265,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037663666008692356,
      "loss": 5.2006,
      "mean_token_accuracy": 0.2112230286002159,
      "num_tokens": 82650252.0,
      "step": 40790
    },
    {
      "entropy": 6.147091245651245,
      "epoch": 3.5650616097177315,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003766088584106667,
      "loss": 5.265,
      "mean_token_accuracy": 0.21006131917238235,
      "num_tokens": 82660035.0,
      "step": 40795
    },
    {
      "entropy": 6.0041975498199465,
      "epoch": 3.565498558070436,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037658105478549875,
      "loss": 5.1996,
      "mean_token_accuracy": 0.21245455294847487,
      "num_tokens": 82669435.0,
      "step": 40800
    },
    {
      "entropy": 6.083066177368164,
      "epoch": 3.565935506423141,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037655324921195295,
      "loss": 5.2422,
      "mean_token_accuracy": 0.2025392860174179,
      "num_tokens": 82679314.0,
      "step": 40805
    },
    {
      "entropy": 6.005801582336426,
      "epoch": 3.5663724547758457,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003765254416905626,
      "loss": 5.1753,
      "mean_token_accuracy": 0.21922339648008346,
      "num_tokens": 82690140.0,
      "step": 40810
    },
    {
      "entropy": 5.994083595275879,
      "epoch": 3.56680940312855,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000376497632221861,
      "loss": 5.2128,
      "mean_token_accuracy": 0.21217521876096726,
      "num_tokens": 82699210.0,
      "step": 40815
    },
    {
      "entropy": 6.079964208602905,
      "epoch": 3.5672463514812547,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003764698208063818,
      "loss": 5.3199,
      "mean_token_accuracy": 0.19945293366909028,
      "num_tokens": 82710398.0,
      "step": 40820
    },
    {
      "entropy": 6.112254476547241,
      "epoch": 3.5676832998339596,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003764420074446581,
      "loss": 5.2154,
      "mean_token_accuracy": 0.2082189217209816,
      "num_tokens": 82721036.0,
      "step": 40825
    },
    {
      "entropy": 6.05685396194458,
      "epoch": 3.5681202481866645,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003764141921372236,
      "loss": 5.2722,
      "mean_token_accuracy": 0.20990186780691147,
      "num_tokens": 82730884.0,
      "step": 40830
    },
    {
      "entropy": 6.031965208053589,
      "epoch": 3.568557196539369,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003763863748846118,
      "loss": 5.2044,
      "mean_token_accuracy": 0.21681686639785766,
      "num_tokens": 82740561.0,
      "step": 40835
    },
    {
      "entropy": 6.123776817321778,
      "epoch": 3.568994144892074,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037635855568735613,
      "loss": 5.2926,
      "mean_token_accuracy": 0.2087497442960739,
      "num_tokens": 82751064.0,
      "step": 40840
    },
    {
      "entropy": 6.053946924209595,
      "epoch": 3.5694310932447784,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003763307345459903,
      "loss": 5.165,
      "mean_token_accuracy": 0.21303742229938508,
      "num_tokens": 82760502.0,
      "step": 40845
    },
    {
      "entropy": 6.009467363357544,
      "epoch": 3.5698680415974833,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003763029114610479,
      "loss": 5.233,
      "mean_token_accuracy": 0.2082640990614891,
      "num_tokens": 82771477.0,
      "step": 40850
    },
    {
      "entropy": 6.039036560058594,
      "epoch": 3.5703049899501877,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003762750864330625,
      "loss": 5.2143,
      "mean_token_accuracy": 0.20949849486351013,
      "num_tokens": 82781572.0,
      "step": 40855
    },
    {
      "entropy": 6.067295360565185,
      "epoch": 3.5707419383028927,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037624725946256793,
      "loss": 5.186,
      "mean_token_accuracy": 0.20766327083110808,
      "num_tokens": 82790217.0,
      "step": 40860
    },
    {
      "entropy": 6.062140417098999,
      "epoch": 3.5711788866555976,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037621943055009786,
      "loss": 5.2326,
      "mean_token_accuracy": 0.21210016459226608,
      "num_tokens": 82800566.0,
      "step": 40865
    },
    {
      "entropy": 6.052109241485596,
      "epoch": 3.571615835008302,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003761915996961861,
      "loss": 5.2349,
      "mean_token_accuracy": 0.21233803778886795,
      "num_tokens": 82810188.0,
      "step": 40870
    },
    {
      "entropy": 6.048530054092407,
      "epoch": 3.5720527833610065,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037616376690136646,
      "loss": 5.2485,
      "mean_token_accuracy": 0.2065952628850937,
      "num_tokens": 82819509.0,
      "step": 40875
    },
    {
      "entropy": 6.0704268455505375,
      "epoch": 3.5724897317137114,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003761359321661727,
      "loss": 5.2247,
      "mean_token_accuracy": 0.20050559788942338,
      "num_tokens": 82830172.0,
      "step": 40880
    },
    {
      "entropy": 6.0874847888946535,
      "epoch": 3.5729266800664163,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037610809549113876,
      "loss": 5.2465,
      "mean_token_accuracy": 0.21206039786338807,
      "num_tokens": 82839245.0,
      "step": 40885
    },
    {
      "entropy": 6.024173259735107,
      "epoch": 3.573363628419121,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003760802568767986,
      "loss": 5.2788,
      "mean_token_accuracy": 0.20218501687049867,
      "num_tokens": 82850479.0,
      "step": 40890
    },
    {
      "entropy": 6.044840335845947,
      "epoch": 3.5738005767718257,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003760524163236862,
      "loss": 5.2436,
      "mean_token_accuracy": 0.20578669011592865,
      "num_tokens": 82859816.0,
      "step": 40895
    },
    {
      "entropy": 6.069573402404785,
      "epoch": 3.57423752512453,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003760245738323354,
      "loss": 5.2851,
      "mean_token_accuracy": 0.2041690543293953,
      "num_tokens": 82870124.0,
      "step": 40900
    },
    {
      "entropy": 6.070343637466431,
      "epoch": 3.574674473477235,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003759967294032803,
      "loss": 5.2241,
      "mean_token_accuracy": 0.20948986411094667,
      "num_tokens": 82880721.0,
      "step": 40905
    },
    {
      "entropy": 6.094638299942017,
      "epoch": 3.5751114218299396,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003759688830370551,
      "loss": 5.2578,
      "mean_token_accuracy": 0.20262978672981263,
      "num_tokens": 82890580.0,
      "step": 40910
    },
    {
      "entropy": 6.068497991561889,
      "epoch": 3.5755483701826445,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037594103473419384,
      "loss": 5.377,
      "mean_token_accuracy": 0.19895949810743332,
      "num_tokens": 82902465.0,
      "step": 40915
    },
    {
      "entropy": 6.026241159439087,
      "epoch": 3.575985318535349,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037591318449523063,
      "loss": 5.1921,
      "mean_token_accuracy": 0.208510160446167,
      "num_tokens": 82912665.0,
      "step": 40920
    },
    {
      "entropy": 6.0418308734893795,
      "epoch": 3.576422266888054,
      "grad_norm": 1.390625,
      "learning_rate": 0.00037588533232069967,
      "loss": 5.1404,
      "mean_token_accuracy": 0.2129328653216362,
      "num_tokens": 82921727.0,
      "step": 40925
    },
    {
      "entropy": 6.053713321685791,
      "epoch": 3.5768592152407583,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003758574782111351,
      "loss": 5.2318,
      "mean_token_accuracy": 0.20717397779226304,
      "num_tokens": 82932449.0,
      "step": 40930
    },
    {
      "entropy": 6.011027574539185,
      "epoch": 3.5772961635934633,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003758296221670713,
      "loss": 5.1503,
      "mean_token_accuracy": 0.21139203161001205,
      "num_tokens": 82941939.0,
      "step": 40935
    },
    {
      "entropy": 5.891794109344483,
      "epoch": 3.577733111946168,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003758017641890425,
      "loss": 5.1459,
      "mean_token_accuracy": 0.22149102240800858,
      "num_tokens": 82951958.0,
      "step": 40940
    },
    {
      "entropy": 5.927540874481201,
      "epoch": 3.5781700602988726,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003757739042775831,
      "loss": 5.2289,
      "mean_token_accuracy": 0.2212016075849533,
      "num_tokens": 82963014.0,
      "step": 40945
    },
    {
      "entropy": 6.072748231887817,
      "epoch": 3.5786070086515775,
      "grad_norm": 1.25,
      "learning_rate": 0.0003757460424332273,
      "loss": 5.2408,
      "mean_token_accuracy": 0.20562423318624495,
      "num_tokens": 82973081.0,
      "step": 40950
    },
    {
      "entropy": 6.0825928211212155,
      "epoch": 3.579043957004282,
      "grad_norm": 1.25,
      "learning_rate": 0.0003757181786565097,
      "loss": 5.2435,
      "mean_token_accuracy": 0.20773928314447404,
      "num_tokens": 82982835.0,
      "step": 40955
    },
    {
      "entropy": 6.0379403591156,
      "epoch": 3.579480905356987,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003756903129479646,
      "loss": 5.2144,
      "mean_token_accuracy": 0.20950407832860946,
      "num_tokens": 82993275.0,
      "step": 40960
    },
    {
      "entropy": 6.087337255477905,
      "epoch": 3.5799178537096914,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003756624453081266,
      "loss": 5.2823,
      "mean_token_accuracy": 0.20740071833133697,
      "num_tokens": 83003844.0,
      "step": 40965
    },
    {
      "entropy": 6.11211838722229,
      "epoch": 3.5803548020623963,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003756345757375302,
      "loss": 5.2647,
      "mean_token_accuracy": 0.20460207164287567,
      "num_tokens": 83014248.0,
      "step": 40970
    },
    {
      "entropy": 6.004701137542725,
      "epoch": 3.580791750415101,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003756067042367098,
      "loss": 5.1667,
      "mean_token_accuracy": 0.2177840143442154,
      "num_tokens": 83023967.0,
      "step": 40975
    },
    {
      "entropy": 6.0777007102966305,
      "epoch": 3.5812286987678057,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003755788308062001,
      "loss": 5.2884,
      "mean_token_accuracy": 0.20744313895702363,
      "num_tokens": 83034167.0,
      "step": 40980
    },
    {
      "entropy": 6.00642466545105,
      "epoch": 3.58166564712051,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037555095544653574,
      "loss": 5.2299,
      "mean_token_accuracy": 0.21009082049131395,
      "num_tokens": 83043822.0,
      "step": 40985
    },
    {
      "entropy": 5.972096395492554,
      "epoch": 3.582102595473215,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037552307815825136,
      "loss": 5.2348,
      "mean_token_accuracy": 0.2172030597925186,
      "num_tokens": 83054152.0,
      "step": 40990
    },
    {
      "entropy": 6.073744297027588,
      "epoch": 3.58253954382592,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037549519894188174,
      "loss": 5.2023,
      "mean_token_accuracy": 0.2115706667304039,
      "num_tokens": 83064553.0,
      "step": 40995
    },
    {
      "entropy": 6.045202827453613,
      "epoch": 3.5829764921786245,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003754673177979614,
      "loss": 5.2369,
      "mean_token_accuracy": 0.21111951321363448,
      "num_tokens": 83073423.0,
      "step": 41000
    },
    {
      "entropy": 6.040544843673706,
      "epoch": 3.583413440531329,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037543943472702536,
      "loss": 5.237,
      "mean_token_accuracy": 0.21782177537679673,
      "num_tokens": 83083774.0,
      "step": 41005
    },
    {
      "entropy": 6.003307867050171,
      "epoch": 3.583850388884034,
      "grad_norm": 1.375,
      "learning_rate": 0.0003754115497296084,
      "loss": 5.1358,
      "mean_token_accuracy": 0.21568583548069,
      "num_tokens": 83092647.0,
      "step": 41010
    },
    {
      "entropy": 6.07802414894104,
      "epoch": 3.5842873372367388,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00037538366280624513,
      "loss": 5.3785,
      "mean_token_accuracy": 0.20344266295433044,
      "num_tokens": 83104070.0,
      "step": 41015
    },
    {
      "entropy": 6.006004714965821,
      "epoch": 3.5847242855894432,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003753557739574707,
      "loss": 5.1957,
      "mean_token_accuracy": 0.21628138422966003,
      "num_tokens": 83113747.0,
      "step": 41020
    },
    {
      "entropy": 6.085330963134766,
      "epoch": 3.585161233942148,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00037532788318382,
      "loss": 5.2789,
      "mean_token_accuracy": 0.20092638432979584,
      "num_tokens": 83123678.0,
      "step": 41025
    },
    {
      "entropy": 6.10540452003479,
      "epoch": 3.5855981822948526,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003752999904858278,
      "loss": 5.2557,
      "mean_token_accuracy": 0.20408979058265686,
      "num_tokens": 83133232.0,
      "step": 41030
    },
    {
      "entropy": 6.096759796142578,
      "epoch": 3.5860351306475575,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037527209586402934,
      "loss": 5.2772,
      "mean_token_accuracy": 0.20376822501420974,
      "num_tokens": 83144351.0,
      "step": 41035
    },
    {
      "entropy": 6.07641921043396,
      "epoch": 3.586472079000262,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003752441993189595,
      "loss": 5.2808,
      "mean_token_accuracy": 0.204872527718544,
      "num_tokens": 83155116.0,
      "step": 41040
    },
    {
      "entropy": 6.060232496261596,
      "epoch": 3.586909027352967,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003752163008511534,
      "loss": 5.2545,
      "mean_token_accuracy": 0.20810915231704713,
      "num_tokens": 83165227.0,
      "step": 41045
    },
    {
      "entropy": 6.085382413864136,
      "epoch": 3.587345975705672,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003751884004611461,
      "loss": 5.2647,
      "mean_token_accuracy": 0.20786261111497878,
      "num_tokens": 83175235.0,
      "step": 41050
    },
    {
      "entropy": 6.036030626296997,
      "epoch": 3.5877829240583763,
      "grad_norm": 1.34375,
      "learning_rate": 0.00037516049814947277,
      "loss": 5.239,
      "mean_token_accuracy": 0.20070448368787766,
      "num_tokens": 83185373.0,
      "step": 41055
    },
    {
      "entropy": 6.121939945220947,
      "epoch": 3.5882198724110808,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003751325939166687,
      "loss": 5.2735,
      "mean_token_accuracy": 0.20279154777526856,
      "num_tokens": 83195046.0,
      "step": 41060
    },
    {
      "entropy": 6.054036426544189,
      "epoch": 3.5886568207637857,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037510468776326894,
      "loss": 5.2464,
      "mean_token_accuracy": 0.2104936346411705,
      "num_tokens": 83203817.0,
      "step": 41065
    },
    {
      "entropy": 6.100069999694824,
      "epoch": 3.5890937691164906,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003750767796898088,
      "loss": 5.3088,
      "mean_token_accuracy": 0.2075034037232399,
      "num_tokens": 83215096.0,
      "step": 41070
    },
    {
      "entropy": 6.040353536605835,
      "epoch": 3.589530717469195,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003750488696968236,
      "loss": 5.3197,
      "mean_token_accuracy": 0.2058231845498085,
      "num_tokens": 83225547.0,
      "step": 41075
    },
    {
      "entropy": 6.055527067184448,
      "epoch": 3.5899676658219,
      "grad_norm": 1.375,
      "learning_rate": 0.00037502095778484866,
      "loss": 5.1956,
      "mean_token_accuracy": 0.21363843828439713,
      "num_tokens": 83235658.0,
      "step": 41080
    },
    {
      "entropy": 6.057150363922119,
      "epoch": 3.5904046141746044,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00037499304395441943,
      "loss": 5.234,
      "mean_token_accuracy": 0.20615829527378082,
      "num_tokens": 83244819.0,
      "step": 41085
    },
    {
      "entropy": 5.9041650772094725,
      "epoch": 3.5908415625273093,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003749651282060711,
      "loss": 5.1198,
      "mean_token_accuracy": 0.226784086227417,
      "num_tokens": 83255729.0,
      "step": 41090
    },
    {
      "entropy": 6.045647716522216,
      "epoch": 3.591278510880014,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003749372105403392,
      "loss": 5.2545,
      "mean_token_accuracy": 0.2113860324025154,
      "num_tokens": 83265687.0,
      "step": 41095
    },
    {
      "entropy": 6.051277780532837,
      "epoch": 3.5917154592327187,
      "grad_norm": 1.25,
      "learning_rate": 0.00037490929095775934,
      "loss": 5.216,
      "mean_token_accuracy": 0.21335383057594298,
      "num_tokens": 83275558.0,
      "step": 41100
    },
    {
      "entropy": 6.027816104888916,
      "epoch": 3.5921524075854236,
      "grad_norm": 1.421875,
      "learning_rate": 0.00037488136945886685,
      "loss": 5.2001,
      "mean_token_accuracy": 0.21501342058181763,
      "num_tokens": 83284744.0,
      "step": 41105
    },
    {
      "entropy": 6.0051477432250975,
      "epoch": 3.592589355938128,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003748534460441975,
      "loss": 5.2442,
      "mean_token_accuracy": 0.21057469099760057,
      "num_tokens": 83294372.0,
      "step": 41110
    },
    {
      "entropy": 6.027797794342041,
      "epoch": 3.5930263042908326,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037482552071428654,
      "loss": 5.2026,
      "mean_token_accuracy": 0.210369336605072,
      "num_tokens": 83305378.0,
      "step": 41115
    },
    {
      "entropy": 6.093115425109863,
      "epoch": 3.5934632526435375,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037479759346966985,
      "loss": 5.296,
      "mean_token_accuracy": 0.2006105065345764,
      "num_tokens": 83315073.0,
      "step": 41120
    },
    {
      "entropy": 6.039422607421875,
      "epoch": 3.5939002009962424,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037476966431088305,
      "loss": 5.1481,
      "mean_token_accuracy": 0.216356460750103,
      "num_tokens": 83324940.0,
      "step": 41125
    },
    {
      "entropy": 6.071395397186279,
      "epoch": 3.594337149348947,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003747417332384618,
      "loss": 5.2956,
      "mean_token_accuracy": 0.20485183596611023,
      "num_tokens": 83335678.0,
      "step": 41130
    },
    {
      "entropy": 5.951079416275024,
      "epoch": 3.594774097701652,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037471380025294186,
      "loss": 5.1889,
      "mean_token_accuracy": 0.21987420171499253,
      "num_tokens": 83345682.0,
      "step": 41135
    },
    {
      "entropy": 6.010871934890747,
      "epoch": 3.5952110460543563,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000374685865354859,
      "loss": 5.2166,
      "mean_token_accuracy": 0.21229759007692336,
      "num_tokens": 83354934.0,
      "step": 41140
    },
    {
      "entropy": 6.011968803405762,
      "epoch": 3.595647994407061,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000374657928544749,
      "loss": 5.2811,
      "mean_token_accuracy": 0.2069059655070305,
      "num_tokens": 83365647.0,
      "step": 41145
    },
    {
      "entropy": 6.009005117416382,
      "epoch": 3.5960849427597656,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003746299898231477,
      "loss": 5.2364,
      "mean_token_accuracy": 0.21208942979574202,
      "num_tokens": 83376040.0,
      "step": 41150
    },
    {
      "entropy": 6.030669736862182,
      "epoch": 3.5965218911124706,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003746020491905909,
      "loss": 5.2214,
      "mean_token_accuracy": 0.21166806817054748,
      "num_tokens": 83386097.0,
      "step": 41155
    },
    {
      "entropy": 6.151680946350098,
      "epoch": 3.5969588394651755,
      "grad_norm": 1.375,
      "learning_rate": 0.00037457410664761475,
      "loss": 5.3665,
      "mean_token_accuracy": 0.19769127815961837,
      "num_tokens": 83395555.0,
      "step": 41160
    },
    {
      "entropy": 6.077396440505981,
      "epoch": 3.59739578781788,
      "grad_norm": 2.046875,
      "learning_rate": 0.000374546162194755,
      "loss": 5.1922,
      "mean_token_accuracy": 0.2190249130129814,
      "num_tokens": 83406195.0,
      "step": 41165
    },
    {
      "entropy": 6.039519834518432,
      "epoch": 3.5978327361705844,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037451821583254764,
      "loss": 5.2098,
      "mean_token_accuracy": 0.2112584576010704,
      "num_tokens": 83415254.0,
      "step": 41170
    },
    {
      "entropy": 6.093936157226563,
      "epoch": 3.5982696845232893,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037449026756152876,
      "loss": 5.2016,
      "mean_token_accuracy": 0.2105535164475441,
      "num_tokens": 83425112.0,
      "step": 41175
    },
    {
      "entropy": 6.050670337677002,
      "epoch": 3.5987066328759942,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037446231738223453,
      "loss": 5.3159,
      "mean_token_accuracy": 0.20194671154022217,
      "num_tokens": 83436185.0,
      "step": 41180
    },
    {
      "entropy": 6.112664031982422,
      "epoch": 3.5991435812286987,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003744343652952008,
      "loss": 5.2843,
      "mean_token_accuracy": 0.20046931505203247,
      "num_tokens": 83446540.0,
      "step": 41185
    },
    {
      "entropy": 6.027129983901977,
      "epoch": 3.5995805295814036,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037440641130096386,
      "loss": 5.1668,
      "mean_token_accuracy": 0.21811046451330185,
      "num_tokens": 83457247.0,
      "step": 41190
    },
    {
      "entropy": 6.069227457046509,
      "epoch": 3.600017477934108,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003743784554000599,
      "loss": 5.2557,
      "mean_token_accuracy": 0.21138365268707277,
      "num_tokens": 83468132.0,
      "step": 41195
    },
    {
      "entropy": 6.010192012786865,
      "epoch": 3.600454426286813,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003743504975930252,
      "loss": 5.2213,
      "mean_token_accuracy": 0.21097091287374498,
      "num_tokens": 83477824.0,
      "step": 41200
    },
    {
      "entropy": 6.064004421234131,
      "epoch": 3.6008913746395175,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003743225378803958,
      "loss": 5.2432,
      "mean_token_accuracy": 0.2058028534054756,
      "num_tokens": 83487352.0,
      "step": 41205
    },
    {
      "entropy": 6.041876411437988,
      "epoch": 3.6013283229922224,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003742945762627081,
      "loss": 5.2245,
      "mean_token_accuracy": 0.2132693499326706,
      "num_tokens": 83497957.0,
      "step": 41210
    },
    {
      "entropy": 5.984076642990113,
      "epoch": 3.6017652713449273,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003742666127404983,
      "loss": 5.1556,
      "mean_token_accuracy": 0.22031001150608062,
      "num_tokens": 83508213.0,
      "step": 41215
    },
    {
      "entropy": 6.000862503051758,
      "epoch": 3.6022022196976318,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0003742386473143029,
      "loss": 5.2065,
      "mean_token_accuracy": 0.2174738049507141,
      "num_tokens": 83518694.0,
      "step": 41220
    },
    {
      "entropy": 6.057119131088257,
      "epoch": 3.6026391680503362,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037421067998465835,
      "loss": 5.1874,
      "mean_token_accuracy": 0.21247451454401017,
      "num_tokens": 83529077.0,
      "step": 41225
    },
    {
      "entropy": 5.980885982513428,
      "epoch": 3.603076116403041,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037418271075210087,
      "loss": 5.2274,
      "mean_token_accuracy": 0.21378295719623566,
      "num_tokens": 83539502.0,
      "step": 41230
    },
    {
      "entropy": 6.011671781539917,
      "epoch": 3.603513064755746,
      "grad_norm": 1.375,
      "learning_rate": 0.000374154739617167,
      "loss": 5.1886,
      "mean_token_accuracy": 0.21520014405250548,
      "num_tokens": 83549073.0,
      "step": 41235
    },
    {
      "entropy": 6.018255424499512,
      "epoch": 3.6039500131084505,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003741267665803934,
      "loss": 5.2564,
      "mean_token_accuracy": 0.21062879115343094,
      "num_tokens": 83560404.0,
      "step": 41240
    },
    {
      "entropy": 5.997541522979736,
      "epoch": 3.6043869614611554,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003740987916423164,
      "loss": 5.1966,
      "mean_token_accuracy": 0.21511661857366562,
      "num_tokens": 83570421.0,
      "step": 41245
    },
    {
      "entropy": 6.035231065750122,
      "epoch": 3.60482390981386,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037407081480347263,
      "loss": 5.2586,
      "mean_token_accuracy": 0.20412037670612335,
      "num_tokens": 83581653.0,
      "step": 41250
    },
    {
      "entropy": 6.028960180282593,
      "epoch": 3.605260858166565,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003740428360643987,
      "loss": 5.1909,
      "mean_token_accuracy": 0.21273931115865707,
      "num_tokens": 83592190.0,
      "step": 41255
    },
    {
      "entropy": 6.06026873588562,
      "epoch": 3.6056978065192693,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003740148554256313,
      "loss": 5.2009,
      "mean_token_accuracy": 0.20850100815296174,
      "num_tokens": 83602495.0,
      "step": 41260
    },
    {
      "entropy": 6.02274808883667,
      "epoch": 3.606134754871974,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003739868728877071,
      "loss": 5.2246,
      "mean_token_accuracy": 0.21175771653652192,
      "num_tokens": 83612772.0,
      "step": 41265
    },
    {
      "entropy": 6.027310419082641,
      "epoch": 3.6065717032246787,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003739588884511628,
      "loss": 5.272,
      "mean_token_accuracy": 0.2149599388241768,
      "num_tokens": 83623472.0,
      "step": 41270
    },
    {
      "entropy": 6.0502136707305905,
      "epoch": 3.6070086515773836,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003739309021165352,
      "loss": 5.1998,
      "mean_token_accuracy": 0.2087159812450409,
      "num_tokens": 83633783.0,
      "step": 41275
    },
    {
      "entropy": 6.0654253482818605,
      "epoch": 3.607445599930088,
      "grad_norm": 1.25,
      "learning_rate": 0.00037390291388436105,
      "loss": 5.2586,
      "mean_token_accuracy": 0.208660788834095,
      "num_tokens": 83643594.0,
      "step": 41280
    },
    {
      "entropy": 5.976083087921142,
      "epoch": 3.607882548282793,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003738749237551772,
      "loss": 5.125,
      "mean_token_accuracy": 0.2216646507382393,
      "num_tokens": 83653405.0,
      "step": 41285
    },
    {
      "entropy": 6.002198362350464,
      "epoch": 3.608319496635498,
      "grad_norm": 1.359375,
      "learning_rate": 0.00037384693172952047,
      "loss": 5.1802,
      "mean_token_accuracy": 0.2128771498799324,
      "num_tokens": 83663233.0,
      "step": 41290
    },
    {
      "entropy": 6.045830106735229,
      "epoch": 3.6087564449882024,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003738189378079278,
      "loss": 5.2443,
      "mean_token_accuracy": 0.21544720679521562,
      "num_tokens": 83672899.0,
      "step": 41295
    },
    {
      "entropy": 6.02002100944519,
      "epoch": 3.609193393340907,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037379094199093603,
      "loss": 5.3099,
      "mean_token_accuracy": 0.20721750259399413,
      "num_tokens": 83683542.0,
      "step": 41300
    },
    {
      "entropy": 6.058401441574096,
      "epoch": 3.6096303416936117,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037376294427908237,
      "loss": 5.3081,
      "mean_token_accuracy": 0.208072030544281,
      "num_tokens": 83693806.0,
      "step": 41305
    },
    {
      "entropy": 6.004207038879395,
      "epoch": 3.6100672900463167,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003737349446729036,
      "loss": 5.159,
      "mean_token_accuracy": 0.21787144839763642,
      "num_tokens": 83704293.0,
      "step": 41310
    },
    {
      "entropy": 6.021062755584717,
      "epoch": 3.610504238399021,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003737069431729369,
      "loss": 5.1874,
      "mean_token_accuracy": 0.2147047758102417,
      "num_tokens": 83714321.0,
      "step": 41315
    },
    {
      "entropy": 5.974083709716797,
      "epoch": 3.610941186751726,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037367893977971933,
      "loss": 5.159,
      "mean_token_accuracy": 0.21192608922719955,
      "num_tokens": 83723815.0,
      "step": 41320
    },
    {
      "entropy": 6.071401119232178,
      "epoch": 3.6113781351044305,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037365093449378796,
      "loss": 5.2747,
      "mean_token_accuracy": 0.2020239159464836,
      "num_tokens": 83734036.0,
      "step": 41325
    },
    {
      "entropy": 6.1262932300567625,
      "epoch": 3.6118150834571354,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037362292731567997,
      "loss": 5.2602,
      "mean_token_accuracy": 0.20615004748106003,
      "num_tokens": 83744343.0,
      "step": 41330
    },
    {
      "entropy": 6.109848403930664,
      "epoch": 3.61225203180984,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003735949182459325,
      "loss": 5.276,
      "mean_token_accuracy": 0.20808148831129075,
      "num_tokens": 83754898.0,
      "step": 41335
    },
    {
      "entropy": 6.113986349105835,
      "epoch": 3.612688980162545,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003735669072850829,
      "loss": 5.2678,
      "mean_token_accuracy": 0.19821044206619262,
      "num_tokens": 83765964.0,
      "step": 41340
    },
    {
      "entropy": 6.04616346359253,
      "epoch": 3.6131259285152497,
      "grad_norm": 1.375,
      "learning_rate": 0.00037353889443366843,
      "loss": 5.1936,
      "mean_token_accuracy": 0.20733582824468613,
      "num_tokens": 83775920.0,
      "step": 41345
    },
    {
      "entropy": 6.032669925689698,
      "epoch": 3.613562876867954,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00037351087969222626,
      "loss": 5.1606,
      "mean_token_accuracy": 0.21643892973661422,
      "num_tokens": 83784371.0,
      "step": 41350
    },
    {
      "entropy": 6.0316338539123535,
      "epoch": 3.6139998252206587,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003734828630612938,
      "loss": 5.2397,
      "mean_token_accuracy": 0.20324414372444152,
      "num_tokens": 83794618.0,
      "step": 41355
    },
    {
      "entropy": 6.0562560081481935,
      "epoch": 3.6144367735733636,
      "grad_norm": 1.25,
      "learning_rate": 0.0003734548445414084,
      "loss": 5.2763,
      "mean_token_accuracy": 0.21058307439088822,
      "num_tokens": 83805070.0,
      "step": 41360
    },
    {
      "entropy": 6.051653575897217,
      "epoch": 3.6148737219260685,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037342682413310746,
      "loss": 5.288,
      "mean_token_accuracy": 0.20313206762075425,
      "num_tokens": 83816257.0,
      "step": 41365
    },
    {
      "entropy": 6.0154603004455565,
      "epoch": 3.615310670278773,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037339880183692855,
      "loss": 5.2513,
      "mean_token_accuracy": 0.2028719574213028,
      "num_tokens": 83826432.0,
      "step": 41370
    },
    {
      "entropy": 6.077649211883545,
      "epoch": 3.615747618631478,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037337077765340905,
      "loss": 5.2479,
      "mean_token_accuracy": 0.21269578933715821,
      "num_tokens": 83836746.0,
      "step": 41375
    },
    {
      "entropy": 6.017109346389771,
      "epoch": 3.6161845669841823,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00037334275158308643,
      "loss": 5.174,
      "mean_token_accuracy": 0.21660541296005248,
      "num_tokens": 83845783.0,
      "step": 41380
    },
    {
      "entropy": 5.988151025772095,
      "epoch": 3.6166215153368872,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003733147236264983,
      "loss": 5.2236,
      "mean_token_accuracy": 0.21005240827798843,
      "num_tokens": 83856981.0,
      "step": 41385
    },
    {
      "entropy": 6.048430585861206,
      "epoch": 3.6170584636895917,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037328669378418236,
      "loss": 5.2961,
      "mean_token_accuracy": 0.2030690923333168,
      "num_tokens": 83867873.0,
      "step": 41390
    },
    {
      "entropy": 6.025294971466065,
      "epoch": 3.6174954120422966,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003732586620566761,
      "loss": 5.2085,
      "mean_token_accuracy": 0.20633778870105743,
      "num_tokens": 83877946.0,
      "step": 41395
    },
    {
      "entropy": 6.0220709323883055,
      "epoch": 3.6179323603950015,
      "grad_norm": 1.390625,
      "learning_rate": 0.00037323062844451713,
      "loss": 5.2062,
      "mean_token_accuracy": 0.2106395959854126,
      "num_tokens": 83888318.0,
      "step": 41400
    },
    {
      "entropy": 6.011982059478759,
      "epoch": 3.618369308747706,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003732025929482433,
      "loss": 5.1788,
      "mean_token_accuracy": 0.20872800499200822,
      "num_tokens": 83899092.0,
      "step": 41405
    },
    {
      "entropy": 6.028029489517212,
      "epoch": 3.6188062571004105,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003731745555683922,
      "loss": 5.1851,
      "mean_token_accuracy": 0.22304904758930205,
      "num_tokens": 83908257.0,
      "step": 41410
    },
    {
      "entropy": 6.027441120147705,
      "epoch": 3.6192432054531154,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037314651630550186,
      "loss": 5.2481,
      "mean_token_accuracy": 0.20652113407850264,
      "num_tokens": 83918821.0,
      "step": 41415
    },
    {
      "entropy": 5.993592977523804,
      "epoch": 3.6196801538058203,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037311847516010975,
      "loss": 5.1978,
      "mean_token_accuracy": 0.21404399424791337,
      "num_tokens": 83928122.0,
      "step": 41420
    },
    {
      "entropy": 6.048395586013794,
      "epoch": 3.620117102158525,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003730904321327539,
      "loss": 5.2672,
      "mean_token_accuracy": 0.20870779305696488,
      "num_tokens": 83939904.0,
      "step": 41425
    },
    {
      "entropy": 6.038040351867676,
      "epoch": 3.6205540505112297,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003730623872239722,
      "loss": 5.2059,
      "mean_token_accuracy": 0.2104217827320099,
      "num_tokens": 83950631.0,
      "step": 41430
    },
    {
      "entropy": 6.079219579696655,
      "epoch": 3.620990998863934,
      "grad_norm": 1.25,
      "learning_rate": 0.0003730343404343025,
      "loss": 5.2945,
      "mean_token_accuracy": 0.2052599161863327,
      "num_tokens": 83960860.0,
      "step": 41435
    },
    {
      "entropy": 6.066054773330689,
      "epoch": 3.621427947216639,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003730062917642828,
      "loss": 5.3084,
      "mean_token_accuracy": 0.20079899728298187,
      "num_tokens": 83971182.0,
      "step": 41440
    },
    {
      "entropy": 5.999891662597657,
      "epoch": 3.6218648955693435,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003729782412144511,
      "loss": 5.1589,
      "mean_token_accuracy": 0.2145593598484993,
      "num_tokens": 83981576.0,
      "step": 41445
    },
    {
      "entropy": 6.066121768951416,
      "epoch": 3.6223018439220485,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003729501887853453,
      "loss": 5.2251,
      "mean_token_accuracy": 0.20955556482076645,
      "num_tokens": 83991526.0,
      "step": 41450
    },
    {
      "entropy": 6.037787675857544,
      "epoch": 3.6227387922747534,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003729221344775036,
      "loss": 5.2702,
      "mean_token_accuracy": 0.20621846318244935,
      "num_tokens": 84001979.0,
      "step": 41455
    },
    {
      "entropy": 5.979437446594238,
      "epoch": 3.623175740627458,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003728940782914641,
      "loss": 5.1261,
      "mean_token_accuracy": 0.21685864478349687,
      "num_tokens": 84011977.0,
      "step": 41460
    },
    {
      "entropy": 6.070886039733887,
      "epoch": 3.6236126889801623,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037286602022776485,
      "loss": 5.2438,
      "mean_token_accuracy": 0.21010438948869706,
      "num_tokens": 84021826.0,
      "step": 41465
    },
    {
      "entropy": 6.02928409576416,
      "epoch": 3.6240496373328672,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037283796028694403,
      "loss": 5.1731,
      "mean_token_accuracy": 0.21932546198368072,
      "num_tokens": 84032180.0,
      "step": 41470
    },
    {
      "entropy": 6.056326150894165,
      "epoch": 3.624486585685572,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037280989846953986,
      "loss": 5.2996,
      "mean_token_accuracy": 0.20454555302858352,
      "num_tokens": 84043698.0,
      "step": 41475
    },
    {
      "entropy": 6.012975358963013,
      "epoch": 3.6249235340382766,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003727818347760906,
      "loss": 5.1722,
      "mean_token_accuracy": 0.21200987249612807,
      "num_tokens": 84053323.0,
      "step": 41480
    },
    {
      "entropy": 6.095404863357544,
      "epoch": 3.6253604823909815,
      "grad_norm": 1.421875,
      "learning_rate": 0.00037275376920713454,
      "loss": 5.3634,
      "mean_token_accuracy": 0.2044745072722435,
      "num_tokens": 84064092.0,
      "step": 41485
    },
    {
      "entropy": 6.05866756439209,
      "epoch": 3.625797430743686,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037272570176320995,
      "loss": 5.2289,
      "mean_token_accuracy": 0.20748103857040406,
      "num_tokens": 84073661.0,
      "step": 41490
    },
    {
      "entropy": 5.995351648330688,
      "epoch": 3.626234379096391,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003726976324448552,
      "loss": 5.264,
      "mean_token_accuracy": 0.21107938140630722,
      "num_tokens": 84083199.0,
      "step": 41495
    },
    {
      "entropy": 6.053361892700195,
      "epoch": 3.6266713274490954,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003726695612526086,
      "loss": 5.2711,
      "mean_token_accuracy": 0.2088525876402855,
      "num_tokens": 84093600.0,
      "step": 41500
    },
    {
      "entropy": 6.0862654685974125,
      "epoch": 3.6271082758018003,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003726414881870087,
      "loss": 5.2711,
      "mean_token_accuracy": 0.20735715478658676,
      "num_tokens": 84103147.0,
      "step": 41505
    },
    {
      "entropy": 5.903131914138794,
      "epoch": 3.627545224154505,
      "grad_norm": 1.65625,
      "learning_rate": 0.00037261341324859386,
      "loss": 5.1687,
      "mean_token_accuracy": 0.2154013141989708,
      "num_tokens": 84113719.0,
      "step": 41510
    },
    {
      "entropy": 6.014184951782227,
      "epoch": 3.6279821725072097,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037258533643790264,
      "loss": 5.1874,
      "mean_token_accuracy": 0.20850200653076173,
      "num_tokens": 84123563.0,
      "step": 41515
    },
    {
      "entropy": 6.05769362449646,
      "epoch": 3.628419120859914,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037255725775547357,
      "loss": 5.2746,
      "mean_token_accuracy": 0.20818746089935303,
      "num_tokens": 84134753.0,
      "step": 41520
    },
    {
      "entropy": 6.101606702804565,
      "epoch": 3.628856069212619,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037252917720184507,
      "loss": 5.265,
      "mean_token_accuracy": 0.1984283909201622,
      "num_tokens": 84146549.0,
      "step": 41525
    },
    {
      "entropy": 6.099269533157349,
      "epoch": 3.629293017565324,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003725010947775559,
      "loss": 5.3362,
      "mean_token_accuracy": 0.2034755304455757,
      "num_tokens": 84157069.0,
      "step": 41530
    },
    {
      "entropy": 6.009864234924317,
      "epoch": 3.6297299659180284,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037247301048314464,
      "loss": 5.2687,
      "mean_token_accuracy": 0.21022784560918809,
      "num_tokens": 84167516.0,
      "step": 41535
    },
    {
      "entropy": 6.04904317855835,
      "epoch": 3.6301669142707333,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037244492431914996,
      "loss": 5.2283,
      "mean_token_accuracy": 0.2140628382563591,
      "num_tokens": 84178539.0,
      "step": 41540
    },
    {
      "entropy": 6.086180591583252,
      "epoch": 3.630603862623438,
      "grad_norm": 1.25,
      "learning_rate": 0.00037241683628611055,
      "loss": 5.2223,
      "mean_token_accuracy": 0.21598641723394393,
      "num_tokens": 84188605.0,
      "step": 41545
    },
    {
      "entropy": 6.040867662429809,
      "epoch": 3.6310408109761427,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00037238874638456524,
      "loss": 5.1993,
      "mean_token_accuracy": 0.21423469632863998,
      "num_tokens": 84197478.0,
      "step": 41550
    },
    {
      "entropy": 6.042101573944092,
      "epoch": 3.631477759328847,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003723606546150527,
      "loss": 5.2625,
      "mean_token_accuracy": 0.20569287240505219,
      "num_tokens": 84207455.0,
      "step": 41555
    },
    {
      "entropy": 6.0023668766021725,
      "epoch": 3.631914707681552,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003723325609781118,
      "loss": 5.1337,
      "mean_token_accuracy": 0.22118629962205888,
      "num_tokens": 84217336.0,
      "step": 41560
    },
    {
      "entropy": 6.019694471359253,
      "epoch": 3.632351656034257,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037230446547428136,
      "loss": 5.1948,
      "mean_token_accuracy": 0.20824607014656066,
      "num_tokens": 84227474.0,
      "step": 41565
    },
    {
      "entropy": 6.04484372138977,
      "epoch": 3.6327886043869615,
      "grad_norm": 1.375,
      "learning_rate": 0.0003722763681041003,
      "loss": 5.2524,
      "mean_token_accuracy": 0.20896682441234588,
      "num_tokens": 84237382.0,
      "step": 41570
    },
    {
      "entropy": 6.079791593551636,
      "epoch": 3.633225552739666,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003722482688681075,
      "loss": 5.2773,
      "mean_token_accuracy": 0.2030806377530098,
      "num_tokens": 84247816.0,
      "step": 41575
    },
    {
      "entropy": 6.057506513595581,
      "epoch": 3.633662501092371,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000372220167766842,
      "loss": 5.2196,
      "mean_token_accuracy": 0.20732390135526657,
      "num_tokens": 84257056.0,
      "step": 41580
    },
    {
      "entropy": 6.138320779800415,
      "epoch": 3.634099449445076,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00037219206480084273,
      "loss": 5.3488,
      "mean_token_accuracy": 0.20527807027101516,
      "num_tokens": 84267499.0,
      "step": 41585
    },
    {
      "entropy": 6.060419321060181,
      "epoch": 3.6345363977977803,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003721639599706487,
      "loss": 5.1809,
      "mean_token_accuracy": 0.2115587830543518,
      "num_tokens": 84277548.0,
      "step": 41590
    },
    {
      "entropy": 6.010973167419434,
      "epoch": 3.634973346150485,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00037213585327679904,
      "loss": 5.1874,
      "mean_token_accuracy": 0.20775680989027023,
      "num_tokens": 84287461.0,
      "step": 41595
    },
    {
      "entropy": 6.032347202301025,
      "epoch": 3.6354102945031896,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037210774471983283,
      "loss": 5.2108,
      "mean_token_accuracy": 0.21051595509052276,
      "num_tokens": 84297878.0,
      "step": 41600
    },
    {
      "entropy": 6.024797058105468,
      "epoch": 3.6358472428558946,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037207963430028915,
      "loss": 5.1584,
      "mean_token_accuracy": 0.2168066009879112,
      "num_tokens": 84307850.0,
      "step": 41605
    },
    {
      "entropy": 6.024498081207275,
      "epoch": 3.636284191208599,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003720515220187072,
      "loss": 5.2872,
      "mean_token_accuracy": 0.20497466623783112,
      "num_tokens": 84318111.0,
      "step": 41610
    },
    {
      "entropy": 5.9614685535430905,
      "epoch": 3.636721139561304,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003720234078756262,
      "loss": 5.1111,
      "mean_token_accuracy": 0.22013984620571136,
      "num_tokens": 84327118.0,
      "step": 41615
    },
    {
      "entropy": 6.069547510147094,
      "epoch": 3.6371580879140084,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003719952918715854,
      "loss": 5.2606,
      "mean_token_accuracy": 0.2087448626756668,
      "num_tokens": 84338428.0,
      "step": 41620
    },
    {
      "entropy": 6.130778217315674,
      "epoch": 3.6375950362667133,
      "grad_norm": 1.25,
      "learning_rate": 0.00037196717400712415,
      "loss": 5.3281,
      "mean_token_accuracy": 0.2010654553771019,
      "num_tokens": 84349117.0,
      "step": 41625
    },
    {
      "entropy": 6.105675315856933,
      "epoch": 3.638031984619418,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003719390542827816,
      "loss": 5.2747,
      "mean_token_accuracy": 0.2052255854010582,
      "num_tokens": 84358807.0,
      "step": 41630
    },
    {
      "entropy": 6.006306886672974,
      "epoch": 3.6384689329721227,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003719109326990973,
      "loss": 5.2417,
      "mean_token_accuracy": 0.21456330567598342,
      "num_tokens": 84368970.0,
      "step": 41635
    },
    {
      "entropy": 6.009486722946167,
      "epoch": 3.6389058813248276,
      "grad_norm": 1.34375,
      "learning_rate": 0.00037188280925661043,
      "loss": 5.1734,
      "mean_token_accuracy": 0.21814386546611786,
      "num_tokens": 84378475.0,
      "step": 41640
    },
    {
      "entropy": 6.007948160171509,
      "epoch": 3.639342829677532,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003718546839558606,
      "loss": 5.1632,
      "mean_token_accuracy": 0.21569331884384155,
      "num_tokens": 84387832.0,
      "step": 41645
    },
    {
      "entropy": 6.0195800304412845,
      "epoch": 3.6397797780302366,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037182655679738713,
      "loss": 5.1779,
      "mean_token_accuracy": 0.21212064772844313,
      "num_tokens": 84398316.0,
      "step": 41650
    },
    {
      "entropy": 6.021406364440918,
      "epoch": 3.6402167263829415,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003717984277817296,
      "loss": 5.1591,
      "mean_token_accuracy": 0.21045834869146346,
      "num_tokens": 84408624.0,
      "step": 41655
    },
    {
      "entropy": 6.036335468292236,
      "epoch": 3.6406536747356464,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037177029690942744,
      "loss": 5.1629,
      "mean_token_accuracy": 0.21588886231184007,
      "num_tokens": 84417422.0,
      "step": 41660
    },
    {
      "entropy": 6.029938268661499,
      "epoch": 3.641090623088351,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003717421641810202,
      "loss": 5.2864,
      "mean_token_accuracy": 0.202339044213295,
      "num_tokens": 84427857.0,
      "step": 41665
    },
    {
      "entropy": 6.115413475036621,
      "epoch": 3.6415275714410558,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037171402959704767,
      "loss": 5.2739,
      "mean_token_accuracy": 0.20019813627004623,
      "num_tokens": 84438468.0,
      "step": 41670
    },
    {
      "entropy": 6.05217604637146,
      "epoch": 3.6419645197937602,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037168589315804944,
      "loss": 5.2584,
      "mean_token_accuracy": 0.20731989294290543,
      "num_tokens": 84449371.0,
      "step": 41675
    },
    {
      "entropy": 6.039803218841553,
      "epoch": 3.642401468146465,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003716577548645651,
      "loss": 5.208,
      "mean_token_accuracy": 0.2060033917427063,
      "num_tokens": 84458569.0,
      "step": 41680
    },
    {
      "entropy": 6.028981351852417,
      "epoch": 3.6428384164991696,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003716296147171343,
      "loss": 5.3094,
      "mean_token_accuracy": 0.21106475293636323,
      "num_tokens": 84469244.0,
      "step": 41685
    },
    {
      "entropy": 5.977675485610962,
      "epoch": 3.6432753648518745,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003716014727162969,
      "loss": 5.1772,
      "mean_token_accuracy": 0.22830240726470946,
      "num_tokens": 84478902.0,
      "step": 41690
    },
    {
      "entropy": 6.02497968673706,
      "epoch": 3.6437123132045794,
      "grad_norm": 1.390625,
      "learning_rate": 0.00037157332886259253,
      "loss": 5.1904,
      "mean_token_accuracy": 0.2137203887104988,
      "num_tokens": 84489166.0,
      "step": 41695
    },
    {
      "entropy": 6.085036754608154,
      "epoch": 3.644149261557284,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003715451831565613,
      "loss": 5.1691,
      "mean_token_accuracy": 0.216258305311203,
      "num_tokens": 84499093.0,
      "step": 41700
    },
    {
      "entropy": 6.057365465164184,
      "epoch": 3.6445862099099884,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00037151703559874273,
      "loss": 5.1959,
      "mean_token_accuracy": 0.21147360205650328,
      "num_tokens": 84508898.0,
      "step": 41705
    },
    {
      "entropy": 6.08314995765686,
      "epoch": 3.6450231582626933,
      "grad_norm": 1.421875,
      "learning_rate": 0.00037148888618967677,
      "loss": 5.2986,
      "mean_token_accuracy": 0.20429212152957915,
      "num_tokens": 84518277.0,
      "step": 41710
    },
    {
      "entropy": 6.082430601119995,
      "epoch": 3.645460106615398,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003714607349299035,
      "loss": 5.2484,
      "mean_token_accuracy": 0.2021609678864479,
      "num_tokens": 84527882.0,
      "step": 41715
    },
    {
      "entropy": 6.018921184539795,
      "epoch": 3.6458970549681027,
      "grad_norm": 1.390625,
      "learning_rate": 0.00037143258181996286,
      "loss": 5.2224,
      "mean_token_accuracy": 0.21359857469797133,
      "num_tokens": 84538065.0,
      "step": 41720
    },
    {
      "entropy": 6.065952730178833,
      "epoch": 3.6463340033208076,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003714044268603947,
      "loss": 5.2511,
      "mean_token_accuracy": 0.20827705562114715,
      "num_tokens": 84547923.0,
      "step": 41725
    },
    {
      "entropy": 6.031115293502808,
      "epoch": 3.646770951673512,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003713762700517392,
      "loss": 5.2226,
      "mean_token_accuracy": 0.20776031911373138,
      "num_tokens": 84558725.0,
      "step": 41730
    },
    {
      "entropy": 6.028364515304565,
      "epoch": 3.647207900026217,
      "grad_norm": 1.5,
      "learning_rate": 0.00037134811139453625,
      "loss": 5.2198,
      "mean_token_accuracy": 0.2169780343770981,
      "num_tokens": 84568945.0,
      "step": 41735
    },
    {
      "entropy": 6.078113651275634,
      "epoch": 3.6476448483789214,
      "grad_norm": 1.234375,
      "learning_rate": 0.000371319950889326,
      "loss": 5.2846,
      "mean_token_accuracy": 0.20914031863212584,
      "num_tokens": 84579230.0,
      "step": 41740
    },
    {
      "entropy": 6.10175199508667,
      "epoch": 3.6480817967316264,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003712917885366487,
      "loss": 5.2414,
      "mean_token_accuracy": 0.2040543645620346,
      "num_tokens": 84588966.0,
      "step": 41745
    },
    {
      "entropy": 5.980372428894043,
      "epoch": 3.6485187450843313,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003712636243370444,
      "loss": 5.1718,
      "mean_token_accuracy": 0.21601906418800354,
      "num_tokens": 84598631.0,
      "step": 41750
    },
    {
      "entropy": 6.025386095046997,
      "epoch": 3.6489556934370357,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037123545829105336,
      "loss": 5.2255,
      "mean_token_accuracy": 0.212026347219944,
      "num_tokens": 84609849.0,
      "step": 41755
    },
    {
      "entropy": 6.052851963043213,
      "epoch": 3.64939264178974,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003712072903992159,
      "loss": 5.2417,
      "mean_token_accuracy": 0.2095771998167038,
      "num_tokens": 84619959.0,
      "step": 41760
    },
    {
      "entropy": 6.085357332229615,
      "epoch": 3.649829590142445,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003711791206620721,
      "loss": 5.2024,
      "mean_token_accuracy": 0.21585071831941605,
      "num_tokens": 84630224.0,
      "step": 41765
    },
    {
      "entropy": 6.07534704208374,
      "epoch": 3.65026653849515,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003711509490801624,
      "loss": 5.2449,
      "mean_token_accuracy": 0.20806976705789565,
      "num_tokens": 84641595.0,
      "step": 41770
    },
    {
      "entropy": 5.9831383228302,
      "epoch": 3.6507034868478545,
      "grad_norm": 1.40625,
      "learning_rate": 0.000371122775654027,
      "loss": 5.1755,
      "mean_token_accuracy": 0.21602923572063445,
      "num_tokens": 84652406.0,
      "step": 41775
    },
    {
      "entropy": 5.965927934646606,
      "epoch": 3.6511404352005594,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003710946003842065,
      "loss": 5.1347,
      "mean_token_accuracy": 0.22291115224361419,
      "num_tokens": 84662319.0,
      "step": 41780
    },
    {
      "entropy": 6.072427606582641,
      "epoch": 3.651577383553264,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003710664232712412,
      "loss": 5.2502,
      "mean_token_accuracy": 0.202026104927063,
      "num_tokens": 84673086.0,
      "step": 41785
    },
    {
      "entropy": 6.028208446502686,
      "epoch": 3.652014331905969,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003710382443156716,
      "loss": 5.2744,
      "mean_token_accuracy": 0.20434786677360534,
      "num_tokens": 84684536.0,
      "step": 41790
    },
    {
      "entropy": 6.023211193084717,
      "epoch": 3.6524512802586733,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003710100635180381,
      "loss": 5.2161,
      "mean_token_accuracy": 0.2136509597301483,
      "num_tokens": 84694975.0,
      "step": 41795
    },
    {
      "entropy": 6.021022176742553,
      "epoch": 3.652888228611378,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003709818808788813,
      "loss": 5.1784,
      "mean_token_accuracy": 0.2088350087404251,
      "num_tokens": 84704290.0,
      "step": 41800
    },
    {
      "entropy": 6.0794933319091795,
      "epoch": 3.653325176964083,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003709536963987417,
      "loss": 5.1957,
      "mean_token_accuracy": 0.210525581240654,
      "num_tokens": 84714280.0,
      "step": 41805
    },
    {
      "entropy": 6.086872816085815,
      "epoch": 3.6537621253167876,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037092551007815997,
      "loss": 5.2668,
      "mean_token_accuracy": 0.21110661923885346,
      "num_tokens": 84725127.0,
      "step": 41810
    },
    {
      "entropy": 6.086040592193603,
      "epoch": 3.654199073669492,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003708973219176767,
      "loss": 5.2618,
      "mean_token_accuracy": 0.20435124784708023,
      "num_tokens": 84736385.0,
      "step": 41815
    },
    {
      "entropy": 5.973568153381348,
      "epoch": 3.654636022022197,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037086913191783247,
      "loss": 5.1861,
      "mean_token_accuracy": 0.22274281680583954,
      "num_tokens": 84746489.0,
      "step": 41820
    },
    {
      "entropy": 6.055372047424316,
      "epoch": 3.655072970374902,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003708409400791681,
      "loss": 5.2458,
      "mean_token_accuracy": 0.2079032465815544,
      "num_tokens": 84756743.0,
      "step": 41825
    },
    {
      "entropy": 6.074552726745606,
      "epoch": 3.6555099187276063,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037081274640222425,
      "loss": 5.2052,
      "mean_token_accuracy": 0.2155635505914688,
      "num_tokens": 84767647.0,
      "step": 41830
    },
    {
      "entropy": 6.063151788711548,
      "epoch": 3.6559468670803112,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003707845508875417,
      "loss": 5.2759,
      "mean_token_accuracy": 0.19984497725963593,
      "num_tokens": 84778122.0,
      "step": 41835
    },
    {
      "entropy": 6.069613075256347,
      "epoch": 3.6563838154330157,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003707563535356614,
      "loss": 5.2626,
      "mean_token_accuracy": 0.2135695531964302,
      "num_tokens": 84787300.0,
      "step": 41840
    },
    {
      "entropy": 6.056424760818482,
      "epoch": 3.6568207637857206,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003707281543471239,
      "loss": 5.2036,
      "mean_token_accuracy": 0.21005377322435378,
      "num_tokens": 84798271.0,
      "step": 41845
    },
    {
      "entropy": 6.039096117019653,
      "epoch": 3.657257712138425,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003706999533224704,
      "loss": 5.2903,
      "mean_token_accuracy": 0.20552886724472047,
      "num_tokens": 84810377.0,
      "step": 41850
    },
    {
      "entropy": 6.051357841491699,
      "epoch": 3.65769466049113,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003706717504622415,
      "loss": 5.2086,
      "mean_token_accuracy": 0.20919393450021745,
      "num_tokens": 84821319.0,
      "step": 41855
    },
    {
      "entropy": 6.101205825805664,
      "epoch": 3.658131608843835,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003706435457669784,
      "loss": 5.2545,
      "mean_token_accuracy": 0.20957478135824203,
      "num_tokens": 84831558.0,
      "step": 41860
    },
    {
      "entropy": 6.092548751831055,
      "epoch": 3.6585685571965394,
      "grad_norm": 1.359375,
      "learning_rate": 0.00037061533923722195,
      "loss": 5.2256,
      "mean_token_accuracy": 0.2086321860551834,
      "num_tokens": 84840983.0,
      "step": 41865
    },
    {
      "entropy": 5.973990869522095,
      "epoch": 3.659005505549244,
      "grad_norm": 1.578125,
      "learning_rate": 0.00037058713087351317,
      "loss": 5.1314,
      "mean_token_accuracy": 0.2204068899154663,
      "num_tokens": 84851304.0,
      "step": 41870
    },
    {
      "entropy": 5.975914382934571,
      "epoch": 3.6594424539019488,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00037055892067639314,
      "loss": 5.2326,
      "mean_token_accuracy": 0.20718680173158646,
      "num_tokens": 84862849.0,
      "step": 41875
    },
    {
      "entropy": 5.9990886688232425,
      "epoch": 3.6598794022546537,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037053070864640294,
      "loss": 5.1719,
      "mean_token_accuracy": 0.22066651582717894,
      "num_tokens": 84873591.0,
      "step": 41880
    },
    {
      "entropy": 6.024298238754272,
      "epoch": 3.660316350607358,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037050249478408364,
      "loss": 5.1645,
      "mean_token_accuracy": 0.2176554024219513,
      "num_tokens": 84883701.0,
      "step": 41885
    },
    {
      "entropy": 6.028000450134277,
      "epoch": 3.660753298960063,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003704742790899765,
      "loss": 5.2367,
      "mean_token_accuracy": 0.2055989533662796,
      "num_tokens": 84893842.0,
      "step": 41890
    },
    {
      "entropy": 6.097389650344849,
      "epoch": 3.6611902473127675,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003704460615646227,
      "loss": 5.2922,
      "mean_token_accuracy": 0.19962456524372102,
      "num_tokens": 84903797.0,
      "step": 41895
    },
    {
      "entropy": 6.087984228134156,
      "epoch": 3.6616271956654725,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037041784220856336,
      "loss": 5.3169,
      "mean_token_accuracy": 0.20752194076776503,
      "num_tokens": 84915459.0,
      "step": 41900
    },
    {
      "entropy": 6.04112548828125,
      "epoch": 3.662064144018177,
      "grad_norm": 1.25,
      "learning_rate": 0.0003703896210223399,
      "loss": 5.2555,
      "mean_token_accuracy": 0.20903389453887938,
      "num_tokens": 84926067.0,
      "step": 41905
    },
    {
      "entropy": 6.09320592880249,
      "epoch": 3.662501092370882,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037036139800649336,
      "loss": 5.2361,
      "mean_token_accuracy": 0.21491145044565202,
      "num_tokens": 84936612.0,
      "step": 41910
    },
    {
      "entropy": 5.992341804504394,
      "epoch": 3.6629380407235863,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003703331731615654,
      "loss": 5.1666,
      "mean_token_accuracy": 0.2166937157511711,
      "num_tokens": 84946620.0,
      "step": 41915
    },
    {
      "entropy": 6.065066909790039,
      "epoch": 3.663374989076291,
      "grad_norm": 1.421875,
      "learning_rate": 0.00037030494648809716,
      "loss": 5.19,
      "mean_token_accuracy": 0.2161489188671112,
      "num_tokens": 84955890.0,
      "step": 41920
    },
    {
      "entropy": 5.930993509292603,
      "epoch": 3.6638119374289957,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003702767179866301,
      "loss": 5.1622,
      "mean_token_accuracy": 0.2099052220582962,
      "num_tokens": 84964694.0,
      "step": 41925
    },
    {
      "entropy": 6.039204788208008,
      "epoch": 3.6642488857817006,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003702484876577057,
      "loss": 5.2135,
      "mean_token_accuracy": 0.21408175975084304,
      "num_tokens": 84974203.0,
      "step": 41930
    },
    {
      "entropy": 6.027640247344971,
      "epoch": 3.6646858341344055,
      "grad_norm": 1.34375,
      "learning_rate": 0.00037022025550186537,
      "loss": 5.1516,
      "mean_token_accuracy": 0.21431401818990709,
      "num_tokens": 84984326.0,
      "step": 41935
    },
    {
      "entropy": 6.0504663467407225,
      "epoch": 3.66512278248711,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003701920215196506,
      "loss": 5.1877,
      "mean_token_accuracy": 0.2094019815325737,
      "num_tokens": 84993284.0,
      "step": 41940
    },
    {
      "entropy": 5.974715089797973,
      "epoch": 3.665559730839815,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003701637857116031,
      "loss": 5.1807,
      "mean_token_accuracy": 0.2167542517185211,
      "num_tokens": 85003239.0,
      "step": 41945
    },
    {
      "entropy": 6.000314092636108,
      "epoch": 3.6659966791925194,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003701355480782642,
      "loss": 5.2056,
      "mean_token_accuracy": 0.21558819264173507,
      "num_tokens": 85015493.0,
      "step": 41950
    },
    {
      "entropy": 6.059848737716675,
      "epoch": 3.6664336275452243,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037010730862017575,
      "loss": 5.1906,
      "mean_token_accuracy": 0.21834474056959152,
      "num_tokens": 85025677.0,
      "step": 41955
    },
    {
      "entropy": 6.0593404293060305,
      "epoch": 3.6668705758979288,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037007906733787913,
      "loss": 5.2081,
      "mean_token_accuracy": 0.21190533488988877,
      "num_tokens": 85034908.0,
      "step": 41960
    },
    {
      "entropy": 6.067487668991089,
      "epoch": 3.6673075242506337,
      "grad_norm": 1.25,
      "learning_rate": 0.0003700508242319163,
      "loss": 5.2667,
      "mean_token_accuracy": 0.2010434165596962,
      "num_tokens": 85045416.0,
      "step": 41965
    },
    {
      "entropy": 6.076376628875733,
      "epoch": 3.667744472603338,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037002257930282887,
      "loss": 5.3171,
      "mean_token_accuracy": 0.20490450263023377,
      "num_tokens": 85056259.0,
      "step": 41970
    },
    {
      "entropy": 6.055783653259278,
      "epoch": 3.668181420956043,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003699943325511585,
      "loss": 5.2102,
      "mean_token_accuracy": 0.21044158190488815,
      "num_tokens": 85065579.0,
      "step": 41975
    },
    {
      "entropy": 6.02490816116333,
      "epoch": 3.6686183693087475,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003699660839774471,
      "loss": 5.206,
      "mean_token_accuracy": 0.21692318916320802,
      "num_tokens": 85075084.0,
      "step": 41980
    },
    {
      "entropy": 5.990868711471558,
      "epoch": 3.6690553176614524,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003699378335822364,
      "loss": 5.2356,
      "mean_token_accuracy": 0.21266890466213226,
      "num_tokens": 85084536.0,
      "step": 41985
    },
    {
      "entropy": 6.0544671535491945,
      "epoch": 3.6694922660141573,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003699095813660683,
      "loss": 5.2116,
      "mean_token_accuracy": 0.21417803019285203,
      "num_tokens": 85094076.0,
      "step": 41990
    },
    {
      "entropy": 6.006749010086059,
      "epoch": 3.669929214366862,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003698813273294848,
      "loss": 5.2018,
      "mean_token_accuracy": 0.2074344903230667,
      "num_tokens": 85103784.0,
      "step": 41995
    },
    {
      "entropy": 6.006327533721924,
      "epoch": 3.6703661627195663,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036985307147302765,
      "loss": 5.2202,
      "mean_token_accuracy": 0.21296968162059784,
      "num_tokens": 85114066.0,
      "step": 42000
    },
    {
      "epoch": 3.6703661627195663,
      "eval_entropy": 5.742777971719495,
      "eval_loss": 5.277601718902588,
      "eval_mean_token_accuracy": 0.21693665564691425,
      "eval_num_tokens": 85114066.0,
      "eval_runtime": 21.6179,
      "eval_samples_per_second": 1712.192,
      "eval_steps_per_second": 214.036,
      "step": 42000
    },
    {
      "entropy": 6.049095106124878,
      "epoch": 3.670803111072271,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000369824813797239,
      "loss": 5.2534,
      "mean_token_accuracy": 0.20988155007362366,
      "num_tokens": 85123881.0,
      "step": 42005
    },
    {
      "entropy": 6.076644802093506,
      "epoch": 3.671240059424976,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003697965543026606,
      "loss": 5.2394,
      "mean_token_accuracy": 0.20957112759351731,
      "num_tokens": 85135159.0,
      "step": 42010
    },
    {
      "entropy": 6.103017950057984,
      "epoch": 3.6716770077776806,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003697682929898347,
      "loss": 5.277,
      "mean_token_accuracy": 0.20955872237682344,
      "num_tokens": 85144718.0,
      "step": 42015
    },
    {
      "entropy": 5.977185916900635,
      "epoch": 3.6721139561303855,
      "grad_norm": 1.46875,
      "learning_rate": 0.00036974002985930315,
      "loss": 5.1905,
      "mean_token_accuracy": 0.21558395773172379,
      "num_tokens": 85154094.0,
      "step": 42020
    },
    {
      "entropy": 6.011161279678345,
      "epoch": 3.67255090448309,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003697117649116083,
      "loss": 5.2005,
      "mean_token_accuracy": 0.2070360839366913,
      "num_tokens": 85163636.0,
      "step": 42025
    },
    {
      "entropy": 6.064304637908935,
      "epoch": 3.672987852835795,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003696834981472921,
      "loss": 5.216,
      "mean_token_accuracy": 0.2143724888563156,
      "num_tokens": 85172977.0,
      "step": 42030
    },
    {
      "entropy": 6.006994152069092,
      "epoch": 3.6734248011884993,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036965522956689683,
      "loss": 5.2118,
      "mean_token_accuracy": 0.21756652146577835,
      "num_tokens": 85182956.0,
      "step": 42035
    },
    {
      "entropy": 6.0186625003814695,
      "epoch": 3.6738617495412043,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003696269591709647,
      "loss": 5.3082,
      "mean_token_accuracy": 0.20198829621076583,
      "num_tokens": 85193317.0,
      "step": 42040
    },
    {
      "entropy": 6.054104518890381,
      "epoch": 3.674298697893909,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00036959868696003785,
      "loss": 5.2584,
      "mean_token_accuracy": 0.20936055183410646,
      "num_tokens": 85203724.0,
      "step": 42045
    },
    {
      "entropy": 6.0924793720245365,
      "epoch": 3.6747356462466136,
      "grad_norm": 1.359375,
      "learning_rate": 0.00036957041293465864,
      "loss": 5.251,
      "mean_token_accuracy": 0.2055113658308983,
      "num_tokens": 85213409.0,
      "step": 42050
    },
    {
      "entropy": 6.0824686050415036,
      "epoch": 3.675172594599318,
      "grad_norm": 1.546875,
      "learning_rate": 0.0003695421370953694,
      "loss": 5.2523,
      "mean_token_accuracy": 0.20840615779161453,
      "num_tokens": 85223975.0,
      "step": 42055
    },
    {
      "entropy": 6.034803247451782,
      "epoch": 3.675609542952023,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036951385944271233,
      "loss": 5.2438,
      "mean_token_accuracy": 0.20918044894933702,
      "num_tokens": 85234522.0,
      "step": 42060
    },
    {
      "entropy": 6.039974403381348,
      "epoch": 3.676046491304728,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036948557997722994,
      "loss": 5.2515,
      "mean_token_accuracy": 0.2009799674153328,
      "num_tokens": 85243449.0,
      "step": 42065
    },
    {
      "entropy": 6.021732187271118,
      "epoch": 3.6764834396574324,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036945729869946464,
      "loss": 5.1964,
      "mean_token_accuracy": 0.2132399022579193,
      "num_tokens": 85254328.0,
      "step": 42070
    },
    {
      "entropy": 6.081392621994018,
      "epoch": 3.6769203880101373,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036942901560995886,
      "loss": 5.2114,
      "mean_token_accuracy": 0.2080259308218956,
      "num_tokens": 85264827.0,
      "step": 42075
    },
    {
      "entropy": 6.042212200164795,
      "epoch": 3.677357336362842,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003694007307092551,
      "loss": 5.1619,
      "mean_token_accuracy": 0.21871697008609772,
      "num_tokens": 85274564.0,
      "step": 42080
    },
    {
      "entropy": 5.975891542434693,
      "epoch": 3.6777942847155467,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003693724439978958,
      "loss": 5.1317,
      "mean_token_accuracy": 0.2212807759642601,
      "num_tokens": 85284086.0,
      "step": 42085
    },
    {
      "entropy": 6.0642897605896,
      "epoch": 3.678231233068251,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003693441554764236,
      "loss": 5.2997,
      "mean_token_accuracy": 0.20595401376485825,
      "num_tokens": 85294145.0,
      "step": 42090
    },
    {
      "entropy": 5.903964328765869,
      "epoch": 3.678668181420956,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003693158651453811,
      "loss": 5.12,
      "mean_token_accuracy": 0.22624156773090362,
      "num_tokens": 85304645.0,
      "step": 42095
    },
    {
      "entropy": 6.046127557754517,
      "epoch": 3.679105129773661,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036928757300531083,
      "loss": 5.1336,
      "mean_token_accuracy": 0.21585464030504226,
      "num_tokens": 85313844.0,
      "step": 42100
    },
    {
      "entropy": 6.049688768386841,
      "epoch": 3.6795420781263655,
      "grad_norm": 1.390625,
      "learning_rate": 0.00036925927905675555,
      "loss": 5.2484,
      "mean_token_accuracy": 0.2166331261396408,
      "num_tokens": 85322660.0,
      "step": 42105
    },
    {
      "entropy": 6.07889723777771,
      "epoch": 3.67997902647907,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003692309833002579,
      "loss": 5.255,
      "mean_token_accuracy": 0.21116147935390472,
      "num_tokens": 85334053.0,
      "step": 42110
    },
    {
      "entropy": 6.056627416610718,
      "epoch": 3.680415974831775,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036920268573636055,
      "loss": 5.131,
      "mean_token_accuracy": 0.22243780940771102,
      "num_tokens": 85343890.0,
      "step": 42115
    },
    {
      "entropy": 5.965247440338135,
      "epoch": 3.6808529231844798,
      "grad_norm": 1.421875,
      "learning_rate": 0.00036917438636560645,
      "loss": 5.1822,
      "mean_token_accuracy": 0.21552686840295793,
      "num_tokens": 85353325.0,
      "step": 42120
    },
    {
      "entropy": 6.084542179107666,
      "epoch": 3.6812898715371842,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003691460851885382,
      "loss": 5.325,
      "mean_token_accuracy": 0.1992670089006424,
      "num_tokens": 85363072.0,
      "step": 42125
    },
    {
      "entropy": 6.010311269760132,
      "epoch": 3.681726819889889,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003691177822056987,
      "loss": 5.1428,
      "mean_token_accuracy": 0.21157970726490022,
      "num_tokens": 85372709.0,
      "step": 42130
    },
    {
      "entropy": 6.072068071365356,
      "epoch": 3.6821637682425936,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00036908947741763084,
      "loss": 5.278,
      "mean_token_accuracy": 0.21761274337768555,
      "num_tokens": 85383143.0,
      "step": 42135
    },
    {
      "entropy": 6.018928480148316,
      "epoch": 3.6826007165952985,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003690611708248774,
      "loss": 5.214,
      "mean_token_accuracy": 0.21720583140850067,
      "num_tokens": 85393245.0,
      "step": 42140
    },
    {
      "entropy": 6.001774454116822,
      "epoch": 3.683037664948003,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036903286242798155,
      "loss": 5.1904,
      "mean_token_accuracy": 0.21095338314771653,
      "num_tokens": 85404031.0,
      "step": 42145
    },
    {
      "entropy": 6.026410865783691,
      "epoch": 3.683474613300708,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003690045522274861,
      "loss": 5.2692,
      "mean_token_accuracy": 0.20855225026607513,
      "num_tokens": 85414561.0,
      "step": 42150
    },
    {
      "entropy": 6.076843929290772,
      "epoch": 3.683911561653413,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036897624022393403,
      "loss": 5.3063,
      "mean_token_accuracy": 0.20593679547309876,
      "num_tokens": 85424755.0,
      "step": 42155
    },
    {
      "entropy": 6.054860544204712,
      "epoch": 3.6843485100061173,
      "grad_norm": 1.390625,
      "learning_rate": 0.00036894792641786853,
      "loss": 5.2509,
      "mean_token_accuracy": 0.20923659950494766,
      "num_tokens": 85434086.0,
      "step": 42160
    },
    {
      "entropy": 6.002258682250977,
      "epoch": 3.6847854583588218,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036891961080983245,
      "loss": 5.2093,
      "mean_token_accuracy": 0.20972370803356172,
      "num_tokens": 85443540.0,
      "step": 42165
    },
    {
      "entropy": 6.042018270492553,
      "epoch": 3.6852224067115267,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003688912934003691,
      "loss": 5.2126,
      "mean_token_accuracy": 0.21707681715488433,
      "num_tokens": 85452813.0,
      "step": 42170
    },
    {
      "entropy": 6.069104194641113,
      "epoch": 3.6856593550642316,
      "grad_norm": 1.484375,
      "learning_rate": 0.00036886297419002153,
      "loss": 5.2266,
      "mean_token_accuracy": 0.21377365440130233,
      "num_tokens": 85461838.0,
      "step": 42175
    },
    {
      "entropy": 6.07232174873352,
      "epoch": 3.686096303416936,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003688346531793329,
      "loss": 5.2898,
      "mean_token_accuracy": 0.20672982037067414,
      "num_tokens": 85472272.0,
      "step": 42180
    },
    {
      "entropy": 6.068764925003052,
      "epoch": 3.686533251769641,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003688063303688465,
      "loss": 5.2035,
      "mean_token_accuracy": 0.2158325955271721,
      "num_tokens": 85482663.0,
      "step": 42185
    },
    {
      "entropy": 6.077282333374024,
      "epoch": 3.6869702001223454,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003687780057591055,
      "loss": 5.2446,
      "mean_token_accuracy": 0.20558717399835585,
      "num_tokens": 85492697.0,
      "step": 42190
    },
    {
      "entropy": 6.038280916213989,
      "epoch": 3.6874071484750504,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003687496793506532,
      "loss": 5.1698,
      "mean_token_accuracy": 0.21915393322706223,
      "num_tokens": 85501785.0,
      "step": 42195
    },
    {
      "entropy": 6.025770425796509,
      "epoch": 3.687844096827755,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036872135114403293,
      "loss": 5.1985,
      "mean_token_accuracy": 0.21529507339000703,
      "num_tokens": 85512493.0,
      "step": 42200
    },
    {
      "entropy": 6.041679811477661,
      "epoch": 3.6882810451804597,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000368693021139788,
      "loss": 5.2302,
      "mean_token_accuracy": 0.21227587014436722,
      "num_tokens": 85521118.0,
      "step": 42205
    },
    {
      "entropy": 6.080942630767822,
      "epoch": 3.6887179935331647,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003686646893384619,
      "loss": 5.3005,
      "mean_token_accuracy": 0.205371131002903,
      "num_tokens": 85531048.0,
      "step": 42210
    },
    {
      "entropy": 6.115847969055176,
      "epoch": 3.689154941885869,
      "grad_norm": 1.15625,
      "learning_rate": 0.000368636355740598,
      "loss": 5.188,
      "mean_token_accuracy": 0.20953846126794815,
      "num_tokens": 85541526.0,
      "step": 42215
    },
    {
      "entropy": 6.030698823928833,
      "epoch": 3.6895918902385736,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003686080203467397,
      "loss": 5.1947,
      "mean_token_accuracy": 0.21286594420671462,
      "num_tokens": 85551588.0,
      "step": 42220
    },
    {
      "entropy": 5.989229202270508,
      "epoch": 3.6900288385912785,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003685796831574305,
      "loss": 5.2008,
      "mean_token_accuracy": 0.2127092570066452,
      "num_tokens": 85560775.0,
      "step": 42225
    },
    {
      "entropy": 5.980642366409302,
      "epoch": 3.6904657869439834,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003685513441732139,
      "loss": 5.2306,
      "mean_token_accuracy": 0.20603976249694825,
      "num_tokens": 85571495.0,
      "step": 42230
    },
    {
      "entropy": 6.066733694076538,
      "epoch": 3.690902735296688,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003685230033946336,
      "loss": 5.2149,
      "mean_token_accuracy": 0.21099516302347182,
      "num_tokens": 85583782.0,
      "step": 42235
    },
    {
      "entropy": 6.024044179916382,
      "epoch": 3.691339683649393,
      "grad_norm": 1.328125,
      "learning_rate": 0.000368494660822233,
      "loss": 5.2865,
      "mean_token_accuracy": 0.2080686166882515,
      "num_tokens": 85594342.0,
      "step": 42240
    },
    {
      "entropy": 6.042062997817993,
      "epoch": 3.6917766320020973,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003684663164565558,
      "loss": 5.2351,
      "mean_token_accuracy": 0.2120063379406929,
      "num_tokens": 85604829.0,
      "step": 42245
    },
    {
      "entropy": 6.045735120773315,
      "epoch": 3.692213580354802,
      "grad_norm": 1.375,
      "learning_rate": 0.00036843797029814575,
      "loss": 5.1813,
      "mean_token_accuracy": 0.21543831527233123,
      "num_tokens": 85614530.0,
      "step": 42250
    },
    {
      "entropy": 6.053924942016602,
      "epoch": 3.6926505287075067,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003684096223475464,
      "loss": 5.23,
      "mean_token_accuracy": 0.21234886050224305,
      "num_tokens": 85624110.0,
      "step": 42255
    },
    {
      "entropy": 6.0854614734649655,
      "epoch": 3.6930874770602116,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003683812726053017,
      "loss": 5.2482,
      "mean_token_accuracy": 0.20643004775047302,
      "num_tokens": 85634345.0,
      "step": 42260
    },
    {
      "entropy": 6.018456077575683,
      "epoch": 3.693524425412916,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003683529210719551,
      "loss": 5.197,
      "mean_token_accuracy": 0.21636656075716018,
      "num_tokens": 85644813.0,
      "step": 42265
    },
    {
      "entropy": 6.005322694778442,
      "epoch": 3.693961373765621,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003683245677480506,
      "loss": 5.2625,
      "mean_token_accuracy": 0.20445431768894196,
      "num_tokens": 85655848.0,
      "step": 42270
    },
    {
      "entropy": 6.036537170410156,
      "epoch": 3.6943983221183254,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000368296212634132,
      "loss": 5.2028,
      "mean_token_accuracy": 0.20938337445259095,
      "num_tokens": 85664672.0,
      "step": 42275
    },
    {
      "entropy": 6.094671678543091,
      "epoch": 3.6948352704710303,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003682678557307432,
      "loss": 5.2777,
      "mean_token_accuracy": 0.2001561164855957,
      "num_tokens": 85674179.0,
      "step": 42280
    },
    {
      "entropy": 6.0556597232818605,
      "epoch": 3.6952722188237352,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000368239497038428,
      "loss": 5.2684,
      "mean_token_accuracy": 0.21091051399707794,
      "num_tokens": 85684244.0,
      "step": 42285
    },
    {
      "entropy": 6.071828699111938,
      "epoch": 3.6957091671764397,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003682111365577304,
      "loss": 5.2887,
      "mean_token_accuracy": 0.20525831282138823,
      "num_tokens": 85694998.0,
      "step": 42290
    },
    {
      "entropy": 6.054821538925171,
      "epoch": 3.696146115529144,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036818277428919433,
      "loss": 5.2515,
      "mean_token_accuracy": 0.20037318468093873,
      "num_tokens": 85705083.0,
      "step": 42295
    },
    {
      "entropy": 6.04963903427124,
      "epoch": 3.696583063881849,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003681544102333638,
      "loss": 5.2423,
      "mean_token_accuracy": 0.209916952252388,
      "num_tokens": 85714442.0,
      "step": 42300
    },
    {
      "entropy": 6.064392757415772,
      "epoch": 3.697020012234554,
      "grad_norm": 1.40625,
      "learning_rate": 0.00036812604439078297,
      "loss": 5.1734,
      "mean_token_accuracy": 0.22145805656909942,
      "num_tokens": 85724124.0,
      "step": 42305
    },
    {
      "entropy": 6.053189277648926,
      "epoch": 3.6974569605872585,
      "grad_norm": 1.421875,
      "learning_rate": 0.00036809767676199585,
      "loss": 5.2263,
      "mean_token_accuracy": 0.21145165264606475,
      "num_tokens": 85734271.0,
      "step": 42310
    },
    {
      "entropy": 6.038369417190552,
      "epoch": 3.6978939089399634,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036806930734754634,
      "loss": 5.2568,
      "mean_token_accuracy": 0.21059663444757462,
      "num_tokens": 85744008.0,
      "step": 42315
    },
    {
      "entropy": 6.0360198497772215,
      "epoch": 3.698330857292668,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036804093614797883,
      "loss": 5.2188,
      "mean_token_accuracy": 0.2153938367962837,
      "num_tokens": 85755377.0,
      "step": 42320
    },
    {
      "entropy": 6.042289447784424,
      "epoch": 3.6987678056453728,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003680125631638375,
      "loss": 5.0973,
      "mean_token_accuracy": 0.22211557477712632,
      "num_tokens": 85764640.0,
      "step": 42325
    },
    {
      "entropy": 5.969058084487915,
      "epoch": 3.6992047539980772,
      "grad_norm": 1.421875,
      "learning_rate": 0.00036798418839566633,
      "loss": 5.1559,
      "mean_token_accuracy": 0.21869859248399734,
      "num_tokens": 85774282.0,
      "step": 42330
    },
    {
      "entropy": 5.998814868927002,
      "epoch": 3.699641702350782,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036795581184400977,
      "loss": 5.1852,
      "mean_token_accuracy": 0.21552979797124863,
      "num_tokens": 85783936.0,
      "step": 42335
    },
    {
      "entropy": 6.117876863479614,
      "epoch": 3.700078650703487,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000367927433509412,
      "loss": 5.2764,
      "mean_token_accuracy": 0.2098218470811844,
      "num_tokens": 85794427.0,
      "step": 42340
    },
    {
      "entropy": 6.122747087478638,
      "epoch": 3.7005155990561915,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036789905339241737,
      "loss": 5.3003,
      "mean_token_accuracy": 0.2022644028067589,
      "num_tokens": 85804367.0,
      "step": 42345
    },
    {
      "entropy": 6.002394676208496,
      "epoch": 3.700952547408896,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003678706714935702,
      "loss": 5.1261,
      "mean_token_accuracy": 0.2199840858578682,
      "num_tokens": 85814393.0,
      "step": 42350
    },
    {
      "entropy": 5.986015129089355,
      "epoch": 3.701389495761601,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003678422878134149,
      "loss": 5.2115,
      "mean_token_accuracy": 0.21009011268615724,
      "num_tokens": 85824979.0,
      "step": 42355
    },
    {
      "entropy": 6.075210618972778,
      "epoch": 3.701826444114306,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036781390235249593,
      "loss": 5.2798,
      "mean_token_accuracy": 0.20197650641202927,
      "num_tokens": 85835181.0,
      "step": 42360
    },
    {
      "entropy": 6.0124183177948,
      "epoch": 3.7022633924670103,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003677855151113576,
      "loss": 5.2476,
      "mean_token_accuracy": 0.20591961592435837,
      "num_tokens": 85844720.0,
      "step": 42365
    },
    {
      "entropy": 6.092852258682251,
      "epoch": 3.702700340819715,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003677571260905445,
      "loss": 5.2839,
      "mean_token_accuracy": 0.20702880769968032,
      "num_tokens": 85854085.0,
      "step": 42370
    },
    {
      "entropy": 6.088937473297119,
      "epoch": 3.7031372891724197,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003677287352906012,
      "loss": 5.2388,
      "mean_token_accuracy": 0.20607806742191315,
      "num_tokens": 85864675.0,
      "step": 42375
    },
    {
      "entropy": 6.0371276378631595,
      "epoch": 3.7035742375251246,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036770034271207204,
      "loss": 5.182,
      "mean_token_accuracy": 0.2112310990691185,
      "num_tokens": 85875122.0,
      "step": 42380
    },
    {
      "entropy": 6.045340490341187,
      "epoch": 3.704011185877829,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036767194835550185,
      "loss": 5.1354,
      "mean_token_accuracy": 0.21583855599164964,
      "num_tokens": 85884485.0,
      "step": 42385
    },
    {
      "entropy": 6.017269325256348,
      "epoch": 3.704448134230534,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000367643552221435,
      "loss": 5.2267,
      "mean_token_accuracy": 0.21431389302015305,
      "num_tokens": 85895033.0,
      "step": 42390
    },
    {
      "entropy": 6.044398498535156,
      "epoch": 3.704885082583239,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036761515431041634,
      "loss": 5.1897,
      "mean_token_accuracy": 0.21457207202911377,
      "num_tokens": 85904882.0,
      "step": 42395
    },
    {
      "entropy": 6.088482093811035,
      "epoch": 3.7053220309359434,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00036758675462299044,
      "loss": 5.2362,
      "mean_token_accuracy": 0.20587941855192185,
      "num_tokens": 85914173.0,
      "step": 42400
    },
    {
      "entropy": 6.031416559219361,
      "epoch": 3.705758979288648,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003675583531597022,
      "loss": 5.2517,
      "mean_token_accuracy": 0.20635349154472352,
      "num_tokens": 85923682.0,
      "step": 42405
    },
    {
      "entropy": 6.036140203475952,
      "epoch": 3.7061959276413527,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036752994992109617,
      "loss": 5.193,
      "mean_token_accuracy": 0.20997668206691741,
      "num_tokens": 85934061.0,
      "step": 42410
    },
    {
      "entropy": 6.053922986984253,
      "epoch": 3.7066328759940577,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003675015449077172,
      "loss": 5.2035,
      "mean_token_accuracy": 0.21712505519390107,
      "num_tokens": 85943568.0,
      "step": 42415
    },
    {
      "entropy": 6.074054336547851,
      "epoch": 3.707069824346762,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00036747313812011007,
      "loss": 5.2166,
      "mean_token_accuracy": 0.2121429041028023,
      "num_tokens": 85953998.0,
      "step": 42420
    },
    {
      "entropy": 6.028381061553955,
      "epoch": 3.707506772699467,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003674447295588198,
      "loss": 5.2285,
      "mean_token_accuracy": 0.21301638334989548,
      "num_tokens": 85963520.0,
      "step": 42425
    },
    {
      "entropy": 6.045846748352051,
      "epoch": 3.7079437210521715,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003674163192243912,
      "loss": 5.2225,
      "mean_token_accuracy": 0.20555851757526397,
      "num_tokens": 85974142.0,
      "step": 42430
    },
    {
      "entropy": 6.05089693069458,
      "epoch": 3.7083806694048764,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036738790711736907,
      "loss": 5.2319,
      "mean_token_accuracy": 0.20942284315824508,
      "num_tokens": 85984724.0,
      "step": 42435
    },
    {
      "entropy": 5.994273900985718,
      "epoch": 3.708817617757581,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003673594932382985,
      "loss": 5.1348,
      "mean_token_accuracy": 0.2221067026257515,
      "num_tokens": 85995475.0,
      "step": 42440
    },
    {
      "entropy": 6.006910610198974,
      "epoch": 3.709254566110286,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003673310775877244,
      "loss": 5.2167,
      "mean_token_accuracy": 0.21831232756376268,
      "num_tokens": 86005672.0,
      "step": 42445
    },
    {
      "entropy": 6.0818479537963865,
      "epoch": 3.7096915144629907,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000367302660166192,
      "loss": 5.2628,
      "mean_token_accuracy": 0.2041953280568123,
      "num_tokens": 86016918.0,
      "step": 42450
    },
    {
      "entropy": 6.058469295501709,
      "epoch": 3.710128462815695,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036727424097424615,
      "loss": 5.2544,
      "mean_token_accuracy": 0.2052785873413086,
      "num_tokens": 86029345.0,
      "step": 42455
    },
    {
      "entropy": 6.065759944915771,
      "epoch": 3.7105654111683997,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036724582001243196,
      "loss": 5.2164,
      "mean_token_accuracy": 0.21076728850603105,
      "num_tokens": 86039368.0,
      "step": 42460
    },
    {
      "entropy": 6.074480581283569,
      "epoch": 3.7110023595211046,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003672173972812946,
      "loss": 5.2807,
      "mean_token_accuracy": 0.19674146324396133,
      "num_tokens": 86048220.0,
      "step": 42465
    },
    {
      "entropy": 6.142723608016968,
      "epoch": 3.7114393078738095,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036718897278137933,
      "loss": 5.2749,
      "mean_token_accuracy": 0.20315338522195817,
      "num_tokens": 86058316.0,
      "step": 42470
    },
    {
      "entropy": 6.023688173294067,
      "epoch": 3.711876256226514,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003671605465132312,
      "loss": 5.1981,
      "mean_token_accuracy": 0.20784102976322175,
      "num_tokens": 86068995.0,
      "step": 42475
    },
    {
      "entropy": 5.974783277511596,
      "epoch": 3.712313204579219,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036713211847739553,
      "loss": 5.1906,
      "mean_token_accuracy": 0.2191367492079735,
      "num_tokens": 86080604.0,
      "step": 42480
    },
    {
      "entropy": 6.017577457427978,
      "epoch": 3.7127501529319233,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036710368867441743,
      "loss": 5.2456,
      "mean_token_accuracy": 0.20830973982810974,
      "num_tokens": 86092636.0,
      "step": 42485
    },
    {
      "entropy": 6.046870183944702,
      "epoch": 3.7131871012846283,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003670752571048425,
      "loss": 5.2295,
      "mean_token_accuracy": 0.21275745928287507,
      "num_tokens": 86101454.0,
      "step": 42490
    },
    {
      "entropy": 6.04411211013794,
      "epoch": 3.7136240496373327,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036704682376921576,
      "loss": 5.2554,
      "mean_token_accuracy": 0.21383124738931655,
      "num_tokens": 86111460.0,
      "step": 42495
    },
    {
      "entropy": 6.108601713180542,
      "epoch": 3.7140609979900376,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003670183886680828,
      "loss": 5.257,
      "mean_token_accuracy": 0.2099501222372055,
      "num_tokens": 86121347.0,
      "step": 42500
    },
    {
      "entropy": 6.114222574234009,
      "epoch": 3.7144979463427426,
      "grad_norm": 1.453125,
      "learning_rate": 0.00036698995180198886,
      "loss": 5.3263,
      "mean_token_accuracy": 0.20078923404216767,
      "num_tokens": 86132084.0,
      "step": 42505
    },
    {
      "entropy": 6.0239345073699955,
      "epoch": 3.714934894695447,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036696151317147944,
      "loss": 5.153,
      "mean_token_accuracy": 0.22014871537685393,
      "num_tokens": 86141300.0,
      "step": 42510
    },
    {
      "entropy": 5.991068696975708,
      "epoch": 3.7153718430481515,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036693307277710005,
      "loss": 5.1707,
      "mean_token_accuracy": 0.21620186567306518,
      "num_tokens": 86151640.0,
      "step": 42515
    },
    {
      "entropy": 5.956051063537598,
      "epoch": 3.7158087914008564,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036690463061939615,
      "loss": 5.1201,
      "mean_token_accuracy": 0.21861974745988846,
      "num_tokens": 86162442.0,
      "step": 42520
    },
    {
      "entropy": 6.060593223571777,
      "epoch": 3.7162457397535613,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003668761866989132,
      "loss": 5.2186,
      "mean_token_accuracy": 0.20718226134777068,
      "num_tokens": 86172824.0,
      "step": 42525
    },
    {
      "entropy": 6.029784202575684,
      "epoch": 3.716682688106266,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003668477410161969,
      "loss": 5.1609,
      "mean_token_accuracy": 0.21597643494606017,
      "num_tokens": 86182829.0,
      "step": 42530
    },
    {
      "entropy": 6.018914604187012,
      "epoch": 3.7171196364589707,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036681929357179285,
      "loss": 5.2902,
      "mean_token_accuracy": 0.20762234926223755,
      "num_tokens": 86193656.0,
      "step": 42535
    },
    {
      "entropy": 6.020424795150757,
      "epoch": 3.717556584811675,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003667908443662466,
      "loss": 5.1805,
      "mean_token_accuracy": 0.21919812709093095,
      "num_tokens": 86203025.0,
      "step": 42540
    },
    {
      "entropy": 5.972493076324463,
      "epoch": 3.71799353316438,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036676239340010386,
      "loss": 5.2046,
      "mean_token_accuracy": 0.21109397262334822,
      "num_tokens": 86213029.0,
      "step": 42545
    },
    {
      "entropy": 6.1216428756713865,
      "epoch": 3.7184304815170846,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003667339406739103,
      "loss": 5.2588,
      "mean_token_accuracy": 0.20056635290384292,
      "num_tokens": 86222725.0,
      "step": 42550
    },
    {
      "entropy": 6.079680967330932,
      "epoch": 3.7188674298697895,
      "grad_norm": 1.25,
      "learning_rate": 0.0003667054861882117,
      "loss": 5.2799,
      "mean_token_accuracy": 0.20143490731716157,
      "num_tokens": 86232514.0,
      "step": 42555
    },
    {
      "entropy": 6.1135392665863035,
      "epoch": 3.7193043782224944,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036667702994355386,
      "loss": 5.3539,
      "mean_token_accuracy": 0.2004944235086441,
      "num_tokens": 86242840.0,
      "step": 42560
    },
    {
      "entropy": 6.136734294891357,
      "epoch": 3.719741326575199,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003666485719404825,
      "loss": 5.3427,
      "mean_token_accuracy": 0.19739107936620712,
      "num_tokens": 86253221.0,
      "step": 42565
    },
    {
      "entropy": 6.098230361938477,
      "epoch": 3.7201782749279033,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036662011217954355,
      "loss": 5.2467,
      "mean_token_accuracy": 0.2053757905960083,
      "num_tokens": 86263154.0,
      "step": 42570
    },
    {
      "entropy": 6.08473801612854,
      "epoch": 3.7206152232806082,
      "grad_norm": 1.375,
      "learning_rate": 0.0003665916506612828,
      "loss": 5.3019,
      "mean_token_accuracy": 0.20337700843811035,
      "num_tokens": 86273041.0,
      "step": 42575
    },
    {
      "entropy": 6.01772723197937,
      "epoch": 3.721052171633313,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036656318738624626,
      "loss": 5.2281,
      "mean_token_accuracy": 0.21532215178012848,
      "num_tokens": 86284669.0,
      "step": 42580
    },
    {
      "entropy": 5.977822780609131,
      "epoch": 3.7214891199860176,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003665347223549797,
      "loss": 5.1048,
      "mean_token_accuracy": 0.22088328152894973,
      "num_tokens": 86294186.0,
      "step": 42585
    },
    {
      "entropy": 6.006438875198365,
      "epoch": 3.7219260683387225,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003665062555680293,
      "loss": 5.1836,
      "mean_token_accuracy": 0.2092280462384224,
      "num_tokens": 86304055.0,
      "step": 42590
    },
    {
      "entropy": 6.056908416748047,
      "epoch": 3.722363016691427,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036647778702594087,
      "loss": 5.236,
      "mean_token_accuracy": 0.2101352855563164,
      "num_tokens": 86314258.0,
      "step": 42595
    },
    {
      "entropy": 6.031444931030274,
      "epoch": 3.722799965044132,
      "grad_norm": 1.25,
      "learning_rate": 0.00036644931672926063,
      "loss": 5.1848,
      "mean_token_accuracy": 0.21273193657398223,
      "num_tokens": 86324829.0,
      "step": 42600
    },
    {
      "entropy": 6.054026556015015,
      "epoch": 3.7232369133968364,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003664208446785346,
      "loss": 5.2519,
      "mean_token_accuracy": 0.2074443981051445,
      "num_tokens": 86334901.0,
      "step": 42605
    },
    {
      "entropy": 6.035529947280883,
      "epoch": 3.7236738617495413,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036639237087430875,
      "loss": 5.235,
      "mean_token_accuracy": 0.20841746181249618,
      "num_tokens": 86345798.0,
      "step": 42610
    },
    {
      "entropy": 6.1479198932647705,
      "epoch": 3.7241108101022458,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003663638953171295,
      "loss": 5.3196,
      "mean_token_accuracy": 0.19659346640110015,
      "num_tokens": 86355676.0,
      "step": 42615
    },
    {
      "entropy": 6.107162570953369,
      "epoch": 3.7245477584549507,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036633541800754277,
      "loss": 5.2403,
      "mean_token_accuracy": 0.2115829437971115,
      "num_tokens": 86366590.0,
      "step": 42620
    },
    {
      "entropy": 6.05617847442627,
      "epoch": 3.724984706807655,
      "grad_norm": 1.25,
      "learning_rate": 0.00036630693894609487,
      "loss": 5.2716,
      "mean_token_accuracy": 0.20358838737010956,
      "num_tokens": 86376495.0,
      "step": 42625
    },
    {
      "entropy": 6.010740947723389,
      "epoch": 3.72542165516036,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036627845813333204,
      "loss": 5.2483,
      "mean_token_accuracy": 0.207031774520874,
      "num_tokens": 86386990.0,
      "step": 42630
    },
    {
      "entropy": 6.091272783279419,
      "epoch": 3.725858603513065,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003662499755698006,
      "loss": 5.2249,
      "mean_token_accuracy": 0.20813189148902894,
      "num_tokens": 86396989.0,
      "step": 42635
    },
    {
      "entropy": 6.053733968734742,
      "epoch": 3.7262955518657694,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036622149125604676,
      "loss": 5.2006,
      "mean_token_accuracy": 0.21373213529586793,
      "num_tokens": 86407870.0,
      "step": 42640
    },
    {
      "entropy": 6.018056011199951,
      "epoch": 3.726732500218474,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003661930051926169,
      "loss": 5.1862,
      "mean_token_accuracy": 0.216156105697155,
      "num_tokens": 86417831.0,
      "step": 42645
    },
    {
      "entropy": 6.05137791633606,
      "epoch": 3.727169448571179,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003661645173800574,
      "loss": 5.3308,
      "mean_token_accuracy": 0.2070523202419281,
      "num_tokens": 86428186.0,
      "step": 42650
    },
    {
      "entropy": 6.049516868591309,
      "epoch": 3.7276063969238837,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036613602781891474,
      "loss": 5.2693,
      "mean_token_accuracy": 0.21630705744028092,
      "num_tokens": 86438592.0,
      "step": 42655
    },
    {
      "entropy": 6.047200345993042,
      "epoch": 3.728043345276588,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036610753650973536,
      "loss": 5.1501,
      "mean_token_accuracy": 0.21866484433412553,
      "num_tokens": 86448488.0,
      "step": 42660
    },
    {
      "entropy": 6.052994585037231,
      "epoch": 3.728480293629293,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003660790434530656,
      "loss": 5.2134,
      "mean_token_accuracy": 0.21426158994436265,
      "num_tokens": 86459524.0,
      "step": 42665
    },
    {
      "entropy": 5.9909649848937985,
      "epoch": 3.7289172419819976,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036605054864945213,
      "loss": 5.0716,
      "mean_token_accuracy": 0.22656408697366714,
      "num_tokens": 86470264.0,
      "step": 42670
    },
    {
      "entropy": 6.019729566574097,
      "epoch": 3.7293541903347025,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003660220520994413,
      "loss": 5.2112,
      "mean_token_accuracy": 0.21454489827156067,
      "num_tokens": 86481155.0,
      "step": 42675
    },
    {
      "entropy": 6.100092124938965,
      "epoch": 3.729791138687407,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036599355380358004,
      "loss": 5.2623,
      "mean_token_accuracy": 0.20094027370214462,
      "num_tokens": 86491666.0,
      "step": 42680
    },
    {
      "entropy": 6.0654970645904545,
      "epoch": 3.730228087040112,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036596505376241455,
      "loss": 5.2667,
      "mean_token_accuracy": 0.2097819983959198,
      "num_tokens": 86501953.0,
      "step": 42685
    },
    {
      "entropy": 5.994852876663208,
      "epoch": 3.730665035392817,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003659365519764917,
      "loss": 5.1621,
      "mean_token_accuracy": 0.21805530786514282,
      "num_tokens": 86512658.0,
      "step": 42690
    },
    {
      "entropy": 6.02426586151123,
      "epoch": 3.7311019837455213,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00036590804844635817,
      "loss": 5.3349,
      "mean_token_accuracy": 0.20335616320371627,
      "num_tokens": 86522856.0,
      "step": 42695
    },
    {
      "entropy": 6.085906553268432,
      "epoch": 3.7315389320982257,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036587954317256064,
      "loss": 5.2639,
      "mean_token_accuracy": 0.2045605704188347,
      "num_tokens": 86533650.0,
      "step": 42700
    },
    {
      "entropy": 6.1085440158844,
      "epoch": 3.7319758804509306,
      "grad_norm": 1.25,
      "learning_rate": 0.0003658510361556458,
      "loss": 5.2739,
      "mean_token_accuracy": 0.21152568310499192,
      "num_tokens": 86543801.0,
      "step": 42705
    },
    {
      "entropy": 6.0885943412780765,
      "epoch": 3.7324128288036356,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003658225273961605,
      "loss": 5.2408,
      "mean_token_accuracy": 0.2045816510915756,
      "num_tokens": 86553637.0,
      "step": 42710
    },
    {
      "entropy": 6.056528043746948,
      "epoch": 3.73284977715634,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003657940168946516,
      "loss": 5.2866,
      "mean_token_accuracy": 0.20590812861919403,
      "num_tokens": 86563531.0,
      "step": 42715
    },
    {
      "entropy": 5.98030309677124,
      "epoch": 3.733286725509045,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003657655046516658,
      "loss": 5.1868,
      "mean_token_accuracy": 0.2203774482011795,
      "num_tokens": 86573675.0,
      "step": 42720
    },
    {
      "entropy": 6.028361988067627,
      "epoch": 3.7337236738617494,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036573699066775006,
      "loss": 5.1852,
      "mean_token_accuracy": 0.2077515408396721,
      "num_tokens": 86583817.0,
      "step": 42725
    },
    {
      "entropy": 6.027526330947876,
      "epoch": 3.7341606222144543,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036570847494345134,
      "loss": 5.2468,
      "mean_token_accuracy": 0.20807473808526994,
      "num_tokens": 86593170.0,
      "step": 42730
    },
    {
      "entropy": 6.056983947753906,
      "epoch": 3.734597570567159,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003656799574793166,
      "loss": 5.2751,
      "mean_token_accuracy": 0.20306687355041503,
      "num_tokens": 86603224.0,
      "step": 42735
    },
    {
      "entropy": 6.024529504776001,
      "epoch": 3.7350345189198637,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036565143827589267,
      "loss": 5.195,
      "mean_token_accuracy": 0.21474947929382324,
      "num_tokens": 86613442.0,
      "step": 42740
    },
    {
      "entropy": 6.017944717407227,
      "epoch": 3.7354714672725686,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003656229173337266,
      "loss": 5.1344,
      "mean_token_accuracy": 0.21797655820846557,
      "num_tokens": 86623459.0,
      "step": 42745
    },
    {
      "entropy": 5.960677289962769,
      "epoch": 3.735908415625273,
      "grad_norm": 1.375,
      "learning_rate": 0.0003655943946533656,
      "loss": 5.1226,
      "mean_token_accuracy": 0.22609993070363998,
      "num_tokens": 86633407.0,
      "step": 42750
    },
    {
      "entropy": 6.019325637817383,
      "epoch": 3.7363453639779776,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003655658702353566,
      "loss": 5.2553,
      "mean_token_accuracy": 0.20936144888401031,
      "num_tokens": 86643902.0,
      "step": 42755
    },
    {
      "entropy": 6.053746557235717,
      "epoch": 3.7367823123306825,
      "grad_norm": 1.25,
      "learning_rate": 0.0003655373440802467,
      "loss": 5.1942,
      "mean_token_accuracy": 0.21592872142791747,
      "num_tokens": 86653871.0,
      "step": 42760
    },
    {
      "entropy": 6.029258775711059,
      "epoch": 3.7372192606833874,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003655088161885831,
      "loss": 5.1703,
      "mean_token_accuracy": 0.2200731799006462,
      "num_tokens": 86663890.0,
      "step": 42765
    },
    {
      "entropy": 6.025850582122803,
      "epoch": 3.737656209036092,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003654802865609131,
      "loss": 5.2064,
      "mean_token_accuracy": 0.21582379788160325,
      "num_tokens": 86672954.0,
      "step": 42770
    },
    {
      "entropy": 6.02887749671936,
      "epoch": 3.7380931573887968,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003654517551977837,
      "loss": 5.2899,
      "mean_token_accuracy": 0.2072115197777748,
      "num_tokens": 86682432.0,
      "step": 42775
    },
    {
      "entropy": 6.037314510345459,
      "epoch": 3.7385301057415012,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036542322209974227,
      "loss": 5.2232,
      "mean_token_accuracy": 0.2109897643327713,
      "num_tokens": 86691832.0,
      "step": 42780
    },
    {
      "entropy": 6.054440927505493,
      "epoch": 3.738967054094206,
      "grad_norm": 1.265625,
      "learning_rate": 0.000365394687267336,
      "loss": 5.2951,
      "mean_token_accuracy": 0.20608464777469634,
      "num_tokens": 86703284.0,
      "step": 42785
    },
    {
      "entropy": 6.079169511795044,
      "epoch": 3.7394040024469106,
      "grad_norm": 1.375,
      "learning_rate": 0.00036536615070111234,
      "loss": 5.2935,
      "mean_token_accuracy": 0.21016643494367598,
      "num_tokens": 86713619.0,
      "step": 42790
    },
    {
      "entropy": 6.054791641235352,
      "epoch": 3.7398409507996155,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003653376124016185,
      "loss": 5.2111,
      "mean_token_accuracy": 0.21055977642536164,
      "num_tokens": 86723802.0,
      "step": 42795
    },
    {
      "entropy": 6.047325372695923,
      "epoch": 3.7402778991523205,
      "grad_norm": 1.3125,
      "learning_rate": 0.000365309072369402,
      "loss": 5.229,
      "mean_token_accuracy": 0.2134592726826668,
      "num_tokens": 86733855.0,
      "step": 42800
    },
    {
      "entropy": 6.067286539077759,
      "epoch": 3.740714847505025,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00036528053060501,
      "loss": 5.2068,
      "mean_token_accuracy": 0.21923886835575104,
      "num_tokens": 86742969.0,
      "step": 42805
    },
    {
      "entropy": 6.0284647941589355,
      "epoch": 3.7411517958577294,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003652519871089903,
      "loss": 5.2506,
      "mean_token_accuracy": 0.2087453469634056,
      "num_tokens": 86753749.0,
      "step": 42810
    },
    {
      "entropy": 6.067316007614136,
      "epoch": 3.7415887442104343,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003652234418818901,
      "loss": 5.201,
      "mean_token_accuracy": 0.21342685371637343,
      "num_tokens": 86764287.0,
      "step": 42815
    },
    {
      "entropy": 6.1122079372406,
      "epoch": 3.742025692563139,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036519489492425704,
      "loss": 5.3058,
      "mean_token_accuracy": 0.20745646953582764,
      "num_tokens": 86774711.0,
      "step": 42820
    },
    {
      "entropy": 6.000048065185547,
      "epoch": 3.7424626409158437,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003651663462366387,
      "loss": 5.1648,
      "mean_token_accuracy": 0.22218079268932342,
      "num_tokens": 86784551.0,
      "step": 42825
    },
    {
      "entropy": 5.94578447341919,
      "epoch": 3.7428995892685486,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003651377958195825,
      "loss": 5.0955,
      "mean_token_accuracy": 0.22231801301240922,
      "num_tokens": 86794258.0,
      "step": 42830
    },
    {
      "entropy": 6.05594162940979,
      "epoch": 3.743336537621253,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003651092436736362,
      "loss": 5.1976,
      "mean_token_accuracy": 0.2073347970843315,
      "num_tokens": 86804332.0,
      "step": 42835
    },
    {
      "entropy": 6.035779428482056,
      "epoch": 3.743773485973958,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003650806897993473,
      "loss": 5.1679,
      "mean_token_accuracy": 0.21564140319824218,
      "num_tokens": 86814593.0,
      "step": 42840
    },
    {
      "entropy": 6.050850200653076,
      "epoch": 3.7442104343266625,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036505213419726364,
      "loss": 5.2842,
      "mean_token_accuracy": 0.2014942690730095,
      "num_tokens": 86824489.0,
      "step": 42845
    },
    {
      "entropy": 5.98567214012146,
      "epoch": 3.7446473826793674,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036502357686793284,
      "loss": 5.2161,
      "mean_token_accuracy": 0.22122272551059724,
      "num_tokens": 86835052.0,
      "step": 42850
    },
    {
      "entropy": 6.047826051712036,
      "epoch": 3.7450843310320723,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003649950178119027,
      "loss": 5.1942,
      "mean_token_accuracy": 0.21613613665103912,
      "num_tokens": 86845183.0,
      "step": 42855
    },
    {
      "entropy": 6.040581703186035,
      "epoch": 3.7455212793847767,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003649664570297209,
      "loss": 5.1685,
      "mean_token_accuracy": 0.2187040314078331,
      "num_tokens": 86854752.0,
      "step": 42860
    },
    {
      "entropy": 6.083758878707886,
      "epoch": 3.745958227737481,
      "grad_norm": 1.25,
      "learning_rate": 0.00036493789452193533,
      "loss": 5.2187,
      "mean_token_accuracy": 0.21047768443822862,
      "num_tokens": 86865876.0,
      "step": 42865
    },
    {
      "entropy": 6.031886959075928,
      "epoch": 3.746395176090186,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036490933028909383,
      "loss": 5.249,
      "mean_token_accuracy": 0.21163988411426543,
      "num_tokens": 86875626.0,
      "step": 42870
    },
    {
      "entropy": 6.038241958618164,
      "epoch": 3.746832124442891,
      "grad_norm": 1.375,
      "learning_rate": 0.00036488076433174426,
      "loss": 5.1619,
      "mean_token_accuracy": 0.21282162964344026,
      "num_tokens": 86884944.0,
      "step": 42875
    },
    {
      "entropy": 6.048069477081299,
      "epoch": 3.7472690727955955,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036485219665043463,
      "loss": 5.2433,
      "mean_token_accuracy": 0.21133662164211273,
      "num_tokens": 86894570.0,
      "step": 42880
    },
    {
      "entropy": 6.0510697841644285,
      "epoch": 3.7477060211483004,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003648236272457127,
      "loss": 5.2113,
      "mean_token_accuracy": 0.209675595164299,
      "num_tokens": 86904093.0,
      "step": 42885
    },
    {
      "entropy": 6.02610034942627,
      "epoch": 3.748142969501005,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003647950561181265,
      "loss": 5.2207,
      "mean_token_accuracy": 0.21018666923046112,
      "num_tokens": 86915019.0,
      "step": 42890
    },
    {
      "entropy": 6.0686352252960205,
      "epoch": 3.74857991785371,
      "grad_norm": 1.359375,
      "learning_rate": 0.00036476648326822403,
      "loss": 5.1783,
      "mean_token_accuracy": 0.21732475757598876,
      "num_tokens": 86924695.0,
      "step": 42895
    },
    {
      "entropy": 6.055949306488037,
      "epoch": 3.7490168662064143,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003647379086965535,
      "loss": 5.15,
      "mean_token_accuracy": 0.22314860820770263,
      "num_tokens": 86935030.0,
      "step": 42900
    },
    {
      "entropy": 6.035250091552735,
      "epoch": 3.749453814559119,
      "grad_norm": 1.453125,
      "learning_rate": 0.00036470933240366276,
      "loss": 5.2745,
      "mean_token_accuracy": 0.2112206444144249,
      "num_tokens": 86945871.0,
      "step": 42905
    },
    {
      "entropy": 6.047849225997925,
      "epoch": 3.7498907629118237,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036468075439010007,
      "loss": 5.3925,
      "mean_token_accuracy": 0.20718965232372283,
      "num_tokens": 86958855.0,
      "step": 42910
    },
    {
      "entropy": 6.072424793243409,
      "epoch": 3.7503277112645286,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036465217465641355,
      "loss": 5.2227,
      "mean_token_accuracy": 0.21195410192012787,
      "num_tokens": 86969071.0,
      "step": 42915
    },
    {
      "entropy": 6.043627405166626,
      "epoch": 3.750764659617233,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036462359320315127,
      "loss": 5.2148,
      "mean_token_accuracy": 0.2094041422009468,
      "num_tokens": 86978985.0,
      "step": 42920
    },
    {
      "entropy": 5.977696704864502,
      "epoch": 3.751201607969938,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003645950100308616,
      "loss": 5.1102,
      "mean_token_accuracy": 0.22502971589565277,
      "num_tokens": 86988527.0,
      "step": 42925
    },
    {
      "entropy": 6.038133573532105,
      "epoch": 3.751638556322643,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003645664251400926,
      "loss": 5.2094,
      "mean_token_accuracy": 0.2133934050798416,
      "num_tokens": 86998218.0,
      "step": 42930
    },
    {
      "entropy": 5.956836700439453,
      "epoch": 3.7520755046753473,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036453783853139266,
      "loss": 5.066,
      "mean_token_accuracy": 0.2261677771806717,
      "num_tokens": 87007147.0,
      "step": 42935
    },
    {
      "entropy": 5.971865224838257,
      "epoch": 3.752512453028052,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003645092502053101,
      "loss": 5.176,
      "mean_token_accuracy": 0.22003575116395951,
      "num_tokens": 87016440.0,
      "step": 42940
    },
    {
      "entropy": 5.953205585479736,
      "epoch": 3.7529494013807567,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036448066016239316,
      "loss": 5.1465,
      "mean_token_accuracy": 0.22096165865659714,
      "num_tokens": 87027969.0,
      "step": 42945
    },
    {
      "entropy": 6.127626276016235,
      "epoch": 3.7533863497334616,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003644520684031902,
      "loss": 5.2298,
      "mean_token_accuracy": 0.20652177333831787,
      "num_tokens": 87038403.0,
      "step": 42950
    },
    {
      "entropy": 6.058510780334473,
      "epoch": 3.753823298086166,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036442347492824983,
      "loss": 5.2436,
      "mean_token_accuracy": 0.20970119684934616,
      "num_tokens": 87047484.0,
      "step": 42955
    },
    {
      "entropy": 6.076388883590698,
      "epoch": 3.754260246438871,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003643948797381203,
      "loss": 5.1917,
      "mean_token_accuracy": 0.2141370803117752,
      "num_tokens": 87058151.0,
      "step": 42960
    },
    {
      "entropy": 5.970731401443482,
      "epoch": 3.7546971947915755,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003643662828333502,
      "loss": 5.1376,
      "mean_token_accuracy": 0.21814700216054916,
      "num_tokens": 87068256.0,
      "step": 42965
    },
    {
      "entropy": 6.09999270439148,
      "epoch": 3.7551341431442804,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003643376842144879,
      "loss": 5.3191,
      "mean_token_accuracy": 0.2032817855477333,
      "num_tokens": 87078346.0,
      "step": 42970
    },
    {
      "entropy": 6.071108436584472,
      "epoch": 3.755571091496985,
      "grad_norm": 1.375,
      "learning_rate": 0.00036430908388208205,
      "loss": 5.1806,
      "mean_token_accuracy": 0.20842546969652176,
      "num_tokens": 87088086.0,
      "step": 42975
    },
    {
      "entropy": 6.014772844314575,
      "epoch": 3.75600803984969,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003642804818366811,
      "loss": 5.2038,
      "mean_token_accuracy": 0.21575691998004914,
      "num_tokens": 87098358.0,
      "step": 42980
    },
    {
      "entropy": 6.079755735397339,
      "epoch": 3.7564449882023947,
      "grad_norm": 1.375,
      "learning_rate": 0.00036425187807883377,
      "loss": 5.2732,
      "mean_token_accuracy": 0.20840841233730317,
      "num_tokens": 87108970.0,
      "step": 42985
    },
    {
      "entropy": 6.027895975112915,
      "epoch": 3.756881936555099,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036422327260908873,
      "loss": 5.2247,
      "mean_token_accuracy": 0.21220632493495942,
      "num_tokens": 87117950.0,
      "step": 42990
    },
    {
      "entropy": 6.090703439712525,
      "epoch": 3.7573188849078036,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003641946654279945,
      "loss": 5.3319,
      "mean_token_accuracy": 0.20191295444965363,
      "num_tokens": 87129177.0,
      "step": 42995
    },
    {
      "entropy": 6.05252366065979,
      "epoch": 3.7577558332605085,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003641660565360997,
      "loss": 5.2646,
      "mean_token_accuracy": 0.20744575411081315,
      "num_tokens": 87140418.0,
      "step": 43000
    },
    {
      "entropy": 6.040815305709839,
      "epoch": 3.7581927816132135,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036413744593395345,
      "loss": 5.218,
      "mean_token_accuracy": 0.20594932436943053,
      "num_tokens": 87151624.0,
      "step": 43005
    },
    {
      "entropy": 6.078603458404541,
      "epoch": 3.758629729965918,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00036410883362210416,
      "loss": 5.2495,
      "mean_token_accuracy": 0.20792084336280822,
      "num_tokens": 87160880.0,
      "step": 43010
    },
    {
      "entropy": 6.020456314086914,
      "epoch": 3.759066678318623,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036408021960110084,
      "loss": 5.1734,
      "mean_token_accuracy": 0.215014086663723,
      "num_tokens": 87170673.0,
      "step": 43015
    },
    {
      "entropy": 6.006911897659302,
      "epoch": 3.7595036266713273,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003640516038714921,
      "loss": 5.2117,
      "mean_token_accuracy": 0.2078605607151985,
      "num_tokens": 87180751.0,
      "step": 43020
    },
    {
      "entropy": 6.046215438842774,
      "epoch": 3.7599405750240322,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036402298643382696,
      "loss": 5.212,
      "mean_token_accuracy": 0.21187750697135926,
      "num_tokens": 87191501.0,
      "step": 43025
    },
    {
      "entropy": 6.033499097824096,
      "epoch": 3.7603775233767367,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003639943672886544,
      "loss": 5.2169,
      "mean_token_accuracy": 0.2115768536925316,
      "num_tokens": 87201382.0,
      "step": 43030
    },
    {
      "entropy": 6.04720106124878,
      "epoch": 3.7608144717294416,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036396574643652316,
      "loss": 5.2257,
      "mean_token_accuracy": 0.20482416749000548,
      "num_tokens": 87212454.0,
      "step": 43035
    },
    {
      "entropy": 6.050407361984253,
      "epoch": 3.7612514200821465,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036393712387798236,
      "loss": 5.1985,
      "mean_token_accuracy": 0.2169852375984192,
      "num_tokens": 87222439.0,
      "step": 43040
    },
    {
      "entropy": 6.00274338722229,
      "epoch": 3.761688368434851,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036390849961358084,
      "loss": 5.2003,
      "mean_token_accuracy": 0.21445995271205903,
      "num_tokens": 87232620.0,
      "step": 43045
    },
    {
      "entropy": 6.062399196624756,
      "epoch": 3.7621253167875555,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003638798736438677,
      "loss": 5.2288,
      "mean_token_accuracy": 0.21048687547445297,
      "num_tokens": 87242031.0,
      "step": 43050
    },
    {
      "entropy": 6.031015682220459,
      "epoch": 3.7625622651402604,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036385124596939204,
      "loss": 5.0988,
      "mean_token_accuracy": 0.21675369441509246,
      "num_tokens": 87251113.0,
      "step": 43055
    },
    {
      "entropy": 6.010961675643921,
      "epoch": 3.7629992134929653,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000363822616590703,
      "loss": 5.1667,
      "mean_token_accuracy": 0.21291868835687638,
      "num_tokens": 87261255.0,
      "step": 43060
    },
    {
      "entropy": 6.0872491836547855,
      "epoch": 3.7634361618456698,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003637939855083495,
      "loss": 5.3428,
      "mean_token_accuracy": 0.19289842545986174,
      "num_tokens": 87271677.0,
      "step": 43065
    },
    {
      "entropy": 6.074043130874633,
      "epoch": 3.7638731101983747,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003637653527228809,
      "loss": 5.3244,
      "mean_token_accuracy": 0.20198300778865813,
      "num_tokens": 87280417.0,
      "step": 43070
    },
    {
      "entropy": 6.087635183334351,
      "epoch": 3.764310058551079,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036373671823484636,
      "loss": 5.2793,
      "mean_token_accuracy": 0.20346499979496002,
      "num_tokens": 87291069.0,
      "step": 43075
    },
    {
      "entropy": 6.09610915184021,
      "epoch": 3.764747006903784,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036370808204479497,
      "loss": 5.3161,
      "mean_token_accuracy": 0.1968403548002243,
      "num_tokens": 87301586.0,
      "step": 43080
    },
    {
      "entropy": 6.037109899520874,
      "epoch": 3.7651839552564885,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003636794441532761,
      "loss": 5.177,
      "mean_token_accuracy": 0.21306880861520766,
      "num_tokens": 87312292.0,
      "step": 43085
    },
    {
      "entropy": 6.033791351318359,
      "epoch": 3.7656209036091934,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036365080456083915,
      "loss": 5.2402,
      "mean_token_accuracy": 0.19970375597476958,
      "num_tokens": 87321896.0,
      "step": 43090
    },
    {
      "entropy": 6.037319135665894,
      "epoch": 3.7660578519618984,
      "grad_norm": 1.25,
      "learning_rate": 0.00036362216326803323,
      "loss": 5.2174,
      "mean_token_accuracy": 0.2152736157178879,
      "num_tokens": 87331615.0,
      "step": 43095
    },
    {
      "entropy": 6.0637462615966795,
      "epoch": 3.766494800314603,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003635935202754077,
      "loss": 5.221,
      "mean_token_accuracy": 0.20912442207336426,
      "num_tokens": 87341345.0,
      "step": 43100
    },
    {
      "entropy": 6.068805122375489,
      "epoch": 3.7669317486673073,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036356487558351214,
      "loss": 5.229,
      "mean_token_accuracy": 0.2088442027568817,
      "num_tokens": 87351015.0,
      "step": 43105
    },
    {
      "entropy": 5.97740740776062,
      "epoch": 3.767368697020012,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036353622919289586,
      "loss": 5.2171,
      "mean_token_accuracy": 0.21818896234035492,
      "num_tokens": 87360976.0,
      "step": 43110
    },
    {
      "entropy": 6.035519027709961,
      "epoch": 3.767805645372717,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036350758110410825,
      "loss": 5.1785,
      "mean_token_accuracy": 0.2123965561389923,
      "num_tokens": 87370863.0,
      "step": 43115
    },
    {
      "entropy": 6.046103143692017,
      "epoch": 3.7682425937254216,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00036347893131769894,
      "loss": 5.2012,
      "mean_token_accuracy": 0.20873522907495498,
      "num_tokens": 87380091.0,
      "step": 43120
    },
    {
      "entropy": 5.942766332626343,
      "epoch": 3.7686795420781265,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036345027983421724,
      "loss": 5.1916,
      "mean_token_accuracy": 0.2188097596168518,
      "num_tokens": 87391005.0,
      "step": 43125
    },
    {
      "entropy": 5.981474304199219,
      "epoch": 3.769116490430831,
      "grad_norm": 1.375,
      "learning_rate": 0.0003634216266542129,
      "loss": 5.1789,
      "mean_token_accuracy": 0.21996499598026276,
      "num_tokens": 87400586.0,
      "step": 43130
    },
    {
      "entropy": 6.056295394897461,
      "epoch": 3.769553438783536,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036339297177823544,
      "loss": 5.2194,
      "mean_token_accuracy": 0.21690647751092912,
      "num_tokens": 87411054.0,
      "step": 43135
    },
    {
      "entropy": 6.036070823669434,
      "epoch": 3.7699903871362404,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003633643152068344,
      "loss": 5.227,
      "mean_token_accuracy": 0.2114200010895729,
      "num_tokens": 87421413.0,
      "step": 43140
    },
    {
      "entropy": 6.048441457748413,
      "epoch": 3.7704273354889453,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036333565694055957,
      "loss": 5.2034,
      "mean_token_accuracy": 0.21068174839019777,
      "num_tokens": 87431592.0,
      "step": 43145
    },
    {
      "entropy": 6.032931756973267,
      "epoch": 3.77086428384165,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036330699697996043,
      "loss": 5.24,
      "mean_token_accuracy": 0.209736867249012,
      "num_tokens": 87441708.0,
      "step": 43150
    },
    {
      "entropy": 6.037705612182617,
      "epoch": 3.7713012321943546,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00036327833532558704,
      "loss": 5.2649,
      "mean_token_accuracy": 0.2065716490149498,
      "num_tokens": 87452083.0,
      "step": 43155
    },
    {
      "entropy": 6.032464122772216,
      "epoch": 3.771738180547059,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003632496719779888,
      "loss": 5.233,
      "mean_token_accuracy": 0.20203275084495545,
      "num_tokens": 87460743.0,
      "step": 43160
    },
    {
      "entropy": 6.084289503097534,
      "epoch": 3.772175128899764,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036322100693771554,
      "loss": 5.2432,
      "mean_token_accuracy": 0.20716746002435685,
      "num_tokens": 87470865.0,
      "step": 43165
    },
    {
      "entropy": 6.057729005813599,
      "epoch": 3.772612077252469,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036319234020531723,
      "loss": 5.2725,
      "mean_token_accuracy": 0.2083009362220764,
      "num_tokens": 87482063.0,
      "step": 43170
    },
    {
      "entropy": 6.0582465648651125,
      "epoch": 3.7730490256051734,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003631636717813435,
      "loss": 5.2345,
      "mean_token_accuracy": 0.20736054033041001,
      "num_tokens": 87492284.0,
      "step": 43175
    },
    {
      "entropy": 6.041291427612305,
      "epoch": 3.7734859739578783,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003631350016663445,
      "loss": 5.2675,
      "mean_token_accuracy": 0.204831326007843,
      "num_tokens": 87501586.0,
      "step": 43180
    },
    {
      "entropy": 6.082434511184692,
      "epoch": 3.773922922310583,
      "grad_norm": 1.4375,
      "learning_rate": 0.00036310632986086993,
      "loss": 5.2835,
      "mean_token_accuracy": 0.20348680168390273,
      "num_tokens": 87510622.0,
      "step": 43185
    },
    {
      "entropy": 6.056958341598511,
      "epoch": 3.7743598706632877,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003630776563654698,
      "loss": 5.2652,
      "mean_token_accuracy": 0.20603003054857255,
      "num_tokens": 87519480.0,
      "step": 43190
    },
    {
      "entropy": 6.080482816696167,
      "epoch": 3.774796819015992,
      "grad_norm": 1.34375,
      "learning_rate": 0.000363048981180694,
      "loss": 5.2425,
      "mean_token_accuracy": 0.2093115285038948,
      "num_tokens": 87529560.0,
      "step": 43195
    },
    {
      "entropy": 6.093604612350464,
      "epoch": 3.775233767368697,
      "grad_norm": 1.40625,
      "learning_rate": 0.00036302030430709263,
      "loss": 5.3072,
      "mean_token_accuracy": 0.20199434608221054,
      "num_tokens": 87539969.0,
      "step": 43200
    },
    {
      "entropy": 6.049832916259765,
      "epoch": 3.775670715721402,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036299162574521585,
      "loss": 5.2415,
      "mean_token_accuracy": 0.20989379435777664,
      "num_tokens": 87550883.0,
      "step": 43205
    },
    {
      "entropy": 6.045687484741211,
      "epoch": 3.7761076640741065,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036296294549561343,
      "loss": 5.1951,
      "mean_token_accuracy": 0.2155774250626564,
      "num_tokens": 87560569.0,
      "step": 43210
    },
    {
      "entropy": 6.058323192596435,
      "epoch": 3.776544612426811,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003629342635588356,
      "loss": 5.2935,
      "mean_token_accuracy": 0.20947330445051193,
      "num_tokens": 87570881.0,
      "step": 43215
    },
    {
      "entropy": 6.023557519912719,
      "epoch": 3.776981560779516,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036290557993543253,
      "loss": 5.1183,
      "mean_token_accuracy": 0.2342497944831848,
      "num_tokens": 87581334.0,
      "step": 43220
    },
    {
      "entropy": 6.071115255355835,
      "epoch": 3.7774185091322208,
      "grad_norm": 1.359375,
      "learning_rate": 0.00036287689462595447,
      "loss": 5.2307,
      "mean_token_accuracy": 0.20632094144821167,
      "num_tokens": 87591554.0,
      "step": 43225
    },
    {
      "entropy": 6.009714412689209,
      "epoch": 3.7778554574849252,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003628482076309515,
      "loss": 5.1375,
      "mean_token_accuracy": 0.2195669010281563,
      "num_tokens": 87600484.0,
      "step": 43230
    },
    {
      "entropy": 6.032329511642456,
      "epoch": 3.77829240583763,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036281951895097384,
      "loss": 5.2578,
      "mean_token_accuracy": 0.2104356363415718,
      "num_tokens": 87610632.0,
      "step": 43235
    },
    {
      "entropy": 5.979619789123535,
      "epoch": 3.7787293541903346,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036279082858657174,
      "loss": 5.0999,
      "mean_token_accuracy": 0.22503851503133773,
      "num_tokens": 87620835.0,
      "step": 43240
    },
    {
      "entropy": 6.091977643966675,
      "epoch": 3.7791663025430395,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003627621365382956,
      "loss": 5.2687,
      "mean_token_accuracy": 0.2056325390934944,
      "num_tokens": 87630437.0,
      "step": 43245
    },
    {
      "entropy": 6.103287267684936,
      "epoch": 3.779603250895744,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036273344280669573,
      "loss": 5.3011,
      "mean_token_accuracy": 0.2007213458418846,
      "num_tokens": 87640293.0,
      "step": 43250
    },
    {
      "entropy": 6.0547887802124025,
      "epoch": 3.780040199248449,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036270474739232244,
      "loss": 5.2019,
      "mean_token_accuracy": 0.2082797035574913,
      "num_tokens": 87651026.0,
      "step": 43255
    },
    {
      "entropy": 6.034888839721679,
      "epoch": 3.7804771476011534,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036267605029572617,
      "loss": 5.2234,
      "mean_token_accuracy": 0.2111925572156906,
      "num_tokens": 87660957.0,
      "step": 43260
    },
    {
      "entropy": 6.045566987991333,
      "epoch": 3.7809140959538583,
      "grad_norm": 1.375,
      "learning_rate": 0.0003626473515174572,
      "loss": 5.2024,
      "mean_token_accuracy": 0.2165924444794655,
      "num_tokens": 87671311.0,
      "step": 43265
    },
    {
      "entropy": 6.0231434345245365,
      "epoch": 3.7813510443065628,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003626186510580661,
      "loss": 5.1519,
      "mean_token_accuracy": 0.21847661286592485,
      "num_tokens": 87681217.0,
      "step": 43270
    },
    {
      "entropy": 6.008683156967163,
      "epoch": 3.7817879926592677,
      "grad_norm": 1.59375,
      "learning_rate": 0.00036258994891810357,
      "loss": 5.2021,
      "mean_token_accuracy": 0.2166168987751007,
      "num_tokens": 87691014.0,
      "step": 43275
    },
    {
      "entropy": 6.078188037872314,
      "epoch": 3.7822249410119726,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003625612450981197,
      "loss": 5.2955,
      "mean_token_accuracy": 0.20570261180400848,
      "num_tokens": 87701125.0,
      "step": 43280
    },
    {
      "entropy": 5.991622257232666,
      "epoch": 3.782661889364677,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036253253959866533,
      "loss": 5.0537,
      "mean_token_accuracy": 0.225146347284317,
      "num_tokens": 87711451.0,
      "step": 43285
    },
    {
      "entropy": 6.096213674545288,
      "epoch": 3.7830988377173815,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000362503832420291,
      "loss": 5.2767,
      "mean_token_accuracy": 0.20886721462011337,
      "num_tokens": 87722753.0,
      "step": 43290
    },
    {
      "entropy": 6.0334526062011715,
      "epoch": 3.7835357860700864,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003624751235635473,
      "loss": 5.1921,
      "mean_token_accuracy": 0.20906418114900588,
      "num_tokens": 87732771.0,
      "step": 43295
    },
    {
      "entropy": 6.077543306350708,
      "epoch": 3.7839727344227914,
      "grad_norm": 1.203125,
      "learning_rate": 0.000362446413028985,
      "loss": 5.2765,
      "mean_token_accuracy": 0.20488071888685228,
      "num_tokens": 87742925.0,
      "step": 43300
    },
    {
      "entropy": 6.032745361328125,
      "epoch": 3.784409682775496,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036241770081715455,
      "loss": 5.2279,
      "mean_token_accuracy": 0.21640223115682602,
      "num_tokens": 87752700.0,
      "step": 43305
    },
    {
      "entropy": 6.001256036758423,
      "epoch": 3.7848466311282007,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003623889869286067,
      "loss": 5.1848,
      "mean_token_accuracy": 0.2133003830909729,
      "num_tokens": 87764054.0,
      "step": 43310
    },
    {
      "entropy": 6.050353717803955,
      "epoch": 3.785283579480905,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003623602713638925,
      "loss": 5.1788,
      "mean_token_accuracy": 0.2189643695950508,
      "num_tokens": 87774861.0,
      "step": 43315
    },
    {
      "entropy": 6.046918201446533,
      "epoch": 3.78572052783361,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00036233155412356244,
      "loss": 5.1756,
      "mean_token_accuracy": 0.21352865248918534,
      "num_tokens": 87785322.0,
      "step": 43320
    },
    {
      "entropy": 6.077304697036743,
      "epoch": 3.7861574761863146,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036230283520816744,
      "loss": 5.2248,
      "mean_token_accuracy": 0.2143728867173195,
      "num_tokens": 87795357.0,
      "step": 43325
    },
    {
      "entropy": 6.00512547492981,
      "epoch": 3.7865944245390195,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003622741146182583,
      "loss": 5.2145,
      "mean_token_accuracy": 0.2187405839562416,
      "num_tokens": 87805642.0,
      "step": 43330
    },
    {
      "entropy": 6.019071960449219,
      "epoch": 3.7870313728917244,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003622453923543858,
      "loss": 5.136,
      "mean_token_accuracy": 0.22552093863487244,
      "num_tokens": 87815700.0,
      "step": 43335
    },
    {
      "entropy": 6.026497507095337,
      "epoch": 3.787468321244429,
      "grad_norm": 1.328125,
      "learning_rate": 0.000362216668417101,
      "loss": 5.2238,
      "mean_token_accuracy": 0.20977207273244858,
      "num_tokens": 87826121.0,
      "step": 43340
    },
    {
      "entropy": 5.999587059020996,
      "epoch": 3.7879052695971334,
      "grad_norm": 1.453125,
      "learning_rate": 0.00036218794280695494,
      "loss": 5.1954,
      "mean_token_accuracy": 0.21072673350572585,
      "num_tokens": 87838152.0,
      "step": 43345
    },
    {
      "entropy": 6.077569103240966,
      "epoch": 3.7883422179498383,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003621592155244984,
      "loss": 5.2533,
      "mean_token_accuracy": 0.20852328091859818,
      "num_tokens": 87848657.0,
      "step": 43350
    },
    {
      "entropy": 6.098685646057129,
      "epoch": 3.788779166302543,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036213048657028233,
      "loss": 5.3171,
      "mean_token_accuracy": 0.20573932230472564,
      "num_tokens": 87858603.0,
      "step": 43355
    },
    {
      "entropy": 6.067118310928345,
      "epoch": 3.7892161146552477,
      "grad_norm": 1.421875,
      "learning_rate": 0.00036210175594485796,
      "loss": 5.226,
      "mean_token_accuracy": 0.21429725736379623,
      "num_tokens": 87868091.0,
      "step": 43360
    },
    {
      "entropy": 6.038251829147339,
      "epoch": 3.7896530630079526,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036207302364877627,
      "loss": 5.1665,
      "mean_token_accuracy": 0.2141103208065033,
      "num_tokens": 87878978.0,
      "step": 43365
    },
    {
      "entropy": 6.0410463333129885,
      "epoch": 3.790090011360657,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036204428968258836,
      "loss": 5.2599,
      "mean_token_accuracy": 0.20293009877204896,
      "num_tokens": 87888547.0,
      "step": 43370
    },
    {
      "entropy": 6.016224336624146,
      "epoch": 3.790526959713362,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036201555404684533,
      "loss": 5.212,
      "mean_token_accuracy": 0.21536038368940352,
      "num_tokens": 87898395.0,
      "step": 43375
    },
    {
      "entropy": 5.986570072174072,
      "epoch": 3.7909639080660664,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003619868167420984,
      "loss": 5.1931,
      "mean_token_accuracy": 0.21634212881326675,
      "num_tokens": 87909127.0,
      "step": 43380
    },
    {
      "entropy": 6.118590545654297,
      "epoch": 3.7914008564187713,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036195807776889864,
      "loss": 5.2639,
      "mean_token_accuracy": 0.20770460665225982,
      "num_tokens": 87919554.0,
      "step": 43385
    },
    {
      "entropy": 6.045253610610962,
      "epoch": 3.7918378047714763,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036192933712779756,
      "loss": 5.197,
      "mean_token_accuracy": 0.20842131823301316,
      "num_tokens": 87930728.0,
      "step": 43390
    },
    {
      "entropy": 5.959505558013916,
      "epoch": 3.7922747531241807,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00036190059481934614,
      "loss": 5.1227,
      "mean_token_accuracy": 0.21894484311342238,
      "num_tokens": 87939928.0,
      "step": 43395
    },
    {
      "entropy": 6.003358554840088,
      "epoch": 3.792711701476885,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003618718508440958,
      "loss": 5.2548,
      "mean_token_accuracy": 0.20414025634527205,
      "num_tokens": 87949978.0,
      "step": 43400
    },
    {
      "entropy": 6.1059516906738285,
      "epoch": 3.79314864982959,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003618431052025978,
      "loss": 5.2987,
      "mean_token_accuracy": 0.2080145686864853,
      "num_tokens": 87960157.0,
      "step": 43405
    },
    {
      "entropy": 6.065590238571167,
      "epoch": 3.793585598182295,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003618143578954036,
      "loss": 5.1726,
      "mean_token_accuracy": 0.21649355739355086,
      "num_tokens": 87971302.0,
      "step": 43410
    },
    {
      "entropy": 6.039089775085449,
      "epoch": 3.7940225465349995,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003617856089230645,
      "loss": 5.1545,
      "mean_token_accuracy": 0.21799081861972808,
      "num_tokens": 87980829.0,
      "step": 43415
    },
    {
      "entropy": 5.97836594581604,
      "epoch": 3.7944594948877044,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036175685828613197,
      "loss": 5.1699,
      "mean_token_accuracy": 0.21626102477312087,
      "num_tokens": 87991003.0,
      "step": 43420
    },
    {
      "entropy": 6.03496823310852,
      "epoch": 3.794896443240409,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003617281059851574,
      "loss": 5.1802,
      "mean_token_accuracy": 0.21765944808721543,
      "num_tokens": 88001475.0,
      "step": 43425
    },
    {
      "entropy": 6.0869713306427,
      "epoch": 3.795333391593114,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003616993520206923,
      "loss": 5.309,
      "mean_token_accuracy": 0.2039695993065834,
      "num_tokens": 88012714.0,
      "step": 43430
    },
    {
      "entropy": 6.013730144500732,
      "epoch": 3.7957703399458183,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036167059639328824,
      "loss": 5.2553,
      "mean_token_accuracy": 0.20942299365997313,
      "num_tokens": 88023195.0,
      "step": 43435
    },
    {
      "entropy": 6.006862783432007,
      "epoch": 3.796207288298523,
      "grad_norm": 1.40625,
      "learning_rate": 0.00036164183910349676,
      "loss": 5.181,
      "mean_token_accuracy": 0.21868284046649933,
      "num_tokens": 88032977.0,
      "step": 43440
    },
    {
      "entropy": 6.024939107894897,
      "epoch": 3.796644236651228,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003616130801518694,
      "loss": 5.2068,
      "mean_token_accuracy": 0.2112330675125122,
      "num_tokens": 88041761.0,
      "step": 43445
    },
    {
      "entropy": 5.990121459960937,
      "epoch": 3.7970811850039325,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003615843195389577,
      "loss": 5.1812,
      "mean_token_accuracy": 0.21512240916490555,
      "num_tokens": 88050709.0,
      "step": 43450
    },
    {
      "entropy": 6.037918281555176,
      "epoch": 3.797518133356637,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036155555726531343,
      "loss": 5.2666,
      "mean_token_accuracy": 0.2076135218143463,
      "num_tokens": 88061381.0,
      "step": 43455
    },
    {
      "entropy": 6.073457956314087,
      "epoch": 3.797955081709342,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00036152679333148816,
      "loss": 5.2562,
      "mean_token_accuracy": 0.21313771456480027,
      "num_tokens": 88070926.0,
      "step": 43460
    },
    {
      "entropy": 6.125643682479859,
      "epoch": 3.798392030062047,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036149802773803374,
      "loss": 5.2958,
      "mean_token_accuracy": 0.21004430949687958,
      "num_tokens": 88081240.0,
      "step": 43465
    },
    {
      "entropy": 6.105504846572876,
      "epoch": 3.7988289784147513,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003614692604855018,
      "loss": 5.3333,
      "mean_token_accuracy": 0.1988932177424431,
      "num_tokens": 88092623.0,
      "step": 43470
    },
    {
      "entropy": 6.0559807300567625,
      "epoch": 3.7992659267674562,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003614404915744441,
      "loss": 5.1928,
      "mean_token_accuracy": 0.2128276363015175,
      "num_tokens": 88102532.0,
      "step": 43475
    },
    {
      "entropy": 6.079052495956421,
      "epoch": 3.7997028751201607,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036141172100541245,
      "loss": 5.3252,
      "mean_token_accuracy": 0.19915588200092316,
      "num_tokens": 88112210.0,
      "step": 43480
    },
    {
      "entropy": 6.093558263778687,
      "epoch": 3.8001398234728656,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036138294877895876,
      "loss": 5.2335,
      "mean_token_accuracy": 0.21078507155179976,
      "num_tokens": 88121800.0,
      "step": 43485
    },
    {
      "entropy": 6.0751410007476805,
      "epoch": 3.80057677182557,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003613541748956349,
      "loss": 5.2036,
      "mean_token_accuracy": 0.21048066020011902,
      "num_tokens": 88131820.0,
      "step": 43490
    },
    {
      "entropy": 6.04481987953186,
      "epoch": 3.801013720178275,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003613253993559926,
      "loss": 5.2688,
      "mean_token_accuracy": 0.2078562930226326,
      "num_tokens": 88141799.0,
      "step": 43495
    },
    {
      "entropy": 6.0259791851043705,
      "epoch": 3.80145066853098,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036129662216058395,
      "loss": 5.1835,
      "mean_token_accuracy": 0.2156072661280632,
      "num_tokens": 88151860.0,
      "step": 43500
    },
    {
      "entropy": 6.043440151214599,
      "epoch": 3.8018876168836844,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003612678433099609,
      "loss": 5.183,
      "mean_token_accuracy": 0.20905645340681075,
      "num_tokens": 88160580.0,
      "step": 43505
    },
    {
      "entropy": 6.025609254837036,
      "epoch": 3.802324565236389,
      "grad_norm": 1.40625,
      "learning_rate": 0.00036123906280467543,
      "loss": 5.2768,
      "mean_token_accuracy": 0.20419339537620546,
      "num_tokens": 88170140.0,
      "step": 43510
    },
    {
      "entropy": 6.04361572265625,
      "epoch": 3.8027615135890938,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003612102806452795,
      "loss": 5.3125,
      "mean_token_accuracy": 0.21203400194644928,
      "num_tokens": 88180288.0,
      "step": 43515
    },
    {
      "entropy": 6.028065013885498,
      "epoch": 3.8031984619417987,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003611814968323252,
      "loss": 5.2165,
      "mean_token_accuracy": 0.20808786600828172,
      "num_tokens": 88190046.0,
      "step": 43520
    },
    {
      "entropy": 6.0531689643859865,
      "epoch": 3.803635410294503,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003611527113663647,
      "loss": 5.1755,
      "mean_token_accuracy": 0.2144423708319664,
      "num_tokens": 88199321.0,
      "step": 43525
    },
    {
      "entropy": 6.059977674484253,
      "epoch": 3.804072358647208,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036112392424794996,
      "loss": 5.2632,
      "mean_token_accuracy": 0.20096899271011354,
      "num_tokens": 88209277.0,
      "step": 43530
    },
    {
      "entropy": 6.04776291847229,
      "epoch": 3.8045093069999125,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003610951354776333,
      "loss": 5.3022,
      "mean_token_accuracy": 0.20214873552322388,
      "num_tokens": 88219664.0,
      "step": 43535
    },
    {
      "entropy": 6.038791561126709,
      "epoch": 3.8049462553526174,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036106634505596693,
      "loss": 5.2526,
      "mean_token_accuracy": 0.20918381363153457,
      "num_tokens": 88230058.0,
      "step": 43540
    },
    {
      "entropy": 6.047621822357177,
      "epoch": 3.805383203705322,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036103755298350277,
      "loss": 5.1653,
      "mean_token_accuracy": 0.21719672679901122,
      "num_tokens": 88240861.0,
      "step": 43545
    },
    {
      "entropy": 6.087436103820801,
      "epoch": 3.805820152058027,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036100875926079335,
      "loss": 5.2864,
      "mean_token_accuracy": 0.2015279933810234,
      "num_tokens": 88250833.0,
      "step": 43550
    },
    {
      "entropy": 6.0368592739105225,
      "epoch": 3.8062571004107313,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003609799638883908,
      "loss": 5.1642,
      "mean_token_accuracy": 0.21476849913597107,
      "num_tokens": 88260209.0,
      "step": 43555
    },
    {
      "entropy": 6.114438438415528,
      "epoch": 3.806694048763436,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003609511668668476,
      "loss": 5.2919,
      "mean_token_accuracy": 0.1983232617378235,
      "num_tokens": 88269961.0,
      "step": 43560
    },
    {
      "entropy": 6.024170160293579,
      "epoch": 3.8071309971161407,
      "grad_norm": 1.375,
      "learning_rate": 0.00036092236819671586,
      "loss": 5.2795,
      "mean_token_accuracy": 0.19988961666822433,
      "num_tokens": 88279916.0,
      "step": 43565
    },
    {
      "entropy": 6.01718692779541,
      "epoch": 3.8075679454688456,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036089356787854816,
      "loss": 5.2585,
      "mean_token_accuracy": 0.20816446244716644,
      "num_tokens": 88290459.0,
      "step": 43570
    },
    {
      "entropy": 6.046852779388428,
      "epoch": 3.8080048938215505,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003608647659128968,
      "loss": 5.2108,
      "mean_token_accuracy": 0.21150607913732528,
      "num_tokens": 88300536.0,
      "step": 43575
    },
    {
      "entropy": 6.085330057144165,
      "epoch": 3.808441842174255,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00036083596230031425,
      "loss": 5.292,
      "mean_token_accuracy": 0.21050371825695038,
      "num_tokens": 88310654.0,
      "step": 43580
    },
    {
      "entropy": 6.015332221984863,
      "epoch": 3.80887879052696,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003608071570413529,
      "loss": 5.2015,
      "mean_token_accuracy": 0.22053583711385727,
      "num_tokens": 88319831.0,
      "step": 43585
    },
    {
      "entropy": 6.1218468189239506,
      "epoch": 3.8093157388796643,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003607783501365653,
      "loss": 5.3642,
      "mean_token_accuracy": 0.20031755566596984,
      "num_tokens": 88330931.0,
      "step": 43590
    },
    {
      "entropy": 6.092230892181396,
      "epoch": 3.8097526872323693,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003607495415865041,
      "loss": 5.2307,
      "mean_token_accuracy": 0.21138605177402497,
      "num_tokens": 88341699.0,
      "step": 43595
    },
    {
      "entropy": 6.0446068286895756,
      "epoch": 3.8101896355850737,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00036072073139172174,
      "loss": 5.257,
      "mean_token_accuracy": 0.20875178426504135,
      "num_tokens": 88351375.0,
      "step": 43600
    },
    {
      "entropy": 6.019933462142944,
      "epoch": 3.8106265839377786,
      "grad_norm": 1.40625,
      "learning_rate": 0.00036069191955277085,
      "loss": 5.2393,
      "mean_token_accuracy": 0.21724432557821274,
      "num_tokens": 88361865.0,
      "step": 43605
    },
    {
      "entropy": 6.032391786575317,
      "epoch": 3.811063532290483,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000360663106070204,
      "loss": 5.1978,
      "mean_token_accuracy": 0.21289519369602203,
      "num_tokens": 88371276.0,
      "step": 43610
    },
    {
      "entropy": 6.01515827178955,
      "epoch": 3.811500480643188,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003606342909445739,
      "loss": 5.1939,
      "mean_token_accuracy": 0.21595952957868575,
      "num_tokens": 88382098.0,
      "step": 43615
    },
    {
      "entropy": 6.077403020858765,
      "epoch": 3.8119374289958925,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003606054741764332,
      "loss": 5.2082,
      "mean_token_accuracy": 0.21154737025499343,
      "num_tokens": 88391488.0,
      "step": 43620
    },
    {
      "entropy": 6.061413288116455,
      "epoch": 3.8123743773485974,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036057665576633466,
      "loss": 5.2271,
      "mean_token_accuracy": 0.20861591249704362,
      "num_tokens": 88401636.0,
      "step": 43625
    },
    {
      "entropy": 5.993462085723877,
      "epoch": 3.8128113257013023,
      "grad_norm": 1.265625,
      "learning_rate": 0.000360547835714831,
      "loss": 5.1677,
      "mean_token_accuracy": 0.2243288442492485,
      "num_tokens": 88412434.0,
      "step": 43630
    },
    {
      "entropy": 6.03119854927063,
      "epoch": 3.813248274054007,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003605190140224751,
      "loss": 5.2525,
      "mean_token_accuracy": 0.19896797239780425,
      "num_tokens": 88421995.0,
      "step": 43635
    },
    {
      "entropy": 6.004465818405151,
      "epoch": 3.8136852224067113,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036049019068981966,
      "loss": 5.1387,
      "mean_token_accuracy": 0.22683893889188766,
      "num_tokens": 88431691.0,
      "step": 43640
    },
    {
      "entropy": 6.018419027328491,
      "epoch": 3.814122170759416,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003604613657174175,
      "loss": 5.2321,
      "mean_token_accuracy": 0.21640366911888123,
      "num_tokens": 88441284.0,
      "step": 43645
    },
    {
      "entropy": 5.973750591278076,
      "epoch": 3.814559119112121,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003604325391058216,
      "loss": 5.2203,
      "mean_token_accuracy": 0.214626444876194,
      "num_tokens": 88450674.0,
      "step": 43650
    },
    {
      "entropy": 6.043655157089233,
      "epoch": 3.8149960674648256,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003604037108555849,
      "loss": 5.2702,
      "mean_token_accuracy": 0.20236847847700118,
      "num_tokens": 88461082.0,
      "step": 43655
    },
    {
      "entropy": 6.102325868606568,
      "epoch": 3.8154330158175305,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003603748809672602,
      "loss": 5.283,
      "mean_token_accuracy": 0.20667562931776046,
      "num_tokens": 88471229.0,
      "step": 43660
    },
    {
      "entropy": 6.102627992630005,
      "epoch": 3.815869964170235,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036034604944140057,
      "loss": 5.319,
      "mean_token_accuracy": 0.20058023780584336,
      "num_tokens": 88481269.0,
      "step": 43665
    },
    {
      "entropy": 6.079911994934082,
      "epoch": 3.81630691252294,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036031721627855905,
      "loss": 5.1886,
      "mean_token_accuracy": 0.21245401948690415,
      "num_tokens": 88490754.0,
      "step": 43670
    },
    {
      "entropy": 6.013193130493164,
      "epoch": 3.8167438608756443,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00036028838147928854,
      "loss": 5.2218,
      "mean_token_accuracy": 0.21915984004735947,
      "num_tokens": 88500487.0,
      "step": 43675
    },
    {
      "entropy": 5.991075086593628,
      "epoch": 3.8171808092283492,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036025954504414223,
      "loss": 5.1681,
      "mean_token_accuracy": 0.21759558618068695,
      "num_tokens": 88510857.0,
      "step": 43680
    },
    {
      "entropy": 6.045414829254151,
      "epoch": 3.817617757581054,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036023070697367314,
      "loss": 5.2314,
      "mean_token_accuracy": 0.21114376187324524,
      "num_tokens": 88522551.0,
      "step": 43685
    },
    {
      "entropy": 6.080643081665039,
      "epoch": 3.8180547059337586,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036020186726843447,
      "loss": 5.3117,
      "mean_token_accuracy": 0.20884378552436828,
      "num_tokens": 88534069.0,
      "step": 43690
    },
    {
      "entropy": 6.0078388214111325,
      "epoch": 3.818491654286463,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003601730259289793,
      "loss": 5.1344,
      "mean_token_accuracy": 0.21417221277952195,
      "num_tokens": 88544912.0,
      "step": 43695
    },
    {
      "entropy": 5.936833047866822,
      "epoch": 3.818928602639168,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003601441829558609,
      "loss": 5.1147,
      "mean_token_accuracy": 0.23480434715747833,
      "num_tokens": 88554445.0,
      "step": 43700
    },
    {
      "entropy": 6.019472408294678,
      "epoch": 3.819365550991873,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003601153383496324,
      "loss": 5.2626,
      "mean_token_accuracy": 0.20648087859153746,
      "num_tokens": 88563589.0,
      "step": 43705
    },
    {
      "entropy": 6.072085762023926,
      "epoch": 3.8198024993445774,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036008649211084716,
      "loss": 5.3237,
      "mean_token_accuracy": 0.2053004339337349,
      "num_tokens": 88574072.0,
      "step": 43710
    },
    {
      "entropy": 6.039950180053711,
      "epoch": 3.8202394476972823,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003600576442400584,
      "loss": 5.2445,
      "mean_token_accuracy": 0.20736179649829864,
      "num_tokens": 88584225.0,
      "step": 43715
    },
    {
      "entropy": 6.030276203155518,
      "epoch": 3.8206763960499868,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003600287947378195,
      "loss": 5.1441,
      "mean_token_accuracy": 0.21681873947381974,
      "num_tokens": 88594364.0,
      "step": 43720
    },
    {
      "entropy": 6.050450372695923,
      "epoch": 3.8211133444026917,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003599999436046837,
      "loss": 5.2855,
      "mean_token_accuracy": 0.20865631550550462,
      "num_tokens": 88605678.0,
      "step": 43725
    },
    {
      "entropy": 6.060613012313842,
      "epoch": 3.821550292755396,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035997109084120446,
      "loss": 5.2102,
      "mean_token_accuracy": 0.2115587741136551,
      "num_tokens": 88615824.0,
      "step": 43730
    },
    {
      "entropy": 6.089890003204346,
      "epoch": 3.821987241108101,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035994223644793516,
      "loss": 5.2482,
      "mean_token_accuracy": 0.20776202231645585,
      "num_tokens": 88625939.0,
      "step": 43735
    },
    {
      "entropy": 6.0666444301605225,
      "epoch": 3.822424189460806,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003599133804254293,
      "loss": 5.2007,
      "mean_token_accuracy": 0.21236979365348815,
      "num_tokens": 88635453.0,
      "step": 43740
    },
    {
      "entropy": 6.0426982879638675,
      "epoch": 3.8228611378135104,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003598845227742402,
      "loss": 5.1672,
      "mean_token_accuracy": 0.2145812213420868,
      "num_tokens": 88646190.0,
      "step": 43745
    },
    {
      "entropy": 6.0230042934417725,
      "epoch": 3.823298086166215,
      "grad_norm": 1.359375,
      "learning_rate": 0.00035985566349492156,
      "loss": 5.2013,
      "mean_token_accuracy": 0.21575212180614473,
      "num_tokens": 88655957.0,
      "step": 43750
    },
    {
      "entropy": 6.054671812057495,
      "epoch": 3.82373503451892,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035982680258802685,
      "loss": 5.2194,
      "mean_token_accuracy": 0.21178064942359925,
      "num_tokens": 88665884.0,
      "step": 43755
    },
    {
      "entropy": 6.032932710647583,
      "epoch": 3.8241719828716247,
      "grad_norm": 1.25,
      "learning_rate": 0.0003597979400541096,
      "loss": 5.2017,
      "mean_token_accuracy": 0.21252846121788024,
      "num_tokens": 88677226.0,
      "step": 43760
    },
    {
      "entropy": 6.02071099281311,
      "epoch": 3.824608931224329,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003597690758937234,
      "loss": 5.1888,
      "mean_token_accuracy": 0.21199328303337098,
      "num_tokens": 88687447.0,
      "step": 43765
    },
    {
      "entropy": 6.003671026229858,
      "epoch": 3.825045879577034,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035974021010742196,
      "loss": 5.1868,
      "mean_token_accuracy": 0.212473826110363,
      "num_tokens": 88696540.0,
      "step": 43770
    },
    {
      "entropy": 6.0469841957092285,
      "epoch": 3.8254828279297386,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035971134269575884,
      "loss": 5.2081,
      "mean_token_accuracy": 0.21038636267185212,
      "num_tokens": 88707945.0,
      "step": 43775
    },
    {
      "entropy": 6.096694660186768,
      "epoch": 3.8259197762824435,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00035968247365928774,
      "loss": 5.3166,
      "mean_token_accuracy": 0.20498718917369843,
      "num_tokens": 88717398.0,
      "step": 43780
    },
    {
      "entropy": 6.100072145462036,
      "epoch": 3.826356724635148,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00035965360299856247,
      "loss": 5.1815,
      "mean_token_accuracy": 0.20416750609874726,
      "num_tokens": 88727030.0,
      "step": 43785
    },
    {
      "entropy": 6.003475046157837,
      "epoch": 3.826793672987853,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003596247307141368,
      "loss": 5.1261,
      "mean_token_accuracy": 0.22135416716337203,
      "num_tokens": 88737413.0,
      "step": 43790
    },
    {
      "entropy": 6.034205770492553,
      "epoch": 3.827230621340558,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003595958568065644,
      "loss": 5.2265,
      "mean_token_accuracy": 0.21882035732269287,
      "num_tokens": 88747577.0,
      "step": 43795
    },
    {
      "entropy": 6.065388536453247,
      "epoch": 3.8276675696932623,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003595669812763991,
      "loss": 5.224,
      "mean_token_accuracy": 0.2096058890223503,
      "num_tokens": 88757023.0,
      "step": 43800
    },
    {
      "entropy": 5.996976852416992,
      "epoch": 3.8281045180459667,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003595381041241948,
      "loss": 5.1422,
      "mean_token_accuracy": 0.22642866522073746,
      "num_tokens": 88767200.0,
      "step": 43805
    },
    {
      "entropy": 5.986136960983276,
      "epoch": 3.8285414663986717,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00035950922535050534,
      "loss": 5.2304,
      "mean_token_accuracy": 0.21161793619394303,
      "num_tokens": 88777448.0,
      "step": 43810
    },
    {
      "entropy": 5.9765747547149655,
      "epoch": 3.8289784147513766,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035948034495588473,
      "loss": 5.1361,
      "mean_token_accuracy": 0.22110623866319656,
      "num_tokens": 88787477.0,
      "step": 43815
    },
    {
      "entropy": 6.135869073867798,
      "epoch": 3.829415363104081,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003594514629408868,
      "loss": 5.2648,
      "mean_token_accuracy": 0.21288446038961412,
      "num_tokens": 88796971.0,
      "step": 43820
    },
    {
      "entropy": 6.035284852981567,
      "epoch": 3.829852311456786,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003594225793060656,
      "loss": 5.1917,
      "mean_token_accuracy": 0.21492722183465957,
      "num_tokens": 88807960.0,
      "step": 43825
    },
    {
      "entropy": 6.055048990249634,
      "epoch": 3.8302892598094904,
      "grad_norm": 1.265625,
      "learning_rate": 0.000359393694051975,
      "loss": 5.3176,
      "mean_token_accuracy": 0.20049853175878524,
      "num_tokens": 88818874.0,
      "step": 43830
    },
    {
      "entropy": 6.070056581497193,
      "epoch": 3.8307262081621953,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003593648071791692,
      "loss": 5.2252,
      "mean_token_accuracy": 0.20741031169891358,
      "num_tokens": 88829216.0,
      "step": 43835
    },
    {
      "entropy": 6.117917203903199,
      "epoch": 3.8311631565149,
      "grad_norm": 1.34375,
      "learning_rate": 0.00035933591868820213,
      "loss": 5.2831,
      "mean_token_accuracy": 0.20265480279922485,
      "num_tokens": 88839261.0,
      "step": 43840
    },
    {
      "entropy": 6.042427635192871,
      "epoch": 3.8316001048676047,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035930702857962804,
      "loss": 5.1753,
      "mean_token_accuracy": 0.21775288879871368,
      "num_tokens": 88849180.0,
      "step": 43845
    },
    {
      "entropy": 5.945922422409057,
      "epoch": 3.8320370532203096,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003592781368540009,
      "loss": 5.2831,
      "mean_token_accuracy": 0.2087254360318184,
      "num_tokens": 88859696.0,
      "step": 43850
    },
    {
      "entropy": 6.109724473953247,
      "epoch": 3.832474001573014,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035924924351187496,
      "loss": 5.2877,
      "mean_token_accuracy": 0.20798275917768477,
      "num_tokens": 88869932.0,
      "step": 43855
    },
    {
      "entropy": 6.061169576644898,
      "epoch": 3.8329109499257186,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035922034855380436,
      "loss": 5.2265,
      "mean_token_accuracy": 0.21644256711006166,
      "num_tokens": 88880946.0,
      "step": 43860
    },
    {
      "entropy": 6.063197994232178,
      "epoch": 3.8333478982784235,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035919145198034337,
      "loss": 5.2403,
      "mean_token_accuracy": 0.21492306441068648,
      "num_tokens": 88890409.0,
      "step": 43865
    },
    {
      "entropy": 6.009378051757812,
      "epoch": 3.8337848466311284,
      "grad_norm": 1.453125,
      "learning_rate": 0.00035916255379204613,
      "loss": 5.3271,
      "mean_token_accuracy": 0.20302467793226242,
      "num_tokens": 88900872.0,
      "step": 43870
    },
    {
      "entropy": 6.078386020660401,
      "epoch": 3.834221794983833,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003591336539894671,
      "loss": 5.1521,
      "mean_token_accuracy": 0.21866815984249116,
      "num_tokens": 88909999.0,
      "step": 43875
    },
    {
      "entropy": 6.036887359619141,
      "epoch": 3.834658743336538,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003591047525731605,
      "loss": 5.1656,
      "mean_token_accuracy": 0.21918347775936126,
      "num_tokens": 88920219.0,
      "step": 43880
    },
    {
      "entropy": 6.006849527359009,
      "epoch": 3.8350956916892422,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035907584954368067,
      "loss": 5.2958,
      "mean_token_accuracy": 0.20617510080337526,
      "num_tokens": 88930115.0,
      "step": 43885
    },
    {
      "entropy": 6.095245409011841,
      "epoch": 3.835532640041947,
      "grad_norm": 1.390625,
      "learning_rate": 0.00035904694490158196,
      "loss": 5.2537,
      "mean_token_accuracy": 0.2080615609884262,
      "num_tokens": 88940114.0,
      "step": 43890
    },
    {
      "entropy": 6.139874124526978,
      "epoch": 3.8359695883946516,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003590180386474188,
      "loss": 5.3168,
      "mean_token_accuracy": 0.20391667187213897,
      "num_tokens": 88951955.0,
      "step": 43895
    },
    {
      "entropy": 6.055931854248047,
      "epoch": 3.8364065367473565,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003589891307817457,
      "loss": 5.1548,
      "mean_token_accuracy": 0.2169928655028343,
      "num_tokens": 88961677.0,
      "step": 43900
    },
    {
      "entropy": 6.034639024734497,
      "epoch": 3.836843485100061,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000358960221305117,
      "loss": 5.2145,
      "mean_token_accuracy": 0.20946093201637267,
      "num_tokens": 88971888.0,
      "step": 43905
    },
    {
      "entropy": 6.107884550094605,
      "epoch": 3.837280433452766,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003589313102180873,
      "loss": 5.3148,
      "mean_token_accuracy": 0.2088872477412224,
      "num_tokens": 88980681.0,
      "step": 43910
    },
    {
      "entropy": 6.070325469970703,
      "epoch": 3.8377173818054704,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003589023975212111,
      "loss": 5.3024,
      "mean_token_accuracy": 0.20892338454723358,
      "num_tokens": 88991410.0,
      "step": 43915
    },
    {
      "entropy": 6.0024608135223385,
      "epoch": 3.8381543301581753,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035887348321504285,
      "loss": 5.1543,
      "mean_token_accuracy": 0.21913469582796097,
      "num_tokens": 89001743.0,
      "step": 43920
    },
    {
      "entropy": 6.008464193344116,
      "epoch": 3.8385912785108802,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00035884456730013735,
      "loss": 5.15,
      "mean_token_accuracy": 0.21597703844308852,
      "num_tokens": 89011138.0,
      "step": 43925
    },
    {
      "entropy": 6.0530013084411625,
      "epoch": 3.8390282268635847,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035881564977704913,
      "loss": 5.2391,
      "mean_token_accuracy": 0.21619384437799455,
      "num_tokens": 89021282.0,
      "step": 43930
    },
    {
      "entropy": 6.037879085540771,
      "epoch": 3.839465175216289,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035878673064633276,
      "loss": 5.2772,
      "mean_token_accuracy": 0.20740268975496293,
      "num_tokens": 89032656.0,
      "step": 43935
    },
    {
      "entropy": 6.125789546966553,
      "epoch": 3.839902123568994,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035875780990854303,
      "loss": 5.2996,
      "mean_token_accuracy": 0.19957833886146545,
      "num_tokens": 89042727.0,
      "step": 43940
    },
    {
      "entropy": 6.038437557220459,
      "epoch": 3.840339071921699,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035872888756423456,
      "loss": 5.2147,
      "mean_token_accuracy": 0.21365428864955902,
      "num_tokens": 89052111.0,
      "step": 43945
    },
    {
      "entropy": 6.0563445568084715,
      "epoch": 3.8407760202744035,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003586999636139622,
      "loss": 5.2807,
      "mean_token_accuracy": 0.20388780534267426,
      "num_tokens": 89062423.0,
      "step": 43950
    },
    {
      "entropy": 6.038159132003784,
      "epoch": 3.8412129686271084,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035867103805828067,
      "loss": 5.235,
      "mean_token_accuracy": 0.21086632907390596,
      "num_tokens": 89072457.0,
      "step": 43955
    },
    {
      "entropy": 6.052265882492065,
      "epoch": 3.841649916979813,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035864211089774487,
      "loss": 5.254,
      "mean_token_accuracy": 0.20554041713476182,
      "num_tokens": 89081919.0,
      "step": 43960
    },
    {
      "entropy": 6.02734146118164,
      "epoch": 3.8420868653325178,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003586131821329094,
      "loss": 5.2153,
      "mean_token_accuracy": 0.20968266278505326,
      "num_tokens": 89092177.0,
      "step": 43965
    },
    {
      "entropy": 6.049314165115357,
      "epoch": 3.8425238136852222,
      "grad_norm": 1.25,
      "learning_rate": 0.0003585842517643294,
      "loss": 5.2244,
      "mean_token_accuracy": 0.21502961963415146,
      "num_tokens": 89102721.0,
      "step": 43970
    },
    {
      "entropy": 6.067979240417481,
      "epoch": 3.842960762037927,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00035855531979255955,
      "loss": 5.1995,
      "mean_token_accuracy": 0.21820572912693023,
      "num_tokens": 89111841.0,
      "step": 43975
    },
    {
      "entropy": 6.054124450683593,
      "epoch": 3.843397710390632,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035852638621815497,
      "loss": 5.2135,
      "mean_token_accuracy": 0.21315988451242446,
      "num_tokens": 89122551.0,
      "step": 43980
    },
    {
      "entropy": 6.054220819473267,
      "epoch": 3.8438346587433365,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003584974510416706,
      "loss": 5.2913,
      "mean_token_accuracy": 0.20750633031129836,
      "num_tokens": 89133388.0,
      "step": 43985
    },
    {
      "entropy": 6.066481876373291,
      "epoch": 3.844271607096041,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035846851426366125,
      "loss": 5.246,
      "mean_token_accuracy": 0.21094100326299667,
      "num_tokens": 89144435.0,
      "step": 43990
    },
    {
      "entropy": 6.070482349395752,
      "epoch": 3.844708555448746,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035843957588468207,
      "loss": 5.2384,
      "mean_token_accuracy": 0.21885087043046952,
      "num_tokens": 89154509.0,
      "step": 43995
    },
    {
      "entropy": 6.063790369033813,
      "epoch": 3.845145503801451,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003584106359052881,
      "loss": 5.3181,
      "mean_token_accuracy": 0.20533374398946763,
      "num_tokens": 89165352.0,
      "step": 44000
    },
    {
      "entropy": 6.055253982543945,
      "epoch": 3.8455824521541553,
      "grad_norm": 1.375,
      "learning_rate": 0.0003583816943260345,
      "loss": 5.212,
      "mean_token_accuracy": 0.21200033724308015,
      "num_tokens": 89175185.0,
      "step": 44005
    },
    {
      "entropy": 6.058289527893066,
      "epoch": 3.84601940050686,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003583527511474763,
      "loss": 5.1819,
      "mean_token_accuracy": 0.214655801653862,
      "num_tokens": 89183864.0,
      "step": 44010
    },
    {
      "entropy": 6.09380784034729,
      "epoch": 3.8464563488595647,
      "grad_norm": 1.375,
      "learning_rate": 0.00035832380637016846,
      "loss": 5.2679,
      "mean_token_accuracy": 0.207011641561985,
      "num_tokens": 89194011.0,
      "step": 44015
    },
    {
      "entropy": 6.037906885147095,
      "epoch": 3.8468932972122696,
      "grad_norm": 1.390625,
      "learning_rate": 0.00035829485999466657,
      "loss": 5.2152,
      "mean_token_accuracy": 0.21459684073925017,
      "num_tokens": 89204213.0,
      "step": 44020
    },
    {
      "entropy": 6.005055332183838,
      "epoch": 3.847330245564974,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003582659120215256,
      "loss": 5.1672,
      "mean_token_accuracy": 0.21928096413612366,
      "num_tokens": 89214160.0,
      "step": 44025
    },
    {
      "entropy": 6.007287311553955,
      "epoch": 3.847767193917679,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035823696245130073,
      "loss": 5.1818,
      "mean_token_accuracy": 0.21649291962385178,
      "num_tokens": 89223399.0,
      "step": 44030
    },
    {
      "entropy": 6.085443115234375,
      "epoch": 3.848204142270384,
      "grad_norm": 1.421875,
      "learning_rate": 0.00035820801128454735,
      "loss": 5.2039,
      "mean_token_accuracy": 0.21237584799528123,
      "num_tokens": 89232703.0,
      "step": 44035
    },
    {
      "entropy": 6.090215826034546,
      "epoch": 3.8486410906230883,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003581790585218207,
      "loss": 5.2673,
      "mean_token_accuracy": 0.2051667183637619,
      "num_tokens": 89243302.0,
      "step": 44040
    },
    {
      "entropy": 6.07885422706604,
      "epoch": 3.849078038975793,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003581501041636761,
      "loss": 5.2753,
      "mean_token_accuracy": 0.19954513758420944,
      "num_tokens": 89253340.0,
      "step": 44045
    },
    {
      "entropy": 6.045441102981568,
      "epoch": 3.8495149873284977,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035812114821066903,
      "loss": 5.134,
      "mean_token_accuracy": 0.21798515915870667,
      "num_tokens": 89263362.0,
      "step": 44050
    },
    {
      "entropy": 5.928891992568969,
      "epoch": 3.8499519356812026,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0003580921906633547,
      "loss": 5.0816,
      "mean_token_accuracy": 0.22609416097402574,
      "num_tokens": 89275386.0,
      "step": 44055
    },
    {
      "entropy": 6.067906904220581,
      "epoch": 3.850388884033907,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003580632315222886,
      "loss": 5.2377,
      "mean_token_accuracy": 0.20697354674339294,
      "num_tokens": 89285325.0,
      "step": 44060
    },
    {
      "entropy": 6.041475248336792,
      "epoch": 3.850825832386612,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003580342707880262,
      "loss": 5.1314,
      "mean_token_accuracy": 0.22035499215126036,
      "num_tokens": 89295850.0,
      "step": 44065
    },
    {
      "entropy": 6.0327150344848635,
      "epoch": 3.8512627807393165,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000358005308461123,
      "loss": 5.1614,
      "mean_token_accuracy": 0.22211634814739228,
      "num_tokens": 89305403.0,
      "step": 44070
    },
    {
      "entropy": 5.98464994430542,
      "epoch": 3.8516997290920214,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035797634454213454,
      "loss": 5.1111,
      "mean_token_accuracy": 0.22003869861364364,
      "num_tokens": 89314551.0,
      "step": 44075
    },
    {
      "entropy": 6.006225490570069,
      "epoch": 3.852136677444726,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003579473790316162,
      "loss": 5.251,
      "mean_token_accuracy": 0.2102419376373291,
      "num_tokens": 89323872.0,
      "step": 44080
    },
    {
      "entropy": 6.014919853210449,
      "epoch": 3.852573625797431,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035791841193012373,
      "loss": 5.2039,
      "mean_token_accuracy": 0.22059657871723176,
      "num_tokens": 89335114.0,
      "step": 44085
    },
    {
      "entropy": 6.076316022872925,
      "epoch": 3.8530105741501357,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003578894432382127,
      "loss": 5.1774,
      "mean_token_accuracy": 0.21401791125535966,
      "num_tokens": 89344382.0,
      "step": 44090
    },
    {
      "entropy": 6.034866046905518,
      "epoch": 3.85344752250284,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003578604729564387,
      "loss": 5.2315,
      "mean_token_accuracy": 0.2061008483171463,
      "num_tokens": 89355327.0,
      "step": 44095
    },
    {
      "entropy": 6.056199169158935,
      "epoch": 3.8538844708555446,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035783150108535737,
      "loss": 5.2239,
      "mean_token_accuracy": 0.2195705220103264,
      "num_tokens": 89365361.0,
      "step": 44100
    },
    {
      "entropy": 6.071156978607178,
      "epoch": 3.8543214192082496,
      "grad_norm": 1.390625,
      "learning_rate": 0.00035780252762552444,
      "loss": 5.2403,
      "mean_token_accuracy": 0.2107245981693268,
      "num_tokens": 89375097.0,
      "step": 44105
    },
    {
      "entropy": 6.023601055145264,
      "epoch": 3.8547583675609545,
      "grad_norm": 1.421875,
      "learning_rate": 0.00035777355257749573,
      "loss": 5.1874,
      "mean_token_accuracy": 0.21538786739110946,
      "num_tokens": 89384934.0,
      "step": 44110
    },
    {
      "entropy": 5.963386106491089,
      "epoch": 3.855195315913659,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035774457594182685,
      "loss": 5.1101,
      "mean_token_accuracy": 0.22581736594438553,
      "num_tokens": 89395493.0,
      "step": 44115
    },
    {
      "entropy": 5.995259189605713,
      "epoch": 3.855632264266364,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003577155977190736,
      "loss": 5.1652,
      "mean_token_accuracy": 0.2234175756573677,
      "num_tokens": 89405057.0,
      "step": 44120
    },
    {
      "entropy": 5.984979820251465,
      "epoch": 3.8560692126190683,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035768661790979193,
      "loss": 5.1682,
      "mean_token_accuracy": 0.21996648162603377,
      "num_tokens": 89415328.0,
      "step": 44125
    },
    {
      "entropy": 6.004046058654785,
      "epoch": 3.8565061609717732,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003576576365145376,
      "loss": 5.2064,
      "mean_token_accuracy": 0.2165834993124008,
      "num_tokens": 89425138.0,
      "step": 44130
    },
    {
      "entropy": 6.008876037597656,
      "epoch": 3.8569431093244777,
      "grad_norm": 1.25,
      "learning_rate": 0.00035762865353386637,
      "loss": 5.2215,
      "mean_token_accuracy": 0.2159249648451805,
      "num_tokens": 89434972.0,
      "step": 44135
    },
    {
      "entropy": 6.050739002227783,
      "epoch": 3.8573800576771826,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003575996689683344,
      "loss": 5.2206,
      "mean_token_accuracy": 0.21539411097764968,
      "num_tokens": 89445661.0,
      "step": 44140
    },
    {
      "entropy": 6.03762412071228,
      "epoch": 3.8578170060298875,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035757068281849747,
      "loss": 5.2828,
      "mean_token_accuracy": 0.2042894557118416,
      "num_tokens": 89455970.0,
      "step": 44145
    },
    {
      "entropy": 6.043173742294312,
      "epoch": 3.858253954382592,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003575416950849115,
      "loss": 5.2575,
      "mean_token_accuracy": 0.21004310995340347,
      "num_tokens": 89466076.0,
      "step": 44150
    },
    {
      "entropy": 6.089432907104492,
      "epoch": 3.8586909027352965,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003575127057681326,
      "loss": 5.2367,
      "mean_token_accuracy": 0.2066671296954155,
      "num_tokens": 89475525.0,
      "step": 44155
    },
    {
      "entropy": 6.1064183712005615,
      "epoch": 3.8591278510880014,
      "grad_norm": 1.375,
      "learning_rate": 0.0003574837148687168,
      "loss": 5.3067,
      "mean_token_accuracy": 0.2020388960838318,
      "num_tokens": 89485913.0,
      "step": 44160
    },
    {
      "entropy": 6.066322231292725,
      "epoch": 3.8595647994407063,
      "grad_norm": 1.390625,
      "learning_rate": 0.00035745472238722013,
      "loss": 5.2261,
      "mean_token_accuracy": 0.20527456402778627,
      "num_tokens": 89496321.0,
      "step": 44165
    },
    {
      "entropy": 6.034045934677124,
      "epoch": 3.8600017477934108,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035742572832419867,
      "loss": 5.2176,
      "mean_token_accuracy": 0.2148586243391037,
      "num_tokens": 89506335.0,
      "step": 44170
    },
    {
      "entropy": 6.055061531066895,
      "epoch": 3.8604386961461157,
      "grad_norm": 1.4375,
      "learning_rate": 0.00035739673268020856,
      "loss": 5.2065,
      "mean_token_accuracy": 0.21383664458990098,
      "num_tokens": 89515683.0,
      "step": 44175
    },
    {
      "entropy": 5.967793083190918,
      "epoch": 3.86087564449882,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00035736773545580594,
      "loss": 5.2535,
      "mean_token_accuracy": 0.2136587157845497,
      "num_tokens": 89525797.0,
      "step": 44180
    },
    {
      "entropy": 6.079654264450073,
      "epoch": 3.861312592851525,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035733873665154696,
      "loss": 5.2663,
      "mean_token_accuracy": 0.20581847727298735,
      "num_tokens": 89535312.0,
      "step": 44185
    },
    {
      "entropy": 5.949971055984497,
      "epoch": 3.8617495412042295,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003573097362679879,
      "loss": 5.0904,
      "mean_token_accuracy": 0.23018173575401307,
      "num_tokens": 89545498.0,
      "step": 44190
    },
    {
      "entropy": 6.0138932228088375,
      "epoch": 3.8621864895569344,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003572807343056849,
      "loss": 5.149,
      "mean_token_accuracy": 0.21860153824090958,
      "num_tokens": 89555421.0,
      "step": 44195
    },
    {
      "entropy": 6.049606704711914,
      "epoch": 3.8626234379096394,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003572517307651944,
      "loss": 5.2124,
      "mean_token_accuracy": 0.21642376780509948,
      "num_tokens": 89565549.0,
      "step": 44200
    },
    {
      "entropy": 5.997225618362426,
      "epoch": 3.863060386262344,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003572227256470725,
      "loss": 5.1976,
      "mean_token_accuracy": 0.21526889652013778,
      "num_tokens": 89575106.0,
      "step": 44205
    },
    {
      "entropy": 6.052852725982666,
      "epoch": 3.8634973346150483,
      "grad_norm": 1.390625,
      "learning_rate": 0.00035719371895187574,
      "loss": 5.1786,
      "mean_token_accuracy": 0.21287223994731902,
      "num_tokens": 89584000.0,
      "step": 44210
    },
    {
      "entropy": 5.9626665115356445,
      "epoch": 3.863934282967753,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003571647106801604,
      "loss": 5.1974,
      "mean_token_accuracy": 0.22005974054336547,
      "num_tokens": 89594557.0,
      "step": 44215
    },
    {
      "entropy": 6.006411218643189,
      "epoch": 3.864371231320458,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003571357008324828,
      "loss": 5.1849,
      "mean_token_accuracy": 0.2140863910317421,
      "num_tokens": 89603710.0,
      "step": 44220
    },
    {
      "entropy": 6.033378791809082,
      "epoch": 3.8648081796731626,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035710668940939935,
      "loss": 5.2715,
      "mean_token_accuracy": 0.20123306810855865,
      "num_tokens": 89614880.0,
      "step": 44225
    },
    {
      "entropy": 6.081842136383057,
      "epoch": 3.8652451280258675,
      "grad_norm": 1.359375,
      "learning_rate": 0.00035707767641146663,
      "loss": 5.2801,
      "mean_token_accuracy": 0.21004445105791092,
      "num_tokens": 89624377.0,
      "step": 44230
    },
    {
      "entropy": 6.014404916763306,
      "epoch": 3.865682076378572,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003570486618392411,
      "loss": 5.2388,
      "mean_token_accuracy": 0.20376088470220566,
      "num_tokens": 89634260.0,
      "step": 44235
    },
    {
      "entropy": 6.029397869110108,
      "epoch": 3.866119024731277,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035701964569327926,
      "loss": 5.2,
      "mean_token_accuracy": 0.20545893013477326,
      "num_tokens": 89643642.0,
      "step": 44240
    },
    {
      "entropy": 6.078573513031006,
      "epoch": 3.8665559730839814,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035699062797413754,
      "loss": 5.2736,
      "mean_token_accuracy": 0.20511059165000917,
      "num_tokens": 89653461.0,
      "step": 44245
    },
    {
      "entropy": 6.064158630371094,
      "epoch": 3.8669929214366863,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0003569616086823727,
      "loss": 5.2715,
      "mean_token_accuracy": 0.20584552586078644,
      "num_tokens": 89663108.0,
      "step": 44250
    },
    {
      "entropy": 6.109140300750733,
      "epoch": 3.8674298697893907,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003569325878185412,
      "loss": 5.2719,
      "mean_token_accuracy": 0.19839015305042268,
      "num_tokens": 89673682.0,
      "step": 44255
    },
    {
      "entropy": 6.09443941116333,
      "epoch": 3.8678668181420957,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003569035653831998,
      "loss": 5.2867,
      "mean_token_accuracy": 0.2102901041507721,
      "num_tokens": 89684795.0,
      "step": 44260
    },
    {
      "entropy": 6.081226634979248,
      "epoch": 3.8683037664948,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00035687454137690514,
      "loss": 5.2435,
      "mean_token_accuracy": 0.212512968480587,
      "num_tokens": 89695134.0,
      "step": 44265
    },
    {
      "entropy": 6.017131757736206,
      "epoch": 3.868740714847505,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035684551580021374,
      "loss": 5.2235,
      "mean_token_accuracy": 0.21740752607584,
      "num_tokens": 89705087.0,
      "step": 44270
    },
    {
      "entropy": 6.05048975944519,
      "epoch": 3.86917766320021,
      "grad_norm": 1.34375,
      "learning_rate": 0.00035681648865368255,
      "loss": 5.1805,
      "mean_token_accuracy": 0.20973471850156783,
      "num_tokens": 89714920.0,
      "step": 44275
    },
    {
      "entropy": 6.062396049499512,
      "epoch": 3.8696146115529144,
      "grad_norm": 1.390625,
      "learning_rate": 0.00035678745993786813,
      "loss": 5.1639,
      "mean_token_accuracy": 0.21441112011671065,
      "num_tokens": 89724022.0,
      "step": 44280
    },
    {
      "entropy": 6.0294890880584715,
      "epoch": 3.870051559905619,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003567584296533275,
      "loss": 5.1879,
      "mean_token_accuracy": 0.21483559161424637,
      "num_tokens": 89734179.0,
      "step": 44285
    },
    {
      "entropy": 6.004508686065674,
      "epoch": 3.870488508258324,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003567293978006173,
      "loss": 5.2179,
      "mean_token_accuracy": 0.21364942491054534,
      "num_tokens": 89743810.0,
      "step": 44290
    },
    {
      "entropy": 6.055318164825439,
      "epoch": 3.8709254566110287,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003567003643802944,
      "loss": 5.2132,
      "mean_token_accuracy": 0.2183991327881813,
      "num_tokens": 89753578.0,
      "step": 44295
    },
    {
      "entropy": 5.997836446762085,
      "epoch": 3.871362404963733,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035667132939291575,
      "loss": 5.1594,
      "mean_token_accuracy": 0.2110862746834755,
      "num_tokens": 89763569.0,
      "step": 44300
    },
    {
      "entropy": 6.022614049911499,
      "epoch": 3.871799353316438,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0003566422928390381,
      "loss": 5.2669,
      "mean_token_accuracy": 0.20601911395788192,
      "num_tokens": 89772759.0,
      "step": 44305
    },
    {
      "entropy": 5.949989128112793,
      "epoch": 3.8722363016691426,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003566132547192187,
      "loss": 5.1337,
      "mean_token_accuracy": 0.21870688647031783,
      "num_tokens": 89782499.0,
      "step": 44310
    },
    {
      "entropy": 6.001076650619507,
      "epoch": 3.8726732500218475,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035658421503401415,
      "loss": 5.247,
      "mean_token_accuracy": 0.20782091319561005,
      "num_tokens": 89792831.0,
      "step": 44315
    },
    {
      "entropy": 6.060327816009521,
      "epoch": 3.873110198374552,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003565551737839816,
      "loss": 5.2237,
      "mean_token_accuracy": 0.20990251153707504,
      "num_tokens": 89802411.0,
      "step": 44320
    },
    {
      "entropy": 6.048014879226685,
      "epoch": 3.873547146727257,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003565261309696781,
      "loss": 5.2382,
      "mean_token_accuracy": 0.21190426051616668,
      "num_tokens": 89812157.0,
      "step": 44325
    },
    {
      "entropy": 6.0073894500732425,
      "epoch": 3.8739840950799618,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003564970865916608,
      "loss": 5.1316,
      "mean_token_accuracy": 0.22309516370296478,
      "num_tokens": 89821849.0,
      "step": 44330
    },
    {
      "entropy": 6.047479438781738,
      "epoch": 3.8744210434326662,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00035646804065048653,
      "loss": 5.2038,
      "mean_token_accuracy": 0.20935521125793458,
      "num_tokens": 89830993.0,
      "step": 44335
    },
    {
      "entropy": 5.969098806381226,
      "epoch": 3.8748579917853707,
      "grad_norm": 1.25,
      "learning_rate": 0.00035643899314671265,
      "loss": 5.2164,
      "mean_token_accuracy": 0.21555138528347015,
      "num_tokens": 89841418.0,
      "step": 44340
    },
    {
      "entropy": 6.0016673564910885,
      "epoch": 3.8752949401380756,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00035640994408089616,
      "loss": 5.1908,
      "mean_token_accuracy": 0.22673381119966507,
      "num_tokens": 89851994.0,
      "step": 44345
    },
    {
      "entropy": 6.043539237976074,
      "epoch": 3.8757318884907805,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035638089345359426,
      "loss": 5.2123,
      "mean_token_accuracy": 0.21598996222019196,
      "num_tokens": 89861127.0,
      "step": 44350
    },
    {
      "entropy": 6.082724905014038,
      "epoch": 3.876168836843485,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035635184126536426,
      "loss": 5.2396,
      "mean_token_accuracy": 0.2141295462846756,
      "num_tokens": 89872346.0,
      "step": 44355
    },
    {
      "entropy": 6.077357578277588,
      "epoch": 3.87660578519619,
      "grad_norm": 1.25,
      "learning_rate": 0.0003563227875167633,
      "loss": 5.2678,
      "mean_token_accuracy": 0.20571626722812653,
      "num_tokens": 89882223.0,
      "step": 44360
    },
    {
      "entropy": 6.053806591033935,
      "epoch": 3.8770427335488944,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035629373220834857,
      "loss": 5.2448,
      "mean_token_accuracy": 0.2091764137148857,
      "num_tokens": 89892305.0,
      "step": 44365
    },
    {
      "entropy": 6.000322866439819,
      "epoch": 3.8774796819015993,
      "grad_norm": 1.25,
      "learning_rate": 0.0003562646753406775,
      "loss": 5.1879,
      "mean_token_accuracy": 0.21645335406064986,
      "num_tokens": 89901547.0,
      "step": 44370
    },
    {
      "entropy": 5.9901426315307615,
      "epoch": 3.8779166302543038,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035623561691430734,
      "loss": 5.2257,
      "mean_token_accuracy": 0.21219248175621033,
      "num_tokens": 89912686.0,
      "step": 44375
    },
    {
      "entropy": 6.081467723846435,
      "epoch": 3.8783535786070087,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003562065569297956,
      "loss": 5.2378,
      "mean_token_accuracy": 0.2136268511414528,
      "num_tokens": 89921728.0,
      "step": 44380
    },
    {
      "entropy": 6.039179039001465,
      "epoch": 3.8787905269597136,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003561774953876995,
      "loss": 5.1884,
      "mean_token_accuracy": 0.21490003168582916,
      "num_tokens": 89931741.0,
      "step": 44385
    },
    {
      "entropy": 6.0189454555511475,
      "epoch": 3.879227475312418,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035614843228857644,
      "loss": 5.2129,
      "mean_token_accuracy": 0.21415937393903733,
      "num_tokens": 89941323.0,
      "step": 44390
    },
    {
      "entropy": 6.0085149765014645,
      "epoch": 3.8796644236651225,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000356119367632984,
      "loss": 5.2725,
      "mean_token_accuracy": 0.20298081636428833,
      "num_tokens": 89950763.0,
      "step": 44395
    },
    {
      "entropy": 6.025460481643677,
      "epoch": 3.8801013720178275,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003560903014214795,
      "loss": 5.2899,
      "mean_token_accuracy": 0.21051902323961258,
      "num_tokens": 89961371.0,
      "step": 44400
    },
    {
      "entropy": 6.089932823181153,
      "epoch": 3.8805383203705324,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003560612336546206,
      "loss": 5.2964,
      "mean_token_accuracy": 0.2084074452519417,
      "num_tokens": 89971880.0,
      "step": 44405
    },
    {
      "entropy": 6.042935180664062,
      "epoch": 3.880975268723237,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003560321643329648,
      "loss": 5.1604,
      "mean_token_accuracy": 0.21432658284902573,
      "num_tokens": 89982040.0,
      "step": 44410
    },
    {
      "entropy": 6.066294717788696,
      "epoch": 3.8814122170759418,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003560030934570696,
      "loss": 5.2812,
      "mean_token_accuracy": 0.2059522420167923,
      "num_tokens": 89992517.0,
      "step": 44415
    },
    {
      "entropy": 6.019790458679199,
      "epoch": 3.881849165428646,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003559740210274925,
      "loss": 5.1957,
      "mean_token_accuracy": 0.21489658057689667,
      "num_tokens": 90002432.0,
      "step": 44420
    },
    {
      "entropy": 5.99413743019104,
      "epoch": 3.882286113781351,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003559449470447914,
      "loss": 5.1253,
      "mean_token_accuracy": 0.22342512160539627,
      "num_tokens": 90012556.0,
      "step": 44425
    },
    {
      "entropy": 6.119521522521973,
      "epoch": 3.8827230621340556,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003559158715095238,
      "loss": 5.2792,
      "mean_token_accuracy": 0.20686281472444534,
      "num_tokens": 90022513.0,
      "step": 44430
    },
    {
      "entropy": 6.032407283782959,
      "epoch": 3.8831600104867605,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003558867944222473,
      "loss": 5.226,
      "mean_token_accuracy": 0.21169938296079635,
      "num_tokens": 90032822.0,
      "step": 44435
    },
    {
      "entropy": 6.049501132965088,
      "epoch": 3.8835969588394654,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003558577157835197,
      "loss": 5.1793,
      "mean_token_accuracy": 0.21018293350934983,
      "num_tokens": 90042685.0,
      "step": 44440
    },
    {
      "entropy": 6.000045919418335,
      "epoch": 3.88403390719217,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003558286355938988,
      "loss": 5.2416,
      "mean_token_accuracy": 0.21361093968153,
      "num_tokens": 90053118.0,
      "step": 44445
    },
    {
      "entropy": 5.948855543136597,
      "epoch": 3.8844708555448744,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003557995538539423,
      "loss": 5.1959,
      "mean_token_accuracy": 0.2108166843652725,
      "num_tokens": 90063041.0,
      "step": 44450
    },
    {
      "entropy": 6.0600261211395265,
      "epoch": 3.8849078038975793,
      "grad_norm": 1.359375,
      "learning_rate": 0.000355770470564208,
      "loss": 5.2296,
      "mean_token_accuracy": 0.21254052519798278,
      "num_tokens": 90072995.0,
      "step": 44455
    },
    {
      "entropy": 6.069825458526611,
      "epoch": 3.885344752250284,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003557413857252538,
      "loss": 5.1999,
      "mean_token_accuracy": 0.21559153944253923,
      "num_tokens": 90084369.0,
      "step": 44460
    },
    {
      "entropy": 6.07654218673706,
      "epoch": 3.8857817006029887,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003557122993376374,
      "loss": 5.2638,
      "mean_token_accuracy": 0.20754247158765793,
      "num_tokens": 90093207.0,
      "step": 44465
    },
    {
      "entropy": 5.98681321144104,
      "epoch": 3.8862186489556936,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003556832114019169,
      "loss": 5.2108,
      "mean_token_accuracy": 0.21803067177534102,
      "num_tokens": 90102452.0,
      "step": 44470
    },
    {
      "entropy": 6.039475917816162,
      "epoch": 3.886655597308398,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035565412191865015,
      "loss": 5.1875,
      "mean_token_accuracy": 0.21135098934173585,
      "num_tokens": 90112200.0,
      "step": 44475
    },
    {
      "entropy": 6.027546024322509,
      "epoch": 3.887092545661103,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035562503088839496,
      "loss": 5.2572,
      "mean_token_accuracy": 0.20680559128522874,
      "num_tokens": 90122627.0,
      "step": 44480
    },
    {
      "entropy": 6.068408823013305,
      "epoch": 3.8875294940138074,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003555959383117096,
      "loss": 5.2189,
      "mean_token_accuracy": 0.20743303298950194,
      "num_tokens": 90131303.0,
      "step": 44485
    },
    {
      "entropy": 6.018639755249024,
      "epoch": 3.8879664423665123,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003555668441891518,
      "loss": 5.1815,
      "mean_token_accuracy": 0.21237660199403763,
      "num_tokens": 90140904.0,
      "step": 44490
    },
    {
      "entropy": 6.070509147644043,
      "epoch": 3.8884033907192173,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003555377485212797,
      "loss": 5.1898,
      "mean_token_accuracy": 0.20965614318847656,
      "num_tokens": 90150729.0,
      "step": 44495
    },
    {
      "entropy": 6.112086534500122,
      "epoch": 3.8888403390719217,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003555086513086514,
      "loss": 5.2742,
      "mean_token_accuracy": 0.1998960942029953,
      "num_tokens": 90160832.0,
      "step": 44500
    },
    {
      "entropy": 6.059109354019165,
      "epoch": 3.889277287424626,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000355479552551825,
      "loss": 5.2118,
      "mean_token_accuracy": 0.21826253682374955,
      "num_tokens": 90171606.0,
      "step": 44505
    },
    {
      "entropy": 5.999364233016967,
      "epoch": 3.889714235777331,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035545045225135864,
      "loss": 5.1766,
      "mean_token_accuracy": 0.2202348992228508,
      "num_tokens": 90181902.0,
      "step": 44510
    },
    {
      "entropy": 6.01136999130249,
      "epoch": 3.890151184130036,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003554213504078104,
      "loss": 5.2686,
      "mean_token_accuracy": 0.20788617581129074,
      "num_tokens": 90192380.0,
      "step": 44515
    },
    {
      "entropy": 5.993135118484497,
      "epoch": 3.8905881324827405,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003553922470217385,
      "loss": 5.1638,
      "mean_token_accuracy": 0.21841271072626114,
      "num_tokens": 90202469.0,
      "step": 44520
    },
    {
      "entropy": 6.06570873260498,
      "epoch": 3.8910250808354454,
      "grad_norm": 1.359375,
      "learning_rate": 0.00035536314209370126,
      "loss": 5.2103,
      "mean_token_accuracy": 0.2154616966843605,
      "num_tokens": 90212706.0,
      "step": 44525
    },
    {
      "entropy": 6.070257997512817,
      "epoch": 3.89146202918815,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00035533403562425676,
      "loss": 5.2195,
      "mean_token_accuracy": 0.21040049195289612,
      "num_tokens": 90222456.0,
      "step": 44530
    },
    {
      "entropy": 6.010206317901611,
      "epoch": 3.891898977540855,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035530492761396337,
      "loss": 5.1726,
      "mean_token_accuracy": 0.21556871086359025,
      "num_tokens": 90232960.0,
      "step": 44535
    },
    {
      "entropy": 6.022184276580811,
      "epoch": 3.8923359258935593,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035527581806337946,
      "loss": 5.175,
      "mean_token_accuracy": 0.2123361974954605,
      "num_tokens": 90242407.0,
      "step": 44540
    },
    {
      "entropy": 6.040283489227295,
      "epoch": 3.892772874246264,
      "grad_norm": 1.53125,
      "learning_rate": 0.00035524670697306324,
      "loss": 5.2058,
      "mean_token_accuracy": 0.21237368136644363,
      "num_tokens": 90252211.0,
      "step": 44545
    },
    {
      "entropy": 6.019613218307495,
      "epoch": 3.8932098225989686,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003552175943435731,
      "loss": 5.2357,
      "mean_token_accuracy": 0.21376916021108627,
      "num_tokens": 90262047.0,
      "step": 44550
    },
    {
      "entropy": 5.991096830368042,
      "epoch": 3.8936467709516736,
      "grad_norm": 1.390625,
      "learning_rate": 0.00035518848017546757,
      "loss": 5.2206,
      "mean_token_accuracy": 0.20115814507007598,
      "num_tokens": 90272870.0,
      "step": 44555
    },
    {
      "entropy": 6.036867094039917,
      "epoch": 3.894083719304378,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003551593644693049,
      "loss": 5.3067,
      "mean_token_accuracy": 0.20120874494314195,
      "num_tokens": 90282581.0,
      "step": 44560
    },
    {
      "entropy": 6.027698230743408,
      "epoch": 3.894520667657083,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035513024722564365,
      "loss": 5.2253,
      "mean_token_accuracy": 0.22112547904253005,
      "num_tokens": 90294100.0,
      "step": 44565
    },
    {
      "entropy": 6.0007726669311525,
      "epoch": 3.894957616009788,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035510112844504223,
      "loss": 5.2018,
      "mean_token_accuracy": 0.21727206856012343,
      "num_tokens": 90305106.0,
      "step": 44570
    },
    {
      "entropy": 6.09762225151062,
      "epoch": 3.8953945643624923,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003550720081280592,
      "loss": 5.2323,
      "mean_token_accuracy": 0.20926695168018342,
      "num_tokens": 90315246.0,
      "step": 44575
    },
    {
      "entropy": 6.0569343090057375,
      "epoch": 3.895831512715197,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00035504288627525313,
      "loss": 5.2378,
      "mean_token_accuracy": 0.21331028044223785,
      "num_tokens": 90324312.0,
      "step": 44580
    },
    {
      "entropy": 6.016189622879028,
      "epoch": 3.8962684610679017,
      "grad_norm": 1.25,
      "learning_rate": 0.0003550137628871825,
      "loss": 5.1607,
      "mean_token_accuracy": 0.22070192098617553,
      "num_tokens": 90333639.0,
      "step": 44585
    },
    {
      "entropy": 6.000194406509399,
      "epoch": 3.8967054094206066,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035498463796440605,
      "loss": 5.1551,
      "mean_token_accuracy": 0.21937346905469896,
      "num_tokens": 90343332.0,
      "step": 44590
    },
    {
      "entropy": 5.995822286605835,
      "epoch": 3.897142357773311,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035495551150748226,
      "loss": 5.177,
      "mean_token_accuracy": 0.21693759113550187,
      "num_tokens": 90355337.0,
      "step": 44595
    },
    {
      "entropy": 6.069763040542602,
      "epoch": 3.897579306126016,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035492638351697,
      "loss": 5.3012,
      "mean_token_accuracy": 0.20470404028892517,
      "num_tokens": 90367126.0,
      "step": 44600
    },
    {
      "entropy": 6.078163051605225,
      "epoch": 3.8980162544787205,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003548972539934277,
      "loss": 5.2257,
      "mean_token_accuracy": 0.2106734946370125,
      "num_tokens": 90377408.0,
      "step": 44605
    },
    {
      "entropy": 6.065605783462525,
      "epoch": 3.8984532028314254,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003548681229374142,
      "loss": 5.1953,
      "mean_token_accuracy": 0.21542245149612427,
      "num_tokens": 90388451.0,
      "step": 44610
    },
    {
      "entropy": 6.02002444267273,
      "epoch": 3.89889015118413,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003548389903494883,
      "loss": 5.2267,
      "mean_token_accuracy": 0.2136280745267868,
      "num_tokens": 90399673.0,
      "step": 44615
    },
    {
      "entropy": 6.034029960632324,
      "epoch": 3.8993270995368348,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035480985623020873,
      "loss": 5.1738,
      "mean_token_accuracy": 0.20918907821178437,
      "num_tokens": 90409567.0,
      "step": 44620
    },
    {
      "entropy": 6.06057071685791,
      "epoch": 3.8997640478895397,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003547807205801342,
      "loss": 5.2903,
      "mean_token_accuracy": 0.20792569667100907,
      "num_tokens": 90419904.0,
      "step": 44625
    },
    {
      "entropy": 6.026166391372681,
      "epoch": 3.900200996242244,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003547515833998237,
      "loss": 5.1844,
      "mean_token_accuracy": 0.20957070589065552,
      "num_tokens": 90431445.0,
      "step": 44630
    },
    {
      "entropy": 6.075293445587159,
      "epoch": 3.9006379445949486,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003547224446898361,
      "loss": 5.2004,
      "mean_token_accuracy": 0.21105246841907502,
      "num_tokens": 90440933.0,
      "step": 44635
    },
    {
      "entropy": 6.004954719543457,
      "epoch": 3.9010748929476535,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035469330445073025,
      "loss": 5.2176,
      "mean_token_accuracy": 0.21722626686096191,
      "num_tokens": 90451145.0,
      "step": 44640
    },
    {
      "entropy": 6.065017127990723,
      "epoch": 3.9015118413003584,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000354664162683065,
      "loss": 5.2752,
      "mean_token_accuracy": 0.20330361872911454,
      "num_tokens": 90460355.0,
      "step": 44645
    },
    {
      "entropy": 6.095972776412964,
      "epoch": 3.901948789653063,
      "grad_norm": 1.25,
      "learning_rate": 0.00035463501938739936,
      "loss": 5.2139,
      "mean_token_accuracy": 0.2141601711511612,
      "num_tokens": 90471162.0,
      "step": 44650
    },
    {
      "entropy": 6.078706407546997,
      "epoch": 3.902385738005768,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003546058745642923,
      "loss": 5.2697,
      "mean_token_accuracy": 0.21231320947408677,
      "num_tokens": 90481807.0,
      "step": 44655
    },
    {
      "entropy": 6.033277225494385,
      "epoch": 3.9028226863584723,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00035457672821430293,
      "loss": 5.1763,
      "mean_token_accuracy": 0.2135167047381401,
      "num_tokens": 90491478.0,
      "step": 44660
    },
    {
      "entropy": 5.949684143066406,
      "epoch": 3.903259634711177,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003545475803379902,
      "loss": 5.1102,
      "mean_token_accuracy": 0.22661496698856354,
      "num_tokens": 90501231.0,
      "step": 44665
    },
    {
      "entropy": 5.999005365371704,
      "epoch": 3.9036965830638817,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035451843093591316,
      "loss": 5.2746,
      "mean_token_accuracy": 0.20816677063703537,
      "num_tokens": 90512021.0,
      "step": 44670
    },
    {
      "entropy": 6.093374824523925,
      "epoch": 3.9041335314165866,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035448928000863096,
      "loss": 5.2856,
      "mean_token_accuracy": 0.2160414695739746,
      "num_tokens": 90522210.0,
      "step": 44675
    },
    {
      "entropy": 6.098274898529053,
      "epoch": 3.9045704797692915,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003544601275567027,
      "loss": 5.2711,
      "mean_token_accuracy": 0.213034588098526,
      "num_tokens": 90533024.0,
      "step": 44680
    },
    {
      "entropy": 5.958355951309204,
      "epoch": 3.905007428121996,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003544309735806876,
      "loss": 5.2052,
      "mean_token_accuracy": 0.2182149976491928,
      "num_tokens": 90541965.0,
      "step": 44685
    },
    {
      "entropy": 6.024481773376465,
      "epoch": 3.9054443764747004,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003544018180811448,
      "loss": 5.1767,
      "mean_token_accuracy": 0.21508955210447311,
      "num_tokens": 90552167.0,
      "step": 44690
    },
    {
      "entropy": 6.062833309173584,
      "epoch": 3.9058813248274054,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003543726610586335,
      "loss": 5.1516,
      "mean_token_accuracy": 0.2194569528102875,
      "num_tokens": 90562383.0,
      "step": 44695
    },
    {
      "entropy": 6.048684024810791,
      "epoch": 3.9063182731801103,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035434350251371295,
      "loss": 5.2192,
      "mean_token_accuracy": 0.2150595173239708,
      "num_tokens": 90572609.0,
      "step": 44700
    },
    {
      "entropy": 5.992998123168945,
      "epoch": 3.9067552215328147,
      "grad_norm": 1.25,
      "learning_rate": 0.0003543143424469425,
      "loss": 5.2262,
      "mean_token_accuracy": 0.21848957240581512,
      "num_tokens": 90582319.0,
      "step": 44705
    },
    {
      "entropy": 6.079633140563965,
      "epoch": 3.9071921698855197,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035428518085888134,
      "loss": 5.2497,
      "mean_token_accuracy": 0.21103215962648392,
      "num_tokens": 90592493.0,
      "step": 44710
    },
    {
      "entropy": 6.1451739311218265,
      "epoch": 3.907629118238224,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035425601775008895,
      "loss": 5.3177,
      "mean_token_accuracy": 0.20671460330486296,
      "num_tokens": 90603504.0,
      "step": 44715
    },
    {
      "entropy": 6.105078315734863,
      "epoch": 3.908066066590929,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003542268531211245,
      "loss": 5.2921,
      "mean_token_accuracy": 0.20713300555944442,
      "num_tokens": 90613957.0,
      "step": 44720
    },
    {
      "entropy": 5.990778827667237,
      "epoch": 3.9085030149436335,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003541976869725476,
      "loss": 5.216,
      "mean_token_accuracy": 0.20854901075363158,
      "num_tokens": 90623986.0,
      "step": 44725
    },
    {
      "entropy": 6.027424001693726,
      "epoch": 3.9089399632963384,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003541685193049175,
      "loss": 5.2248,
      "mean_token_accuracy": 0.21089646369218826,
      "num_tokens": 90633879.0,
      "step": 44730
    },
    {
      "entropy": 6.075221109390259,
      "epoch": 3.9093769116490433,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003541393501187937,
      "loss": 5.2404,
      "mean_token_accuracy": 0.20701107233762742,
      "num_tokens": 90644031.0,
      "step": 44735
    },
    {
      "entropy": 6.077065658569336,
      "epoch": 3.909813860001748,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035411017941473576,
      "loss": 5.2466,
      "mean_token_accuracy": 0.21291156262159347,
      "num_tokens": 90654550.0,
      "step": 44740
    },
    {
      "entropy": 6.075545787811279,
      "epoch": 3.9102508083544523,
      "grad_norm": 1.375,
      "learning_rate": 0.00035408100719330304,
      "loss": 5.2566,
      "mean_token_accuracy": 0.20610218048095702,
      "num_tokens": 90664117.0,
      "step": 44745
    },
    {
      "entropy": 6.059450387954712,
      "epoch": 3.910687756707157,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003540518334550552,
      "loss": 5.2377,
      "mean_token_accuracy": 0.2083273336291313,
      "num_tokens": 90674400.0,
      "step": 44750
    },
    {
      "entropy": 6.090710115432739,
      "epoch": 3.911124705059862,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035402265820055185,
      "loss": 5.2734,
      "mean_token_accuracy": 0.20817346274852752,
      "num_tokens": 90684759.0,
      "step": 44755
    },
    {
      "entropy": 6.139222526550293,
      "epoch": 3.9115616534125666,
      "grad_norm": 1.34375,
      "learning_rate": 0.00035399348143035244,
      "loss": 5.2857,
      "mean_token_accuracy": 0.20623773634433745,
      "num_tokens": 90694346.0,
      "step": 44760
    },
    {
      "entropy": 6.067502737045288,
      "epoch": 3.9119986017652715,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035396430314501665,
      "loss": 5.2515,
      "mean_token_accuracy": 0.2139562949538231,
      "num_tokens": 90704239.0,
      "step": 44765
    },
    {
      "entropy": 6.026588821411133,
      "epoch": 3.912435550117976,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003539351233451042,
      "loss": 5.2163,
      "mean_token_accuracy": 0.209856379032135,
      "num_tokens": 90714240.0,
      "step": 44770
    },
    {
      "entropy": 6.077186155319214,
      "epoch": 3.912872498470681,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003539059420311747,
      "loss": 5.2633,
      "mean_token_accuracy": 0.20107121765613556,
      "num_tokens": 90724575.0,
      "step": 44775
    },
    {
      "entropy": 6.118413019180298,
      "epoch": 3.9133094468233853,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003538767592037879,
      "loss": 5.2413,
      "mean_token_accuracy": 0.20907132476568221,
      "num_tokens": 90735217.0,
      "step": 44780
    },
    {
      "entropy": 6.066597986221313,
      "epoch": 3.9137463951760902,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003538475748635035,
      "loss": 5.1965,
      "mean_token_accuracy": 0.22110716104507447,
      "num_tokens": 90745437.0,
      "step": 44785
    },
    {
      "entropy": 6.055582284927368,
      "epoch": 3.914183343528795,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003538183890108813,
      "loss": 5.2393,
      "mean_token_accuracy": 0.20751957595348358,
      "num_tokens": 90755286.0,
      "step": 44790
    },
    {
      "entropy": 6.080459308624268,
      "epoch": 3.9146202918814996,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003537892016464811,
      "loss": 5.2652,
      "mean_token_accuracy": 0.20104107409715652,
      "num_tokens": 90764734.0,
      "step": 44795
    },
    {
      "entropy": 6.076907682418823,
      "epoch": 3.915057240234204,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035376001277086276,
      "loss": 5.2436,
      "mean_token_accuracy": 0.21003336012363433,
      "num_tokens": 90774569.0,
      "step": 44800
    },
    {
      "entropy": 6.129848623275757,
      "epoch": 3.915494188586909,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003537308223845861,
      "loss": 5.3135,
      "mean_token_accuracy": 0.1988552525639534,
      "num_tokens": 90785029.0,
      "step": 44805
    },
    {
      "entropy": 6.133103752136231,
      "epoch": 3.915931136939614,
      "grad_norm": 1.34375,
      "learning_rate": 0.00035370163048821095,
      "loss": 5.3016,
      "mean_token_accuracy": 0.20844779312610626,
      "num_tokens": 90795382.0,
      "step": 44810
    },
    {
      "entropy": 5.985546970367432,
      "epoch": 3.9163680852923184,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035367243708229736,
      "loss": 5.2013,
      "mean_token_accuracy": 0.21770042181015015,
      "num_tokens": 90805354.0,
      "step": 44815
    },
    {
      "entropy": 6.0240589618682865,
      "epoch": 3.9168050336450233,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035364324216740516,
      "loss": 5.2772,
      "mean_token_accuracy": 0.2042780965566635,
      "num_tokens": 90815967.0,
      "step": 44820
    },
    {
      "entropy": 6.131615352630615,
      "epoch": 3.9172419819977278,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035361404574409436,
      "loss": 5.2444,
      "mean_token_accuracy": 0.21279466450214385,
      "num_tokens": 90826717.0,
      "step": 44825
    },
    {
      "entropy": 6.097381496429444,
      "epoch": 3.9176789303504327,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000353584847812925,
      "loss": 5.1723,
      "mean_token_accuracy": 0.21508197039365767,
      "num_tokens": 90836592.0,
      "step": 44830
    },
    {
      "entropy": 6.011948680877685,
      "epoch": 3.918115878703137,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035355564837445707,
      "loss": 5.2731,
      "mean_token_accuracy": 0.20584758818149568,
      "num_tokens": 90847323.0,
      "step": 44835
    },
    {
      "entropy": 6.078777885437011,
      "epoch": 3.918552827055842,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035352644742925066,
      "loss": 5.3481,
      "mean_token_accuracy": 0.19436980038881302,
      "num_tokens": 90858471.0,
      "step": 44840
    },
    {
      "entropy": 6.107532978057861,
      "epoch": 3.918989775408547,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035349724497786577,
      "loss": 5.2726,
      "mean_token_accuracy": 0.20879143923521043,
      "num_tokens": 90868983.0,
      "step": 44845
    },
    {
      "entropy": 6.05928373336792,
      "epoch": 3.9194267237612515,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00035346804102086266,
      "loss": 5.2394,
      "mean_token_accuracy": 0.2109227418899536,
      "num_tokens": 90879904.0,
      "step": 44850
    },
    {
      "entropy": 6.070004463195801,
      "epoch": 3.919863672113956,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00035343883555880143,
      "loss": 5.2683,
      "mean_token_accuracy": 0.20980285555124284,
      "num_tokens": 90889923.0,
      "step": 44855
    },
    {
      "entropy": 6.107934999465942,
      "epoch": 3.920300620466661,
      "grad_norm": 1.375,
      "learning_rate": 0.00035340962859224217,
      "loss": 5.2252,
      "mean_token_accuracy": 0.2090401291847229,
      "num_tokens": 90899991.0,
      "step": 44860
    },
    {
      "entropy": 6.044262790679932,
      "epoch": 3.9207375688193657,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003533804201217452,
      "loss": 5.1851,
      "mean_token_accuracy": 0.2143040731549263,
      "num_tokens": 90910359.0,
      "step": 44865
    },
    {
      "entropy": 6.003077840805053,
      "epoch": 3.92117451717207,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035335121014787063,
      "loss": 5.214,
      "mean_token_accuracy": 0.21096185743808746,
      "num_tokens": 90920033.0,
      "step": 44870
    },
    {
      "entropy": 6.080852270126343,
      "epoch": 3.921611465524775,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003533219986711789,
      "loss": 5.2511,
      "mean_token_accuracy": 0.21282220780849456,
      "num_tokens": 90929301.0,
      "step": 44875
    },
    {
      "entropy": 6.083338689804077,
      "epoch": 3.9220484138774796,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003532927856922301,
      "loss": 5.1779,
      "mean_token_accuracy": 0.21396201252937316,
      "num_tokens": 90939197.0,
      "step": 44880
    },
    {
      "entropy": 6.096610498428345,
      "epoch": 3.9224853622301845,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035326357121158463,
      "loss": 5.2333,
      "mean_token_accuracy": 0.21095086634159088,
      "num_tokens": 90949013.0,
      "step": 44885
    },
    {
      "entropy": 6.0333899974823,
      "epoch": 3.922922310582889,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003532343552298029,
      "loss": 5.2071,
      "mean_token_accuracy": 0.21159491837024688,
      "num_tokens": 90960010.0,
      "step": 44890
    },
    {
      "entropy": 6.009474086761474,
      "epoch": 3.923359258935594,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003532051377474453,
      "loss": 5.2624,
      "mean_token_accuracy": 0.20689420998096467,
      "num_tokens": 90969768.0,
      "step": 44895
    },
    {
      "entropy": 6.025135326385498,
      "epoch": 3.9237962072882984,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003531759187650721,
      "loss": 5.2316,
      "mean_token_accuracy": 0.2169200912117958,
      "num_tokens": 90980051.0,
      "step": 44900
    },
    {
      "entropy": 6.068010425567627,
      "epoch": 3.9242331556410033,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003531466982832438,
      "loss": 5.1989,
      "mean_token_accuracy": 0.21429988741874695,
      "num_tokens": 90990144.0,
      "step": 44905
    },
    {
      "entropy": 6.030220031738281,
      "epoch": 3.9246701039937077,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003531174763025209,
      "loss": 5.1185,
      "mean_token_accuracy": 0.220527546107769,
      "num_tokens": 91000353.0,
      "step": 44910
    },
    {
      "entropy": 6.052541494369507,
      "epoch": 3.9251070523464127,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003530882528234639,
      "loss": 5.2072,
      "mean_token_accuracy": 0.21847995221614838,
      "num_tokens": 91010819.0,
      "step": 44915
    },
    {
      "entropy": 6.074267244338989,
      "epoch": 3.9255440006991176,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035305902784663333,
      "loss": 5.2631,
      "mean_token_accuracy": 0.20386486649513244,
      "num_tokens": 91020454.0,
      "step": 44920
    },
    {
      "entropy": 6.093632507324219,
      "epoch": 3.925980949051822,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035302980137258964,
      "loss": 5.3229,
      "mean_token_accuracy": 0.20737387239933014,
      "num_tokens": 91030710.0,
      "step": 44925
    },
    {
      "entropy": 6.083997583389282,
      "epoch": 3.9264178974045265,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003530005734018935,
      "loss": 5.2271,
      "mean_token_accuracy": 0.21970997005701065,
      "num_tokens": 91039907.0,
      "step": 44930
    },
    {
      "entropy": 6.106987476348877,
      "epoch": 3.9268548457572314,
      "grad_norm": 1.390625,
      "learning_rate": 0.00035297134393510543,
      "loss": 5.2481,
      "mean_token_accuracy": 0.21031178683042526,
      "num_tokens": 91049377.0,
      "step": 44935
    },
    {
      "entropy": 6.051066923141479,
      "epoch": 3.9272917941099363,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003529421129727862,
      "loss": 5.2422,
      "mean_token_accuracy": 0.20467465072870256,
      "num_tokens": 91060341.0,
      "step": 44940
    },
    {
      "entropy": 6.048925065994263,
      "epoch": 3.927728742462641,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035291288051549644,
      "loss": 5.2082,
      "mean_token_accuracy": 0.21474958062171937,
      "num_tokens": 91069737.0,
      "step": 44945
    },
    {
      "entropy": 6.0579366207122805,
      "epoch": 3.9281656908153457,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035288364656379666,
      "loss": 5.2475,
      "mean_token_accuracy": 0.20679220110177993,
      "num_tokens": 91079290.0,
      "step": 44950
    },
    {
      "entropy": 6.040918064117432,
      "epoch": 3.92860263916805,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035285441111824784,
      "loss": 5.2137,
      "mean_token_accuracy": 0.21149878203868866,
      "num_tokens": 91090316.0,
      "step": 44955
    },
    {
      "entropy": 6.066441011428833,
      "epoch": 3.929039587520755,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035282517417941056,
      "loss": 5.2059,
      "mean_token_accuracy": 0.2107372373342514,
      "num_tokens": 91100521.0,
      "step": 44960
    },
    {
      "entropy": 6.012794876098633,
      "epoch": 3.9294765358734596,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003527959357478457,
      "loss": 5.266,
      "mean_token_accuracy": 0.2098572388291359,
      "num_tokens": 91110207.0,
      "step": 44965
    },
    {
      "entropy": 6.051195812225342,
      "epoch": 3.9299134842261645,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000352766695824114,
      "loss": 5.253,
      "mean_token_accuracy": 0.2006568893790245,
      "num_tokens": 91120993.0,
      "step": 44970
    },
    {
      "entropy": 6.014159536361694,
      "epoch": 3.9303504325788694,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003527374544087764,
      "loss": 5.1876,
      "mean_token_accuracy": 0.21584532856941224,
      "num_tokens": 91131769.0,
      "step": 44975
    },
    {
      "entropy": 6.065791034698487,
      "epoch": 3.930787380931574,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003527082115023936,
      "loss": 5.3155,
      "mean_token_accuracy": 0.20369801223278045,
      "num_tokens": 91141958.0,
      "step": 44980
    },
    {
      "entropy": 6.062927198410034,
      "epoch": 3.9312243292842783,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035267896710552667,
      "loss": 5.2084,
      "mean_token_accuracy": 0.21133185923099518,
      "num_tokens": 91151725.0,
      "step": 44985
    },
    {
      "entropy": 6.053237056732177,
      "epoch": 3.9316612776369833,
      "grad_norm": 1.375,
      "learning_rate": 0.00035264972121873633,
      "loss": 5.2197,
      "mean_token_accuracy": 0.21159012913703917,
      "num_tokens": 91161756.0,
      "step": 44990
    },
    {
      "entropy": 6.055783796310425,
      "epoch": 3.932098225989688,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035262047384258376,
      "loss": 5.3226,
      "mean_token_accuracy": 0.20282163172960282,
      "num_tokens": 91172160.0,
      "step": 44995
    },
    {
      "entropy": 6.092729091644287,
      "epoch": 3.9325351743423926,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003525912249776298,
      "loss": 5.2903,
      "mean_token_accuracy": 0.2032739832997322,
      "num_tokens": 91183030.0,
      "step": 45000
    },
    {
      "epoch": 3.9325351743423926,
      "eval_entropy": 5.7840268160010195,
      "eval_loss": 5.26773738861084,
      "eval_mean_token_accuracy": 0.21840926486477885,
      "eval_num_tokens": 91183030.0,
      "eval_runtime": 21.6224,
      "eval_samples_per_second": 1711.832,
      "eval_steps_per_second": 213.991,
      "step": 45000
    }
  ],
  "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": 2.023487019961344e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}