rus-cyrl-10mb-after-ppt-Dp-100mb-ckpt500_seed3407/checkpoint-6000/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5243380232456524,
  "eval_steps": 3000,
  "global_step": 6000,
  "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
    }
  ],
  "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": 2700955567226880.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}